Access to this full-text is provided by Springer Nature.
Content available from Information Systems and e-Business Management
This content is subject to copyright. Terms and conditions apply.
Vol.:(0123456789)
Information Systems and e-Business Management
https://doi.org/10.1007/s10257-024-00688-w
ORIGINAL ARTICLE
Prescriptive analytics systems revised: asystematic
literature review fromaninformation systems perspective
ChristopherWissuchek1 · PatrickZschech2
Received: 15 October 2023 / Revised: 31 May 2024 / Accepted: 9 August 2024
© The Author(s) 2024
Abstract
Prescriptive Analytics Systems (PAS) represent the most mature iteration of busi-
ness analytics, significantly enhancing organizational decision-making. Recently,
research has gained traction, with various technological innovations, including
machine learning and artificial intelligence, significantly influencing the design of
PAS. Although recent studies highlight these developments, the rising trend focuses
on broader implications, such as the synergies and delegation between systems and
users in organizational decision-making environments. Against this backdrop, we
utilized a systematic literature review of 262 articles to build on this evolving per-
spective. Guided by general systems theory and socio-technical thinking, the con-
cept of an information systems artifact directed this review. Our first objective was
to clarify the essential subsystems, identifying 23 constituent components of PAS.
Subsequently, we delved into the meta-level design of PAS, emphasizing the syn-
ergy and delegation between the human decision-maker and prescriptive analytics
in supporting organizational decisions. From this exploration, four distinct system
archetypes emerged: advisory, executive, adaptive, and self-governing PAS. Lastly,
we engaged with affordance theory, illuminating the action potential of PAS. Our
study advances the perspective on PAS, specifically from a broader socio-technical
and information systems viewpoint, highlighting six distinct research directions, act-
ing as a launchpad for future research in the domain.
Keywords Prescriptive analytics· Business analytics· Decision support system·
Systematic literature review· Decision-making· Artificial intelligence
* Christopher Wissuchek
christopher.wissuchek@fau.de
Patrick Zschech
patrick.zschech@uni-leipzig.de
1 FAU Erlangen-Nürnberg, Lange Gasse 20, 90403Nuremberg, Germany
2 Universität Leipzig, Grimmaische Str. 12, 04109Leipzig, Germany
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
1 Introduction
Decision-making is a cognitive process where individuals select from multiple
alternatives. Historically, decisions were primarily based on personal experience,
direct observation, or shared knowledge (Santos and Rosati 2015). However, with
the widespread use of modern information technology, the interconnectedness
of society, and the exponentially growing amount of generated data, decision-
making has become increasingly complex. As a result, humans began employing
mathematical models and algorithms for advanced decision-making to delegate
complex decision tasks to computers.
Against this backdrop, business analytics (BA) research to improve organiza-
tional decision-making has gained significant traction. The origin of BA is deeply
rooted in operations research (OR), commonly linked with decision support sys-
tems. Subsequently, the universal adoption of integrated information systems
(IS) has enabled organizations to accumulate substantial quantities of data, cul-
minating in the emergence of concepts such as business intelligence (BI) and,
more contemporarily, big data analytics (Mikalef etal. 2018). Despite the field’s
evolving landscape, the core objective of BA has remained steadfast: to delegate
analytical tasks to IS to fortify and streamline decision-making processes within
organizations. BA shapes well-informed decisions by providing decision-makers
with accurate, comprehensive, and timely information, ultimately driving organi-
zational performance and fostering a competitive advantage in the ever-changing
business environment (Holsapple etal. 2014; Mikalef etal. 2020).
Prescriptive Analytics Systems (PAS) embody the most advanced iteration of
IS utilized within BA and surpass the capabilities of descriptive analytics, which
focus on understanding historical data, and predictive analytics, which forecast
the likely future. PAS are designed to guide the best action, considering vari-
ous factors and constraints to achieve desired outcomes. These systems lever-
age descriptive and predictive analytics results as inputs, harnessing the insights
derived from past events and probable future scenarios to inform their recommen-
dations (Lepenioti etal. 2020).
In recent years, research activity has gained traction, with various technologi-
cal innovations significantly influencing the design of PAS, foremost machine
learning and artificial intelligence (AI), such as deep learning, reinforcement
learning, and biologically inspired algorithms (Lepenioti et al. 2020). Survey
papers keep track of these developments by classifying and conceptualizing PAS
aspects from different perspectives. However, their primary focus is often on
algorithmic facets. Further, analytics has been predominantly viewed as a pas-
sive tool to be used by the human decision-maker (e.g., Frazzetto et al. 2019;
Poornima and Pushpalatha 2020; Lepenioti etal. 2020).
Recently, there has been a noticeable shift in BA research, with researchers and
practitioners increasingly examining the broader implications of analytics sys-
tems. The synergistic relationship between analytics or AI systems and their users
in decision-making processes has emerged as a critical area of investigation (e.g.,
Rzepka and Berger 2018; Niehaus and Wiesche 2021; Hinsen etal. 2022). The
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
dynamics in the relationship between analytics tools and human decision-makers
are changing drastically, with increased delegation between the two and shifting
responsibilities, with analytical systems taking agency and ownership of critical
steps in the decision-making process. Specifically, PAS demonstrate a significant
bidirectional relationship and an expansive decision-making latitude compared to
other analytical systems, which are rather passive, reactive, or anticipatory. Such
prescriptive agents can act as human partners or substitutes for behavior-based
or outcome-based decision-making (Baird and Maruping 2021). We argue that
comprehending these factors within the IS community is imperative for illustrat-
ing the integration of algorithmic or technical elements into an overarching view,
especially given the heightened interest in PAS and their anticipated expansion in
research and practice, necessitating the consolidation of the existing knowledge.
To this end, the IS artifact is an established theoretical framework underpinned
by general systems theory (GST) to describe, design, or examine systems in a
broader organizational context (Chatterjee etal. 2021). Following socio-technical
thinking, an IS artifact comprises two closely interrelated and connected subsys-
tems: the social system, with humans as its central component, and the techni-
cal system, encompassing elements such as technical infrastructure, hardware,
and software. The subsystems are nested in an open system that receives inputs
and produces outputs in an environment (i.e., organizational or industry context).
They are in synergy with each other and adaptable, meaning they can change over
time. Thereby, IS artifacts are intentionally designed to meet a specific objective
(Bostrom and Heinen 1977; Niehaus and Wiesche 2021; Chatterjee etal. 2021).
We endeavor to build on this perspective and advance the understanding of
PAS with three research objectives nested along the principles of GST (Chatter-
jee etal. 2021):
(1) First, we aim to consolidate existing literature on PAS, clarifying the essential
subsystems, their constituent components, and their interplay and connection to
the decision environment. This effort lays the foundation for future research and
is crucial in bridging the current knowledge gap. Understanding these aspects
is pivotal for successfully deploying PAS, particularly when considering their
integration into organizational decision-making processes.
(2) Second, expanding upon the consistent components, we aim to explore the PAS
artifacts from a meta-level perspective. We seek to determine whether the litera-
ture unveils recurring archetypes or system designs with distinct characteristics,
emphasizing the synergy or delegation between the human decision-maker and
prescriptive agent and the degree of adaptation (Baird and Maruping 2021).
Here, we fall back on human decision-making to better understand how PAS
can support organizational decision-making.
(3) Third, as a final objective of our study, we seek to discern the action potential,
defining a technology’s capabilities to an individual, organization, or industry for
a particular purpose. For this, we fall back on affordance theory, which focuses
on the action possibilities arising from the relationship between technologies
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
and their users, often referred to as technology affordances (e.g., Anderson and
Robey 2017; Mettler etal. 2017; Effah etal. 2021).
In summary, presenting this synthesized perspective, we posit that it paves the
way for subsequent investigations into the core dimensions of PAS within a GST
framework, which is especially important for the IS community, effectively setting a
research agenda.
To achieve this, we conduct a systematic literature review (SLR), adhering
to established methods in IS research (Cooper 1988; Webster and Watson 2002;
vomBrocke etal. 2009, 2015). This approach ensures a comprehensive and rigorous
examination of pertinent studies, allowing us to derive meaningful insights and iden-
tify knowledge gaps in the field of PAS. The paper’s structure follows: Sect.2 pro-
vides the research context, encompassing decision-theoretic foundations, BA, GST,
and related studies. Section3 introduces our SLR methodology, detailing the steps
to analyze the relevant body of work. Section4 presents the results of our review,
focusing on the key concepts and findings that emerged from the literature. Sec-
tion5 discusses the implications of our results, highlighting potential future research
streams and avenues for further exploration. Finally, in Sect.6, we offer conclud-
ing remarks, summarizing the contributions of our study and its implications for the
field of PAS.
2 Background
To facilitate comprehension of the core facets, in the following sections, we examine
human decision-making, BA’s role in improving organizational decision-making,
and introducing prescriptive analytics in the GST context as an IT artifact. Subse-
quently, we assess related studies to identify research gaps and underscore the neces-
sity for further investigation.
2.1 Decision‑making
Decision-making is fundamentally a biological process rooted in evolution (San-
tos and Rosati 2015), and decision theory as a research area focuses on examining
human choice-making. This field is typically divided into two interrelated aspects
(Slovic et al. 1977): normative theory, which prescribes ideal decision-making
behavior, and descriptive theory, which describes actual human behavior. In the con-
text of our research questions, the normative theory is of greater relevance, as it
assumes decision-makers adhere to rules for consistent and optimal outcomes under
given conditions. In practical terms, specifically in IS research, normative theory
aims to develop tools that enhance human decision-making (Straub and Welpe
2014).
In this context, human decision-making generally follows a systematic process.
While interpretations may vary across domains, the fundamental structure remains
consistent (e.g., Simon 1960; Svenson 1992; Schoenfeld 2010; Darioshi and Lahav
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
2021; Darioshi and Lahav 2021), also in an organizational setting (e.g., Trunk etal.
2020). The process begins with problem identification, followed by alternative genera-
tion, evaluation, selection of the most suitable option, decision execution, effectiveness
assessment, and iteration for similar problems. This iterative approach constitutes a
continuous learning process, yielding increasingly optimized results over time.
Based on the different interpretations, several authors break down the process into
overarching phases (e.g., Ren etal. 2006; Leyer etal. 2020). In our work, we adopt
a triphase decision-making process consisting of the stages before, during, and after
the decision (refer to Fig.1). Phase 1, evaluation of alternatives, encompasses problem
identification, alternative generation, and ranking. Phase 2, decision-making, involves
selecting the most appropriate alternative based on the situation, considering the loss
and utility of potential consequences, and executing the decision. Phase 3, adaptation
and learning, entails assessing the effectiveness of the outcomes in order to modify
behavior for subsequent iterations. The triphase process and its components are crucial
in designing a PAS and serve as a guiding framework in the synthesis of this study.
2.2 Business analytics
The rapid data volume growth in recent times made BA and Big Data Analytics (BDA)
central topics in IS and e-business research (e.g., Pappas etal. 2018; Mikalef et al.
2020; Jensen etal. 2023). Analyzing extensive and diverse data can offer organizations
a competitive edge, help achieve strategic and tactical goals, and enhance operational
performance by optimizing decision-making processes (Holsapple etal. 2014; Knabke
and Olbrich 2018; Oesterreich etal. 2022; Shiau etal. 2023). However, the sheer vol-
ume of big data, coupled with uncertainty and noisiness, renders it non-self-explan-
atory (Lepenioti et al. 2020). Consequently, extracting value from data necessitates
sophisticated techniques, processes, and practices. BA, in this context, is a multidimen-
sional and interdisciplinary concept, drawing on technologies from computer science
and engineering, quantitative methods from mathematics, statistics and econometrics,
and decision-theoretic aspects from psychological and behavior sciences (Mortenson
etal. 2015).
BA can be conceptualized using domain, technique, and orientation (Holsapple etal.
2014). The domain (i) pertains to the context in which BA is applied (e.g., a decision
problem in manufacturing). The second dimension, technique (ii), denotes the meth-
ods employed to perform an analytics task, such as linear programming or specific AI
or ML techniques. Lastly, orientation (iii) characterizes the objective or direction of
thought, addressing questions like ‘what does analytics do?’ or ‘why is it performed?’
and can be regarded as the central dimension. A commonly utilized taxonomy to illus-
trate the orientation of BA applications is the categorization into maturity levels, which
Evaluation of alternatives (1) Decision-making(2) Adaptation and learning (3)
Fig. 1 General phases of decision-making processes
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
consist of three levels based on their potential and complexity: descriptive analytics,
predictive analytics, and prescriptive analytics (e.g., Delen and Ram 2018; Frazzetto
etal. 2019; Lepenioti etal. 2020).
The three maturity levels create a synergistic relationship, as depicted in Fig.2.
Descriptive analytics focuses on the past and present by answering questions such as
‘What is happening?’ or ‘What happened?’ utilizing traditional BI techniques like
Online Analytical Processing (OLAP) or data mining (Delen and Zolbanin 2018). In
contrast, predictive analytics anticipates the likely future by addressing the question
‘What will happen?’ and employs ML methods, such as classification and regres-
sion models (Lepenioti etal. 2020). Prescriptive analytics seeks to identify optimal
decisions, recommendations, or actions by tackling the question, “What should be
done?” (Delen & Ram 2018). This advanced approach employs sophisticated analyt-
ics, operations research, and machine learning techniques, including deep learning,
mathematical programming, evolutionary computation, and reinforcement learning
(Lepenioti etal. 2020).
The full potential of predictive analytics can only be harnessed when combined
with prescriptive analytics, which streamlines decision-making processes proac-
tively. Reducing the time interval between event prediction and proactive decision-
making is paramount to maximizing business value. Prescriptive analytics generates
well-informed decisions based on the outcomes of predictive analytics, considering
the most suitable timing for executing actions preceding the anticipated event. On
the other hand, descriptive analytics can be utilized after the event to scrutinize its
underlying causes and consequences while operating on diverse timescales for reac-
tive or long-term actions. In this regard, the prompt detection of the current state
and precise forecasting of emerging events are crucial factors in mitigating potential
losses in business value (Krumeich etal. 2016; Lepenioti etal. 2020).
2.3 Prescriptive analytics systems asanISartifact
The concept of an IS artifact remains ambiguously defined (Chatterjee
et al. 2021). However, consensus suggests that it can direct research, clarify
Fig. 2 An overview of the synergies and dynamics of descriptive, predictive, and prescriptive analytics
(Krumeich etal. 2016; Lepenioti etal. 2020)
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
understanding, set boundaries, provide a design framework, and foster novel
research perspectives, among other applications (Orlikowski and Iacono 2001;
Aier and Fischer 2011; Chatterjee etal. 2021).
One way of conceptualizing an IS artifact is with GST (Kast and Rosenz-
weig 1972; Chatterjee et al. 2021). Within this theoretic framework, and draw-
ing upon socio-technical thinking, two principal subsystems emerge: the social
and the technical (Sarker etal. 2019). The social subsystem is characterized by
its components, encompassing individuals with their inherent knowledge, skills,
and values, as well as structural facets like organizational hierarchies and reward
systems. Conversely, the technical subsystem is described as an assembly of con-
stituent technical components, such as hardware, software, or methodologies,
to transmute inputs into outputs, enhancing the performance of an organization
(Bostrom and Heinen 1977; Chatterjee etal. 2021). These subsystems interact in
synergy, allowing for the exchange of information to fulfill mutual objectives or
purposes. Situated as an open system, an IS artifact is embedded within its envi-
ronment (i.e., in an organizational or industry context), influenced by external
factors, and concurrently impacting its surroundings. Central to its design is the
adaptability of its subsystems, ensuring stability amid changes (Kast and Rosenz-
weig 1972; Sarker etal. 2019; Chatterjee etal. 2021).
Further, Chatterjee etal. (2021) underscore the significance of examining the
interactions between subsystems using an affording-constraining lens. The affor-
dance theory, introduced initially by Gibson (1986), identifies action possibilities
stemming from the relationship between an object and its observer. In IS research,
this concept translates to technology affordances, highlighting the potential
actions enabled by the relationship between technologies and their users (Ander-
son and Robey 2017; Mettler etal. 2017; Leidner etal. 2018). Here, affordance
can be defined as”what an individual or organization with a particular purpose
can do with a technology” (Majchrzak and Markus 2013), further emphasized by
Markus and Silver (2008), who describe them as a user’s interaction potential
with a technical object.
The IS artifact as a theoretical concept will guide our SLR, so we elucidate the
aspects and their interplay in an exemplary and simplified PAS-supported organi-
zational decision-making problem within manufacturing, precisely, maintenance
operations (e.g., Liu etal. 2019; Ansari etal. 2019; Gordon etal. 2020; Wanner
etal. 2023), as illustrated in Fig.3.
At its core, the social subsystem encompasses the individuals involved in the
decision-making process, serving as decision-makers responsible for managing
and overseeing maintenance processes. Concurrently, the prescriptive agent, as
the technical subsystem, comprises the infrastructure, analytics models, and visu-
alization tools to present findings to decision-makers based on inputs from the
decision environment. The decision-maker interacts with the technology compo-
nents. This interplay affords the optimal maintenance schedules to the user, which
then can be actioned upon and implemented in the environment, for instance, a
production line with multiple machines.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
2.4 Related work
Given the significant research interest in the field, several studies have investigated
key topics related to prescriptive analytics. Our analysis emphasizes the impor-
tance of our research objectives, and we compile findings, including the purpose
addressed in the related work (cf. Table1). We considered existing systematic and
unstructured literature reviews to ensure a well-rounded understanding.
Previous research has primarily concentrated on the technical subsystem of PAS,
analyzing its technology components and affordances or applications. For instance,
Lepenioti etal. (2020) begin their review with an in-depth analysis of predictive
and prescriptive analytics methods before outlining challenges and future directions.
Meanwhile, Frazzetto etal. (2019) take a system-oriented approach, emphasizing
the various features of PAS, including productivity, infrastructural considerations,
and analytical capabilities. Vanani etal. (2021) focus specifically on employing deep
learning algorithms in PAS in the Internet of Things, while Stefani and Zschech
(2018) provide a conceptualization that considers decision theory as a fundamen-
tal aspect. They consolidate various perspectives to derive technology components.
Lastly, Poornima and Pushpalatha (2020) adopt an application-oriented approach,
providing a comprehensive overview of the usage of PAS in diverse industries.
Despite their differences, these studies collectively emphasize the importance of
considering various technical factors when developing PAS.
Aside from general reviews of PAS, some studies focus on specific industries
or contexts. For instance, Fox etal. (2022) emphasized the importance of mainte-
nance tasks in wind farms and conducted a PAS review specifically for this indus-
try. Soeffker etal. (2022) identified unique requirements for dynamic vehicle
Environment (e.g., manufacturing line with production machines)
IS artifact (e.g., PAS)
Social Subsystem
Decision-Maker
(e.g., maintenace
engineers)
Technical Subsystem
(e.g., prescriptiveagent)
Component
(e.g., Visualization)
Component
(e.g., Infrastructure)
Component
(e.g., analytics model)
Affording/constraining
relationship
(e.g., affords assistance
by prescribing
maintenance schedules)
Input
(e.g., machine sensor data, spare parts inventory)
Output
(e.g., maintenance actions)
Fig. 3 IS artifact with exemplary PAS-supported decision-making problem (own depiction based on
Bostrom and Heinen 1977; Chatterjee etal. 2021)
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
Table 1 Previous literature reviews on prescriptive analytics
Comparison element Data collection/method
(literature reviewed)
Purpose
Lepenioti etal. (2020)SLR (N = 56) Synthesis of analytics techniques and methods and proposition of a research agenda
Frazzetto etal. (2019) Unstructured review Analysis of technical components for analytics, infrastructure, and productivity
Vanani etal. (2021) Unstructured review Prescriptive analytics, specifically using deep learning in an IoT context
Stefani and Zschech (2018)SLR (N = 30) Overview of PAS technology components in a coherent view, incorporating aspects from decision theory
Poornima and Pushpalatha (2020) Unstructured review Overview of application domains, including analytics methods and infrastructure considerations
Fox etal. (2022) Unstructured review PAS for wind-farm operation and maintenance, including perspectives on analytics methods and technol-
ogy, but also insight into affordances related to the context
Soeffker etal. (2022) Unstructured review Conceptualization of decision models, prescriptive methodologies, and computational methods
Bhatt etal. (2023)SLR (N = 147) PAS use-cases for sustainable operations research
Kubrak etal. (2022)SLR (N = 37) PAS for process monitoring, including aspects such as purpose/objective, data considerations, and ana-
lytics techniques summarized in a framework
This Study SLR (N = 262) PAS from an IS perspective, constituent components of subsystems, a meta-system level view, and
technology affordances
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
routing and reviewed relevant literature. Bhatt etal. (2023) provided a frame-
work for developing PAS in sustainable operations by identifying five application
themes. Meanwhile, Kubrak etal. (2022) explored challenges and suggested areas
for future research to enhance the usefulness of prescriptive process monitoring
methods. These studies highlight the significance of context-specific considera-
tions and affordances in developing and applying PAS.
In summary, much of the existing research on PAS has been primarily anchored
in its technical components and underlying concepts. This review seeks to weave
these diverse strands of thought, capitalizing on the foundational works to offer
a more holistic perspective. Specifically, we aim to synthesize the current land-
scape of prescriptive analytics, positioning it as an IS artifact within the broader
context of the decision-making process and revealing the delegation of tasks and
responsibilities of both the human decision-maker and the prescriptive agent.
3 SLR methodology
In this section, we employ the established SLR methodology in IS research,
as outlined by vom Brocke et al. (2009; 2015), incorporating extensions from
Cooper (1988) and Webster and Watson (2002). This method consists of five
phases: (1) definition of review scope, (2) conceptualization, (3) literature search
process, (4) literature analysis and synthesis, and (5) research agenda. Further,
we take a descriptive approach to show the current understanding of the literature
and reveal patterns, trends, or gaps in current PAS research (Paré etal. 2015).
In this section, we begin by addressing the review scope and conceptualiza-
tion of the topic, drawing on the theoretical foundations from the previous sec-
tion. Subsequently, we introduce the literature search process, followed by an ini-
tial analysis of the literature sample. The synthesis results will be presented in
Sect.4, while Sect.5 of this paper will cover the research agenda.
3.1 Scope
To define and present the scope of our SLR, we employed Cooper’s (1988) tax-
onomy with six dimensions. Our (1) focus encompasses research outcomes and
applications, including mathematical, conceptual, technological, and infrastruc-
tural contributions related to using PAS to understand their aspects better. The (2)
goal of our review is to integrate GST perspectives by (3) organizing the results
conceptually, following the method outlined by Webster and Watson (2002).
We aim for a (4) neutral representation to reveal the current state of PAS-based
research. Our review targets a (5) diverse audience, including IS scholars, practi-
tioners, and specialized researchers from the BA community. Lastly, we endeavor
to provide a (6) representative coverage of the relevant literature.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
3.2 Conceptualization
In our conceptualization, we draw upon the research context and related work and
utilize the GST-based IS artifact to guide the organization and structure of our
review and its findings. We aim to uncover crucial concepts and elements within
PAS to enable a research launchpad. To effectively address our research goals, we
will divide our SLR into three distinct foci:
• Constituent components: Understanding the constituent components of a PAS
is essential. Here, we adopt GST, precisely the notion that IT artifacts can be
viewed as transformational models, receiving inputs, transforming them, and
generating outputs (Kast and Rosenzweig 1972; Chatterjee etal. 2021).
• System archetypes: We aim to explore the meta-system level of PAS artifacts
by building on the constituent elements. We are keen to ascertain if the literature
reveals recurrent PAS archetypes or designs marked by unique features, spot-
lighting the delegation and responsibilities between the human decision-maker
and prescriptive agent and the degree of adaptation. To achieve this, we lean into
the decision-making process as a lens to better understand how PAS are nested
here.
• Technology affordances: From an IS perspective, we aim to understand the
purposes for which PAS are implemented and applied. Focusing our analysis on
industry or industry-agnostic use cases and breaking the investigation into spe-
cific technology affordances, we will delve deeper into how PAS contribute to
and influence decision-making tasks across diverse industry and organizational
settings.
3.3 Literature search process
The third phase of the SLR methodology, the literature search process, consists
of three subphases: database search, keyword search, and backward and forward
search. We outline our procedure as follows (cf. Figure4).
Initially, we selected interdisciplinary databases such as Web of Science and
Scopus, technology-related databases like ACM Digital Library and IEEE Xplore,
and AISeL for IS-related outlets. Our search string combined the term ‘prescriptive’
with several core concepts derived from prior survey articles (Stefani and Zschech
2018; Frazzetto etal. 2019; Lepenioti etal. 2020) (cf. Appendix A for details).
This search yielded 2,597 results (date of search: March 30, 2023). Two research-
ers collaborated to analyze and screen the papers, using existing conceptual and
review papers to establish a shared understanding. Inclusion criteria were set (cf.
Table2) to consider only papers that explicitly address the overall design of PAS
or describe specific PAS elements, components, or properties. For example, this
includes a diverse spectrum of studies, such as conceptual (Levasseur 2015; e.g.,
Appelbaum etal. 2017), review (e.g., Poornima and Pushpalatha 2020; Lepenioti
etal. 2020), technological/architectural (e.g., Vater etal. 2019; Basdere etal. 2019),
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
or mathematical papers (e.g., Bertsimas and Kallus 2020; Elmachtoub and Grigas
2022). By contrast, we aimed to exclude papers that only briefly mentioned pre-
scriptive analytics without providing more detailed descriptions (e.g., Swaminathan
Search String:
(prescriptive)
AND
(analytics OR model OR machine learning OR optimization OR evolutionary OR expert system OR heuristics OR simulation OR
artificial intelligence).
Database
Web of Science
(1509)
Scopus
(292)
IEEE Explore
(487)
ACM DL
(73)
AISeL
(236)
Running string in database
= 2597
n= 799
n= 350
n= 198
n= 262
Title and keyword analysis (removed
1798)
Abstract analysis and check for
duplicates (removed 449)
Full-text analysis (removed 152)
Citation chaining (backward and
forward) (added 64)
Fig. 4 The literature search process
Table 2 Inclusion and exclusion criteria used in the SLR
Inclusion criteria Exclusion criteria
• Explicitly addresses the overall design or
specific elements, components, or properties of
prescriptive analytics systems
• Not written in English
• Published before 2010 (and, thus, does not discuss
contemporary PAS)
• Is a presentation or a research-in-progress paper
without significant interim results
• Prescriptive analytics is only mentioned without
providing more detailed descriptions
• Non-academic papers (e.g., practical case studies
or reports)
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
2018; Pereira etal. 2021). We also excluded non-academic articles. However, we
did not limit our search to only high-ranking journals and conferences to ensure a
comprehensive SLR. Webster and Watson (2002) argue that a topic-centric view of
the literature is more valuable than a view limited to a few top journals. Further, we
excluded articles that were not written in English.
To incrementally exclude irrelevant papers, we applied a stepwise procedure.
First, a title and keyword analysis identified 799 relevant articles. After reviewing
abstracts and removing duplicates, we removed 449, reducing the number of papers
to 350. Full-text screening further reduced the number of relevant articles to 198. To
supplement our findings, we conducted citation chaining, both forward search (via
Google Scholar) and backward search (via bibliography), adding 64 relevant papers
and increasing the total to 262 articles. The complete list of the literature sample is
available in Appendix B.
3.4 Overview oftheliterature sample
Our findings indicate a significant growth in research interest in prescriptive ana-
lytics. More than half of our sample was published after 2020, demonstrating the
increasing relevance of this area of research. Concerning publication types, approxi-
mately 60% of articles are in journals, and 34% of our sample comprises conference
papers. A smaller proportion, 6%, is book sections or chapters.
An initial sample analysis allowed us to explore thematic trends by observing the
top research outlets. Much of the sample is published in IEEE and ACM proceed-
ings focusing on computer science and technology. Similarly, we observed a clear
indication of the operations research and management domain, with publications in
journals such as Management Science, European Journal of Operations Research,
and others at the intersection between computing, operations research, and industrial
engineering (Fig.5).
In 2021 and 2022, we noticed a considerable drop in conference papers, possibly
due to widespread lockdowns in light of the COVID-19 pandemic. Nonetheless, the
publication of numerous journal papers during this period contributed to the con-
tinued growth of interest in those years. In summary, after initial observation of our
Overview of literature sample, including publication type Top 13 research outlets
IEEE proceedings (various
)2
1
European Journal of Operations Research 9
INFORMS Journal on Applied Analytics 9
ACM proceedings (various
)9
Management Science 6
IEEE Acces
s6
Annals of Operations Research 4
Sustainability 3
Production and Operations Managemen
t3
Expert Systems with Applications 3
Decision Support Systems 3
Computers & Operations Research 3
Computers & Industrial Engineerin
g3
0
10
20
30
40
50
60
70
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022/2023
Book chapters Conference proceedings Journals
Fig. 5 Overview of literature sample and top research outlets
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
literature sample, it is evident that the research is heavily weighted toward techno-
logical and mathematical disciplines, which is to be expected given the core of pre-
scriptive analytics.
4 Results
Below, we showcase the findings ofthe synthesis. We will begin by detailing the
constituent components, system archetypes, and finally, the technology affordances
of PAS.
4.1 Constituent components
We observed a predominant emphasis on technical components within our literature
sample, characterized by its technical focus and the inherent nature of prescriptive
analytics. The decision-maker naturally emerges as the pivotal entity in the social
subsystem. However, a detailed exploration of the decision processes or structures
surrounding PAS-based decision-making is absent in current research, with just a
few authors focusing on the decision-maker. For example, Käki etal. (2019) inves-
tigate the deviations of decision-makers from model-based recommendations and
their impact on the effectiveness of decision-support processes. By examining these
discrepancies and discerning the underlying motivations, the authors emphasize the
potential for improving the planning process, optimizing model-driven decision-
making, and refining the lifecycle management of PAS. Similarly, Caro & de Tejada
Cuenca (2023) study the adherence to prescriptive analytics recommendations, high-
lighting trust as a deterrent, with interpretability as a crucial intervention.
Consequently, the following sections will mirror this emphasis on the technical
subsystem and its components. Here, we present a multi-layered concept matrix
with 23 concepts. The basic structure of the concept matrix follows GST, that IT
artifacts can be viewed as transformational models, receiving inputs, transforming
and processing them, and generating outputs (Kast and Rosenzweig 1972; Chatter-
jee etal. 2021), which we coin “decision formulation”, “decision input”, “decision
processing”, and “ decision output” in our study. We added “ancillary components”
to address additional aspects (Frazzetto etal. 2019) that are not situated in the core
of the prescriptive agent but support its integration into organizational decision-
making processes and structures.
Table3 presents a concise summary of the outcomes derived from the concept
matrix, featuring exemplary studies corresponding to each concept and the num-
ber of hits where the concepts are discussed or mentioned in the text corpus. A
detailed overview of all concepts correlated with the literature sample can be found
in Appendix C. Further, we added a trend analysis in Appendix H, visualizing the
concepts’ development across the years.
Each concept (emphasized in bold) is expounded upon in greater detail in the
subsequent sections. Further, Fig.6 integrates and summarizes the findings, offering
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
Table 3 Concept matrix of constituent components
Constituent components of PAS Example studies Frequency
Decision formulation Decision variables [7], [19], [28], [48], [53], [69], [110], [117], [176], [243] 147
Objectives [4], [27], [45], [56], [76], [110], [146], [166], [195], [262] 158
Constraints [8], [33], [45], [72], [92], [117], [152], [185], [217], [243] 137
Decision input Current state [15], [30], [53], [67], [90], [105], [130], [163], [200], [236] 133
Probabilities [13], [27], [58], [67], [92], [105], [139], [162], [202], [234] 201
Decision processing Mathematical programming [10], [35], [42], [74], [97], [131], [152], [178], [204], [233] 149
Machine learning [18], [34], [45], [58], [67], [84], [110], [148], [198], [226] 171
Evolutionary computation [59], [92], [176], [190], [18], [145], [128], [83], [67], [229] 27
Simulation [19], [61], [72], [87], [92], [101], [117], [139], [197], [213] 67
Logic-based models [90], [113], [118], [165], [171], [190], [204], [211], [226] 12
Probabilistic models [21], [27], [50], [73], [90], [110], [129], [144], [168], [229] 38
Decision output Single decision [14], [118], [183], [29], [65], [56], [131], [40], [178], [142] 128
Multiple decisions [4], [84], [92], [115], [136], [190], [211], [213], [222], [231] 21
Action mechanisms Execution [21], [54], [60], [91], [98], [112], [163], [192], [202], [223] 25
Adaptation [16], [40], [60], [79], [92], [123], [162], [194], [196], [220] 34
Ancillary features Integration [9], [31], [35], [36], [38], [49], [67], [125], [163], [227] 48
Distributed computing [12], [23], [30], [31], [54], [114], [189], [194], [208], [223] 16
Modulization [14], [25], [34], [54], [79], [92], [112], [123], [194], [233] 80
Security- and privacy-preserving [34], [38], [59], [75], [82], [98], [159], [182], [190], [210] 10
Workflow interface [6], [12], [30], [38], [65], [82], [92], [100], [173], [218] 43
Explainability [84], [147], [203] 3
Visualization [3], [6], [38], [65], [84], [91], [104], [132], [189], [257] 71
Extensibility [30], [40], [49], [73], [189], [208], [218], [236] 8
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
a cohesive representation of the PAS while elucidating the interactions among its
core components.
4.1.1 Decision formulation
Decision formulation refers to the essential elements for structuring a decision prob-
lem, subdivided into decision variables, objectives, and constraints. Decision vari-
ables define the object of interest within a decision (Stefani and Zschech 2018). For
instance, production planning may involve mapping the manufacturing workforce,
machinery, and material flow allocation to each other most profitably (Elmachtoub
and Grigas 2022). In this context, the complete set of all potential mappings consti-
tutes the set of all alternatives or competing decisions. When considering in con-
junction with the encompassing environmental conditions and various contextual
factors, the specification of decision variables plays a pivotal role in delineating spe-
cific states and their corresponding outcomes. These states frequently exhibit asso-
ciations with utility values, encompassing metrics such as costs, profits, or revenues,
thus serving as quantitative indicators for overarching objectives that demand either
minimization or maximization. These objectives are commonly denoted as objective
functions, optimization functions, or simply objectives. Additionally, it is pertinent
to acknowledge the presence of constraints, which often encircle decision spaces,
Current stateProbabilities
Decision
variables
Constraints Objectives
Single
decision
Multiple
decisions
Prescriptive Agent
Execution mechanism
Decision-processing technique(s)
Adaptation mechanism
Workflow
interface
Visual
interface
Decision-maker
Action in
environment
Automated
action
Human action
Tracking of consequences
Results from descriptive
and predictive analytics
Decision environment
Fig. 6 Exemplary visualization of constituent components of PAS in a coherent view
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
emanating from natural limitations (e.g., capacity limitations of a machine) or man-
agerial imperatives (Stefani and Zschech 2018; Elmachtoub and Grigas 2022).
4.1.2 Decision input
The input is the foundational component in data-driven decision-making, encom-
passing essential attributes of analytics processes. While not intrinsic to the core
components, numerous authors emphasize these attributes in the context of PAS (cf.
Appendix G for an overview). These properties encompass the structural charac-
teristics of the data (Lash and Zhao 2016), its origin, whether external or internal
(Bertsimas and Kallus 2020), and the manner of data generation, whether reliant
on human-based assumptions, empirical methods or synthetic means (Stefani and
Zschech 2018; Ceselli etal. 2019). Additionally, consideration is given to the veloc-
ity of data, differentiating between historical and real-time data (Krumeich et al.
2016; Miikkulainen et al. 2021). Subsequently, data undergoes various preproc-
essing or data engineering steps to prepare the data input for analytical processing
(McFowland III etal. 2021), often handed over to descriptive and predictive func-
tions, resulting in the current state and probabilities. Many authors see these preced-
ing analytical results as a foundation of PAS (e.g., Wang etal. 2018; Miikkulainen
etal. 2021), thus making them constituents.
Descriptive analytics provide insights into the current state, such as patterns or
key performance indicators, to assess the existing conditions within the decision
context. This information helps identify areas that require modification compared
to the current state and serves as a baseline for evaluating the ramifications, e.g., in
terms of gains or losses, of a decision (Li etal. 2021). Moreover, it is imperative to
recognize that decision problems inherently encompass a degree of uncertainty. This
uncertainty may be effectively quantified by utilizing probabilities derived from
predictive analytics, facilitating an elucidation of the likelihood associated with the
impending occurrence of a particular outcome. Probabilities frequently serve as
integral components, directly integrated as weightings within the delineation of the
objectives within the overarching decision framework (Stefani and Zschech 2018;
Wang etal. 2018; Lepenioti etal. 2020; Miikkulainen etal. 2021).
4.1.3 Decision processing
Both the input and formulation frame the decision processing, correspondingly gen-
erating prescriptions. Here, the literature refers to various techniques, which can
be grouped into mathematical programming, evolutionary computations, machine
learning, probabilistic, and logic-based models, which are not mutually exclusive
and can be utilized interactively or sequentially, confirming the results of prior lit-
erature reviews (Lepenioti etal. 2020). Additionally, we refer to Appendix F for an
overview of more detailed techniques and potential subcategories.
Mathematical programming is widely adopted for optimizing objective func-
tions within constrained solution spaces (e.g., McFowland III etal. 2021). In con-
trast, evolutionary computations offer bio-inspired optimization techniques (e.g.,
Miikkulainen etal. 2021), while machine learning (ML) algorithms enable learning
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
without explicit instructions (Janiesch etal. 2021). Supervised ML facilitates antici-
patory decision-making by predicting unseen data (Lash and Zhao 2016). As a sub-
set of ML, reinforcement learning (RL) aims to maximize cumulative rewards in
given environments, proving effective for well-formalized decision-making prob-
lems (Lepenioti et al. 2021). Probabilistic models, such as Bayesian inference
or Markov models, calculate event likelihoods and represent causal relationships
(Lepenioti et al. 2020). Logic-based models examine chains of cause-and-effect
relationships leading to specific outcomes (Lepenioti etal. 2020). Lastly, simula-
tions enable exploring hypothetical or real-life processes to improve decision-mak-
ing by generating scenarios and uncovering optimal behaviors for specific situations
(Lepenioti etal. 2020).
4.1.4 Decision output
The decision output refers to the result or outcome produced in decision process-
ing. It represents the prescribed course of action or solution from competing deci-
sions based on the decision variable (e.g., all alternative configurations of human
resources, machinery, and equipment). While the optimal or single decision is typi-
cally required, authors emphasize the importance of making the alternatives or mul-
tiple decisions transparent to the decision-maker and accommodating more com-
plex situations, such as dynamic environments. For example, the prescriptions are
tailored to sequential process stages, varying time points, or ever-changing states,
ensuring a more adaptive response (e.g., Liu etal. 2019; Brandt etal. 2021).
4.1.5 Action mechanisms
Following the decision-making process, the human decision-maker traditionally
performs the subsequent actions within the decision environment. In this context, it
is essential to emphasize that the implications and outcomes are detached from the
technical subsystem, which primarily functions as a passive tool at the user’s dis-
posal. However, recent advancements in AI and ML have triggered a notable shift in
IS research (Baird and Maruping 2021). This shift acknowledges the agency of the
IS artifact and promotes synergistic collaboration between the technical and social
subsystems. In the context of PAS, this transition necessitates the development of
specific mechanisms to facilitate action execution, tracking, and adaptation, enabling
the technical components to engage in the decision environment autonomously.
Implementing execution mechanisms is essential to facilitate actions within the
decision environment, distinguishing between two primary execution modes: (i)
autonomous execution, where the prescriptive agent independently carries out the
decision (e.g., Mazon-Olivo etal. 2018; Soroush etal. 2020), and (ii) execution with
human intervention, which may require human confirmation of the decision output
before implementation (e.g., Rizzo etal. 2020). As automation levels increase, the
importance of tracking becomes more pronounced to document the actions taken in
the decision environment, leading to adaptation mechanisms to change iteratively
by using insights from tracked actions and their outcomes (e.g., Bousdekis etal.
2020; Zhang etal. 2021),. This adaptation is driven by the implications of decisions
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
made within the decision environment and dynamic shifts in decision inputs, par-
ticularly in dynamic and evolving contexts. Also, this can be done autonomously
by observing the decision-maker or the outcomes in the decision environment (Liu
et al. 2019; Tamimi et al. 2019), but also with the decision-maker’s input (e.g.,
Krumeich etal. 2016; Kim etal. 2020; Vater etal. 2020; Miikkulainen etal. 2021).
Action mechanisms are crucial in distinguishing between the PAS archetypes we
identified in our synthesis. Therefore, we will delve deeper into these aspects in the
respective Sect.(4.2).
4.1.6 Ancillary features
In addition to the formulation, input, processing, and action mechanism compo-
nents, our literature review has revealed further concepts within the context of PAS.
We term these concepts "ancillary" since they are not at the core but somewhat sec-
ondary to PAS functionality. The literature encompasses many features, from data
properties and general infrastructure considerations to integrating PAS within manu-
facturing systems (e.g., Vater etal. 2019; Ansari etal. 2019; Consilvio etal. 2019).
Our focus, however, remains on overarching aspects and concepts that align with
our research objectives. Specifically, we concentrate on features directly connecting
to how PAS is situated within the organizational context and their role in decision-
making processes.
Integration is a central concept determining a PAS’s positioning in the broader
(inter-)organizational landscape. Vertical integration allows data and decision out-
puts to be available across hierarchical levels, while horizontal integration incorpo-
rates data throughout processes or business functions, even externally (Appelbaum
etal. 2017; Kumari and Kulkarni 2022). The growing data volume and sophisti-
cated algorithms demand increased computing power, addressed through distrib-
uted computing, linking computational resources for shared data and processing
power (Lepenioti etal. 2020). Modularization reduces complexity by, for instance,
separating descriptive, predictive, and prescriptive analytics or specific functions
(Appelbaum etal. 2017; Frazzetto etal. 2019). Additionally, security- and privacy-
preserving features, though a niche in current research, are crucial due to rising
cyber threats. For example, Harikumar etal. (2022) propose an algorithm for private
prescription vectors.
With a focus on the decision-maker’s perspective, our review has unearthed
studies discussing explainability within PAS-based decision-making. The objec-
tive is to bolster user trust in the decision-making process, facilitating the adop-
tion and effective implementation of system recommendations in real-world
scenarios (e.g., Mehdiyev and Fettke 2020; Notz 2020; Suvarna etal. 2022). Vis-
ualization is pivotal as a design feature in PAS, guiding users visually through
the decision process. Visualized results prove instrumental in enabling users to
swiftly grasp decision outcomes and potential consequences (e.g., Appelbaum
etal. 2017). The workflow interface serves as a guiding element, allowing users
to navigate the decision process with the flexibility to adjust input parameters,
underlying models, or output validation. These adjustments can be facilitated
through no-code or traditional programming interfaces (e.g., Frazzetto et al.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
2019). Furthermore, extensibility options are paramount in PAS, allowing users
to install or develop components tailored to specific use cases (e.g., Frazzetto
etal. 2019).
4.2 System archetypes
Per our study’s objectives, this section delves into PAS from a meta-level perspec-
tive. The overarching goal is the enhancement of organizational decision-making.
Building on the decision phases in the background section, we use these as the
foundation to conceptualize archetypes, denoting overarching designs or setups
with distinct characteristics. Through this lens, we aim to illuminate the synergy
among the technical and the social subsystems, the prescriptive agent and human
decision-maker, respectively, underscoring their collective role in refining organi-
zational decision-making processes.
Given the extensive body of literature, configurations of the constituent com-
ponents are diverse, depending on the industry, application, or specific use case.
While many authors primarily emphasize technical aspects, the analysis of this lit-
erature reveals recurring patterns in the overall structure of PAS. The patterns are
often anchored in the general decision-making phases (evaluation of alternatives,
decision-making, and adaptation). One noteworthy observation is that technical sub-
systems are not uniform in their role within the decision-making process, nor their
interaction with human decision-makers, and within the reviewed literature, a dis-
cernible shift emerges. Prescriptive analytics is evolving from a passive tool used
by human decision-makers to having agency and assuming responsibilities of their
own in the decision-making process. Prescriptive agents exhibit a growing decision-
making latitude, and they can assume the role of substitutes for behavior-based or
outcome-based decision-making by prescribing, autonomously executing actions,
and adapting to changes in the decision environment (Baird and Maruping 2021).
To conceptualize these findings, we draw upon the theoretical framework of IS
delegation proposed by Baird and Maruping (2021), anchored in agent interaction
theories. Specifically, we adopt delegation mechanisms to delineate and explain four
distinct system archetypes: advisory, executive, adaptive, and self-governing PAS.
Our focus is directed toward understanding the (i) levels of delegation and the (2)
roles or responsibilities played by human decision-makers and prescriptive agents
across the three decision-making phases. Table4 summarizes the key characteristics
of each archetype, and a complete overview of the identified archetypes in our litera-
ture sample is available in Appendix E. Additionally, Table5 demonstrates the roles
and authority of the prescriptive agent and human decision-maker in each archetype.
The agency of the other is not always entirely removed, and it rather pertains to the
primary responsibility of a delegator or proxy-based relationship (Baird and Marup-
ing 2021). We detail this in the following sections, where we will analyze the arche-
types identified through our study, supported by exemplary visualization (refer to
Figs.7, 8, 9, and 10) of the responsibilities and delegation mechanisms in the PAS-
based decision-making process.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
4.2.1 Advisory PAS
The advisory archetype is notably the most common variant in the literature sam-
ple by a significant margin. In this archetype, prescriptive agents contribute only
to the initial phase by assessing alternatives and presenting the optimal decision or
course of action to the user, who maintains full decision-making authority. These
prescriptive agents are static and do not adapt to the consequences of a decision
or changing environments, necessitating manual adjustments or reconfigurations of
inputs or underlying models by humans. The prescriptive agents are, in this sense,
mostly passive tools to be used by the decision-maker with minimal delegation or
agency in the decision-making process and entirely disconnected from the problem
environment. For instance, Abdollahnejadbarough etal. (2020) explore a telecom-
munications provider employing an advisory PAS for supplier management. The
system collects data from internal ERP and external supplier sources before employ-
ing machine learning to cluster suppliers. Subsequently, an optimization engine pro-
cesses the results to recommend the most efficient suppliers for sourcing decisions.
The decision-maker handles the following steps, such as contacting suppliers or exe-
cuting purchase orders, without further involvement of the prescriptive agent.
Though less extensively researched, there are some examples in the litera-
ture where delegation does happen between the decision-maker and the prescrip-
tive agent during the evaluation of alternatives phase (cf. Figure7). This interac-
tion might involve adjusting inputs or decision variables to accommodate real-world
Table 4 Summary of key characteristics of PAS archetypes
Archetypes Key characteristics
Advisory PAS • Prescriptive agents support the first phase of the decision-making process by pre-
scribing an output to the decision-maker as a primarily passive tool to be used
• The decision-maker takes this output and manually takes action in the decision
environment
• The prescriptive agents are disconnected from the decision environment and do not
adapt to changes, which humans must perform manually
Executive PAS • The prescriptive agent evaluates the alternatives and takes action in the decision
environment
• The prescriptive agent remains static and does not adapt to the effects of the out-
comes on the environment
• The human decision-maker remains responsible for observing the effects and adapt-
ing
Adaptive PAS • The prescriptive agent prescribes an output to the decision-maker, who executes the
decision in the environment
• By observing the decision-maker or incorporating human feedback, the prescriptive
agents adapt to the consequences and effects of the prescribed output in the environ-
ment
Self-governing
PAS
• The archetype combines the capabilities of the three former archetypes
• The prescriptive agent has agency and responsibilities in the whole decision-making
process, mainly with some involvement from the human decision-maker but with the
potential for full automation
• There is a high potential for delegation between prescriptive agents and human
decision-makers in the phases
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
Table 5 Overview of PAS archetypes and the respective primary authority of human decision-makers and prescriptive agents in the decision-making phases
PAS archetypes Evaluation of alternatives Decision-making Adaptation and learning Frequency
in literature
sample
Example articles
Advisory Prescriptive agent Human decision-maker Human decision-maker 197 [29], [66], [122], [179], [244], [36], [64], [260]
Executive Prescriptive agent Prescriptive agent Human decision-maker 12 [23], [98], [128], [163], [223], [199], [251]
Adaptive Prescriptive agent Human decision-maker Prescriptive agent 21 [11], [198], [218], [220], [227], ([32], [79], [92], [162],
[234]
Self-governing Prescriptive agent prescriptive agent Prescriptive agent 13 [16], [24], [51], [67], [159], [21], [60], [77], [196], [210]
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
factors, expert knowledge, or risk preferences by the decision-maker. For example,
Kawas etal. (2013) outline a PAS for sales team assignments that recommends opti-
mal allocations while allowing decision-makers to fine-tune output through what-if
analyses. This approach incorporates expert judgment (i.e., expert-in-the-loop), such
as customer sentiment or subjective preferences, enabling experimentation with
diverse sales team configurations.
4.2.2 Executive PAS
Executive PAS is the least common archetype in our sample. With only twelve
papers, this archetype represents a niche in current PAS research. Humans tra-
ditionally hold the mandate to act upon a prescriptive output. However, the lit-
erature also suggests some PAS designs in which the prescriptive agent receives
the authority to execute decisions autonomously in the problem environment. In
these systems, adaptation remains static, and the prescriptive agent is responsi-
ble for the initial two phases, with minimal interaction with the decision-maker.
Prescriptive
Agent
Human
Decision-maker
Delegation MechanismsDecision Phases
I
Evaluation of
Alternatives
II
Decision-making
III
Adaptation
and learning
Provide optimal decision or course of action
Adjust decision inputs or variables
Execute
decision
Generate
prescription
Adapt to
consequences
Fig. 7 Exemplary visualization of delegation mechanisms and responsibilities in advisory PAS
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
Executive PAS are predominantly utilized in domains with high automation,
standardization, or repetitiveness, where rapid decision-making is necessary.
For example, Soroush etal. (2020) introduce a PAS recommending optimal
safety actions to detect and address process operation hazards by implement-
ing mitigative chemical-process measures. Similarly, Mazon-Olivo etal. (2018)
describe a PAS in precision agriculture that autonomously sends repetitive and
planned actions to IoT devices in the field. Additional examples include intel-
ligent call center routings where an optimal service employee is matched to a
customer (Ali 2011) and a data allocation scheme across a Hadoop cluster for
enhanced data security and privacy (Revathy and Mukesh 2020). In some cases
(cf. Figure 8), there is a higher degree of delegation, where the prescriptive
agent executes the decision in the environment, but a decision-maker must first
approve or validate the output (Rizzo etal. 2020). This approach can benefit
high-stakes decision-making with significant financial implications or safety and
compliance concerns.
Prescriptive
Agent
Human
Decision-maker
Delegation MechanismsDecision Phases
I
Evaluation of
Alternatives
II
Decision-making
III
Adaptation
and learning
Generate
prescription
Adapt to
consequences
Ask for approval or validation for execution
Accept, adjust, or decline action
Execute
decision
Fig. 8 Exemplary visualization of delegation mechanisms and responsibilities in executive PAS
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
4.2.3 Adaptive PAS
In the case of an Adaptive PAS, while human decision-makers maintain authority
over the decision-making phase, the prescriptive agent assists in the adaptation and
learning phases, contributing to a more effective and well-informed decision-mak-
ing process. The prescriptive agent monitors decision outcomes and their impact on
the decision environment, incorporating observations into subsequent iterations by
adding new data as input or dynamically adjusting the decision model. As problem
environments often change due to shifting requirements, priorities, or new knowl-
edge, adaptive PAS, as a dynamic archetype, holds significant potential compared to
static counterparts.
For example, Liu etal. (2019) propose a system for optimizing locomotive
wheel maintenance operations, recommending inspection schedules to mini-
mize long-term cost rates. Similarly, Zhang etal. (2021) present a reinforcement
learning-based maintenance optimization model that determines optimal actions
based on a machine’s ever-changing degradation state. Bousdekis etal. (2020)
emphasize the importance of feedback and learning mechanisms in a generic
Prescriptive
Agent
Human
Decision-maker
Delegation MechanismsDecision Phases
I
Evaluation of
Alternatives
II
Decision-making
III
Adaptation
and learning
Provide optimal decision or course of action
Adjust decision inputs or variables
Execute
decision
Generate
prescription
Adapt to
consequences
Monitor outcomes of executed action in
decision environment
Provide information about outcomes in
decision environment
Fig. 9 Exemplary visualization of delegation mechanisms and responsibilities in adaptive PAS
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
IoT scenario, where an agent suggests optimal actions to users and updates the
prescriptive model dynamically based on real-time IoT sensor data. Prescriptive
Agents can also observe the decision environment and decision-makers while
actioning. Tamimi et al. (2019) discuss a PAS for field development design,
recommending optimal designs and deriving the decision-maker’s utility func-
tion for subsequent iterations of prediction and optimization models. Käki etal.
(2019) highlight the deviation from model recommendations in production plan-
ning, often resulting in deteriorated performance, and emphasize the added value
of the adaptive PAS compared to static counterparts.
There are also examples in the literature where decision-makers observe action
consequences or judge potential outcomes based on domain knowledge (i.e.,
expert-in-the-loop systems). Here, the prescriptive agent requires active input
from the human to adapt for subsequent iterations, with decision-makers provid-
ing information by, for example, relabeling outputs or aggregating real-world
outcomes into the training set for future cycles (e.g., Krumeich etal. 2016; Kim
etal. 2020; Vater etal. 2020; Miikkulainen etal. 2021). This archetype indicates
that expert knowledge and human judgment remain vital in the adaptation and
Prescriptive
Agent
Human
Decision-maker
Delegation MechanismsDecision Phases
I
Evaluation of
Alternatives
II
Decision-making
III
Adaptation
and learning
Provide optimal decision or course of action
Adjust decision inputs or variables
Generate
prescription
Adapt to
consequences
Monitor outcomes of executed action in
decision environment
Provide information about outcomes in
decision environment
Execute
Decision
Ask for approval or validation of execution
Accept, adjust, or decline action
Fig. 10 Exemplary visualization of delegation mechanisms and responsibilities in self-governing PAS
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
learning phase. Also, from the perspective of GST, adaptation is crucial, as the
social and technical subsystems naturally change over time (Chatterjee et al.
2021).
4.2.4 Self‑governing PAS
The fourth archetype, self-governing PAS, represents a potentially fully autonomous
system where the prescriptive agent has agency and responsibilities in the entire
decision process independently or with the decision-maker’s involvement. Combin-
ing the capabilities of the other three archetypes, the self-governing PAS is the most
sophisticated version, offering the highest added business value due to automated
execution, adaptability, dynamic self-learning mechanisms, reduced manual work,
and enabling rapid, fact-based decision-making in dynamic environments.
Self-governing PAS relates to well-researched and practiced areas such as route
optimization and data load distribution (cf., Wang etal. 2008; Jozefowiez et al.
2008), used in highly structured environments, which could be considered precur-
sors or early manifestations of the archetype. However, they differ from more recent
examples in aspects like their integration into the broader infrastructural landscape
and their use of historical and real-time data. Moreover, being less disjointed from
the decision environment, they support the entire decision-making process while
showing a high potential for delegation between the human decision-maker and the
prescriptive agent.
Examples from our literature sample primarily come from domains with high
technological maturity and sophistication, such as cyber-physical systems, IoT,
modern energy distribution systems, or smart-sensor-driven environments. These
technologies are inherently data-driven, automated, and integrated—ideal precondi-
tions for advanced agents. For instance, Ceselli etal. (2019) propose a data-driven
framework for optimally distributing data traffic from mobile access points across
capacity-constrained mobile edge cloud networks. A fully autonomous orchestrator
module executes the best data assignment plans. The selected plans, their effects,
and the access points’ demands are continuously logged and validated for subse-
quent iterations. Similarly, Vater etal. (2020) introduce an IoT-based architecture for
real-time error detection in automotive manufacturing, utilizing edge-/cloud-archi-
tecture including modules for preprocessing, prediction, prescription, action-taking,
and validation to close the loop for a fully autonomous decision-making process.
Gutierrez-Franco et al. (2021) present a PAS for last-mile delivery operations
as another example. The system leverages historical data such as traffic, customer
behavior, and driver performance as input. This data is initially preprocessed and
descriptively analyzed, forming a foundation for predicting future operations and
prescribing optimal routes or schedules. The generated output is fed into an execu-
tion module, providing optimal routes for drivers. Furthermore, real-time circum-
stances, including traffic or route deviations by drivers, are captured via GPS or
the vehicle’s sensors, allowing continuous recalculation of the optimal schedule.
In this instance, the decision process is not linear but dynamically adapts based on
the current state of the problem environment. At the end of each shift, a learning
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
mechanism initiates, collecting accumulated data and best practices to enhance
delivery operations, serving as historical data input for subsequent days.
4.3 Technology affordances
As outlined in the scope of our SLR, our goal is to uncover affordances that repre-
sent specific purposes or decision-making tasks supported by PAS. The resulting
concept matrix begins with industry (bold text) and is divided into specific affor-
dances (italic text). We use affordances in the sense of action potential. The techni-
cal nature of our literature sample poses a challenge in extracting both the percep-
tion and actualization of affordance. In this context, we can only derive how the
social subsystem perceives the affordance, for example, through visualization. Con-
versely, actualization can only be detailed based on the actor responsible for execut-
ing the decision, either the technical component or the human decision-maker (Pozzi
etal. 2014; Leidner etal. 2018) (cf. Figure11 for details), which we detail in the
previous section on system archetypes.
Further, our literature sample includes papers that do not address a specific affor-
dance but provide a more general perspective, such as mathematical or algorithmic
formulations, infrastructural considerations, reviews, and conceptual papers. There-
fore, they are excluded from our considerations in this section. In the following, we
will focus on more prevalent industries (N > 5) and affordances, referring to Table6
for niche examples. We conclude this section with a summary and overarching affor-
dance patterns. Additionally, Appendix D provides a comprehensive overview of all
affordances mapped to the literature sample. Further, Appendix H includes a trend
analysis across the more prominent industries.
4.3.1 Manufacturing
As the most researched industry, manufacturing has historically relied on mathemat-
ical models to optimize processes, workforce allocation, and schedules, which are
Decision environment
Prescriptive analytics system
Affordance
Effect
Prescriptive
agent
Human
decision-maker
Affording
relationship
through
interaction
Social subsystem
Technical subsystem
Affordance
Perception
Affordance
Actualization
The decision
affordance is perceived
after the evaluation of
the alterantives
The action (decision)
taken by either the
decision maker or
prescriptive agent
Outcome of decision
and potential for
adaptation and learning
Focus of this
literature review
Fig. 11 Affordance theoretical framework mapped to PAS and the decision-making phases (own depic-
tion based on Pozzi etal. 2014)
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
Table 6 Concept matrix of PAS-driven decision-making affordances
Industry; frequency Affordance (effect); frequency Example studies (IDs)
Manufacturing; n = 44 Maintenance planning for optimal maintenance schedule; n = 24 [11], [54], [137], [196]
Production planning for optimal manufacturing schedule; n = 13 [43], [61], [123], [226]
Product (portfolio) design optimization; n = 4 [66], [202], [213], [216]
Operations safety improvement and planning; n = 1 [128]
Industrial worker training optimization; n = 1 [91]
Deformation control in additive manufacturing; n = 1 [174]
Transportation and logistics; n = 25 Optimization of routing and scheduling; n = 15 [19], [67], [131], [237]
Capacity/cargo management and improvement; n = 8 [53], [88], [173], [184]
Vehicle maintenance planning; n = 2 [10], [200]
Health and MedTech; n = 26 Patient treatment planning and improvement; n = 8 [39], [99], [142], [172]
Patient scheduling; n = 8 [8], [130], [146], [197]
Pandemic/epidemic intervention planning; n = 5 [6], [83], [92], [122]
Human health tracking and improvement; n = 3 [1], [71], [168]
Assortment and inventory planning; n = 1 [158]
Clinical investment management; n = 1 [15]
Energy and environment; n = 24 Optimizing power system/grid operations; n = 5 [7], [25], [31], [262]
Disaster preparation/recovery planning; n = 3 [117], [138], [214]
Electricity brokerage optimization; n = 3 [24], [205], [206]
Maintenance planning; n = 3 [27], [45], [201]
Wastewater treatment improvement; n = 2 [18], [253]
Waste collection and management planning; n = 2 [107], [114]
Optimization of deepwater casing exits; n = 2 [186], [195]
Waterflooding process optimization; n = 1 [103]
Battery lifetime optimization; n = 1 [218]
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
Table 6 (continued)
Industry; frequency Affordance (effect); frequency Example studies (IDs)
Reservoir design planning; n = 1 [32]
Soil slope analysis; n = 1 [119]
Retail and trade; n = 12 Price optimization; n = 4 [46], [76], [135], [209]
Assortment and inventory planning; n = 3 [94], [110], [152]
Sales team assignments; n = 2 [36], [161]
Customer characterization; n = 1 [52]
Customer service recommendation; n = 1 [96]
Theft surveillance and automated checkout; n = 1 [51]
Education; n = 10 Academic performance improving; n = 6 [34], [50], [109], [245]
Dropout prevention planning; n = 2 [113], [212]
Admissions planning and selection; n = 2 [59], [252]
Chemicals and resources; n = 8 Maximize oil/gas recovery; n = 4 [85], [143], [156], [163]
Mining fleet scheduling; n = 1 [33]
Sand molding process improvement; n = 1 [63]
Laboratory task allocation and planning; n = 1 [35]
Biodiesel properties optimization; n = 1 [147]
Technology and communications; n = 6 Social media usage optimization; n = 2 [234], [238]
Network and computing resource orchestration; n = 2 [136], [169]
Software development estimation; n = 1 [235]
Website performance analysis and optimization; n = 1 [81]
Academia; n = 6 Research advising; n = 6 [170], [179], [180], [219]
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
Table 6 (continued)
Industry; frequency Affordance (effect); frequency Example studies (IDs)
Agriculture; n = 5 Harvesting operations planning and optimization, n = 3 [90], [105], [223]
Crop yield optimization; n = 1 [256]
Fish trawling routing and optimization; n = 1 [221]
Crime and law enforcement, n = 3 Law enforcement resource allocation and planning; n = 2 [111], [187]
Imprisonment decision planning and recommendation; n = 1 [248]
Sports and recreation; n = 3 Tournament lodging planning; n = 1 [164]
Sports event safety management and planning; n = 1 [224]
Athlete training process improvement; n = 1 [124]
Defense and military; n = 2 Infantry engagement planning; n = 1 [118]
Military logistics planning; n = 1 [261]
E-Commerce; n = 2 Markdown planning and price optimization; n = 1 [28]
Oder delivery scheduling; n = 1 [14]
Finance; n = 2 Teller machine replenishment planning and allocation; n = 1 [217]
Stock purchase recommendations; n = 1 [241]
Media; n = 2 Radio advertising scheduling; n = 1 [126]
Movie planning and profit-maximizing; n = 1 [78]
Professional ser vices; n = 2 Project staffing planning and allocation; n = 2 [167], [228]
Public sector; n = 2 Child welfare assessment; n = 1 [153]
Infrastructure planning and optimization; n = 1 [178]
Tourism & hospitalitiy; n = 2 Optimal tour pricing; n = 1 [127]
Restaurant recommendations; n = 1 [57]
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
Table 6 (continued)
Industry; frequency Affordance (effect); frequency Example studies (IDs)
Industry-agnostic; n = 20 Prescriptive process management; n = 8 [40], [95], [192], [203]
Employee recruiting and staffing; n = 3 [13], [80], [160]
Procurement and supplier management; n = 2 [177], [258]
Marketing management; n = 2 [70], [125]
Facility/asset management; n = 2 [116], [129]
Managerial Accounting; n = 1 [104]
Call center routing; n = 1 [112]
Server incident management and prevention; n = 1 [132]
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
crucial in enhancing efficiency and productivity. With Industry 4.0, manufacturing
has transformed into a highly developed sector integrating advanced technologies
like robotics and IIoT to establish intelligent, interconnected production systems
(Wanner etal. 2023). As reflected in our literature sample, these developments have
spurred extensive research interest in data-driven analytics. Maintenance planning,
a dominant affordance for PAS, has emerged as a research stream called prescriptive
maintenance. Here, the primary purpose is to afford optimal maintenance schedules,
often incorporating spare parts management, primarily driven by machine sensor
data. For example, these systems employ descriptive analytics to analyze the current
machine state and predictive analytics to foresee potential failures, fueling a pre-
scriptive model to propose the optimal schedule (e.g., Liu etal. 2019; Ansari etal.
2019; Fox etal. 2022). Similarly, production planning, a longstanding research area,
has begun to harness sensor-driven data to improve production schedules, processes,
quality, and operations. Examples include PAS to optimize shop floor operations
(Stein etal. 2018) and schedule diffusion furnaces (Vimala Rani and Mathirajan
2021).
In addition to the two dominant affordances, there are other, less explored exam-
ples within the manufacturing sector. Some of these include supporting product
development, optimizing product portfolio designs (Jank etal. 2019), and enhancing
the design of industrial products (Dey etal. 2019). Moreover, research has proposed
utilizing PAS for training industrial workers by offering training schedules based
on digital twins (Longo et al. 2023), prescribing optimal safety actions (Soroush
etal. 2020), and improving additive manufacturing processes through deformation
control (Jin etal. 2016). As the manufacturing industry continues to advance, the
potential applications of PAS are expected to grow, fostering further innovation and
efficiency.
4.3.2 Transportation andlogistics
Facing significant pressure regarding cost efficiency and sustainability, the transpor-
tation and logistics industry, like manufacturing, has a history of using mathemati-
cal optimization models for routing and scheduling tasks (Konstantakopoulos etal.
2022). With vehicles becoming increasingly connected and generating digital traces
through sensors and networks, PAS can further harness this data to afford efficiency,
enabling the industry to capitalize on various applications. Routing and scheduling
naturally emerge as the most researched affordance.
Examples span various modes of transportation, such as ground (Gutierrez-
Franco etal. 2021), air (Ayhan etal. 2018), and public transport (Xylia et al.
2016), showcasing the versatility and potential of PAS in enhancing efficiency
across diverse systems. Another affordance is capacity management. Affordance
effects include optimizing freight or cargo distribution (Rizzo etal. 2020) and
passenger seat assignments (Moore etal. 2021), ensuring efficient resource allo-
cation, and enhancing overall operational performance. Lastly, considering that
vehicles undergo continuous degradation while in use, our sample also includes
prescriptive maintenance affordances to effectively address wear and tear,
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
optimize maintenance schedules, and prolong the service life of vehicles (Consil-
vio etal. 2019; Anglou etal. 2021).
4.3.3 Health andMedTech
Health and MedTech are the second most researched sectors in our literature sam-
ple. Implementing PAS can significantly improve resource allocation and over-
all patient outcomes with the increasing complexity and demand for healthcare
services. The sector features two dominant affordances: patient treatment plan-
ning and scheduling. Patient treatment planning primarily focuses on improv-
ing health outcomes by optimizing treatments, reducing hospital readmissions,
enhancing precision medicine, and boosting clinical staff efficiency (Rider etal.
2021; Zheng etal. 2021). On the other hand, patient scheduling emphasizes the
efficient management of appointment scheduling and bed occupancy (Belciug
and Gorunescu 2016; Srinivas and Ravindran 2018). In the clinical context, niche
examples of affordances include assortment, inventory planning (Galli et al.
2021), and investment management (Fang etal. 2021).
Further, in light of the recent COVID-19 pandemic, several contributions have
focused on pandemic or epidemic intervention planning. These studies consider var-
ious aspects, such as mobility intervention and the rapid deployment of medical staff
and equipment (Miikkulainen etal. 2021; Ahmed et al. 2021). Lastly, a group of
researchers has shifted their focus to the patients or their bodies directly, for exam-
ple, incorporating sensor data for health tracking, enhancing safety and consumption
decisions, and preventing impulsive behavior among patients (Sedighi Maman etal.
2020; Raychaudhuri etal. 2021). By leveraging PAS in these areas, healthcare pro-
viders can offer more personalized care, empower patients to make better-informed
decisions, and ultimately improve overall health outcomes.
4.3.4 Energy andenvironment
The energy and environment industry showcases more diverse affordances in our
literature sample than in previous sectors. A key focus in this domain is power
generation systems (e.g., wind farms), where PAS are utilized to afford optimized
performance (Tektaş etal. 2022), electricity brokerage (Peters etal. 2013), and
prescriptive maintenance (Goyal etal. 2016) for these systems. Additionally, PAS
applications extend to optimizing waste collection and planning (Vargas et al.
2022) and enhancing wastewater treatment processes (Zadorojniy etal. 2019).
Some PAS afford disaster preparation and recovery planning, addressing chal-
lenges posed by wildfires, hurricanes, or floods (Hu etal. 2019; Yang etal. 2022).
Niche examples within this sector include soil slope analysis (Li etal. 2019) and
optimization of battery lifetime (Eider and Berl 2020). As demonstrated for this
sector, implementing AI systems can significantly impact society by improving
sustainability, efficiency, and environmental protection (Schoormann etal. 2023).
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
4.3.5 Retail andtrade
The retail and trade industry encompasses both B2B and B2C interactions. Despite
its size, there has been relatively little PAS research in this area compared to other
industries. One possible reason may be the traditional set-up often found in brick-
and-mortar stores and a comparatively lower level of digitization than in industries
like manufacturing or logistics. Nevertheless, there are instances of PAS research
in our sample. One example is dynamic price optimization, which incorporates fac-
tors such as customers, competition, business partners, and environmental aspects
(Ito and Fujimaki 2017). A key challenge in this industry is assortment and inven-
tory planning, which is a significant cost driver when considering perishable goods
or inventory costs (Jin etal. 2016; Flamand etal. 2018). In the B2B context, sales
teams drive revenue, making optimal assignment a potential affordance. Other
examples include customer characterization (Perugini and Perugini 2014), customer
service recommendations (Lo and Pachamanova 2015), theft surveillance, and facil-
itating automated store checkouts (Hauser etal. 2021).
4.3.6 Education
Three primary affordances have emerged in education: dropout prevention plan-
ning, improving students’ academic performance, and admissions planning. Firstly,
dropout prevention planning focuses on identifying students at risk of leaving their
educational programs prematurely. PAS enables institutions to target support and
interventions, ensuring that students receive the help they need to stay on track
and complete their studies (Yanta etal. 2021; de Jesus and Ledda 2021). Secondly,
improving academic performance is an essential priority for educational institutions.
By utilizing PAS, educators can gain insights into students’ learning patterns and
areas of difficulty, enabling them to tailor teaching approaches and offer personal-
ized learning experiences that foster success (Uskov etal. 2019; Islam etal. 2021).
Finally, admissions planning is essential to maintaining a thriving educational insti-
tution. PAS can help optimize the admissions process by analyzing student demo-
graphics and academic performance to ensure that institutions admit the most
suitable candidates (Kiaghadi and Hoseinpour 2023). Applying PAS in education
can improve traditional academic decision-making processes, enhance student out-
comes, and streamline institutional operations.
4.3.7 Chemicals andresources
The chemicals and resources industry has some affordances identified in our review,
aiming to optimize processes, boost efficiency, and promote sustainable resource uti-
lization. Maximizing oil and gas recovery is the most researched affordance effect,
with PAS used to optimize extraction techniques, reservoir modeling, and resource
management. Other examples include laboratory task allocation (Silva and Cortez
2022), mining fleet scheduling (Nakousi etal. 2018), optimizing biodiesel proper-
ties (Suvarna etal. 2022), and improving sand molding processes (Chowdhary and
Khandelwal 2018).
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
4.3.8 Technology andcommunication
Despite its technological maturity, the technology and communications sector has lim-
ited research on PAS applications. This sector can benefit from PAS in various affor-
dances, such as network and computing resource orchestration (Ceselli etal. 2018,
2019), social media optimization (Ballings etal. 2016), software development estima-
tion (Pospieszny 2017), and website performance analysis (Salvio and Palaoag 2019).
PAS can streamline network and computing systems, enhance social media campaigns,
improve project management, and optimize website performance. Although current
literature is limited, this industry has the potential for further exploration. Harnessing
PAS can improve performance, efficiency, and user satisfaction across technology and
communication operations.
4.3.9 Academia
Furthermore, several studies have investigated the use of PAS in academia. They focus
on improving and enhancing academic research performance. By analyzing data related
to research output and other relevant factors, such as citations or related work, PAS can
provide valuable insights and recommendations for researchers, such as journals or ref-
erences. Additionally, some research has explored the concept of system thinking and
crafting scenarios. These approaches help researchers better understand their work’s
potential outcomes and consequences, enabling them to make more informed decisions
about their research directions (e.g., Song etal. 2014; Jeong and Joo 2019).
4.3.10 Industry‑agnostic
Beyond sector-specific applications, much research investigates affordances from an
industry-agnostic perspective. In these cases, authors often apply prescriptive ana-
lytics in a context, which we coin prescriptive process management (Krumeich etal.
2016; Kubrak etal. 2022). The primary affordance effects include process moni-
toring, controlling execution, and recommending the most appropriate subsequent
actions. Interestingly, two contributions specifically discuss the explainability of
decision outputs in this setting (Mehdiyev and Fettke 2020; Notz 2020). Moreover,
PAS can be used to optimize employee recruitment (Pessach etal. 2020), facility and
asset management (Lavy etal. 2014), and supplier selection (Abdollahnejadbarough
et al. 2020). Some niche systems involve enhancing accounting procedures and
improving call center routing (Ali 2011). Overall, these versatile applications dem-
onstrate the potential of PAS to streamline operations and support decision-making
across a wide range of industry contexts.
4.3.11 Summary
In conclusion, PAS have demonstrated the potential to affect various industries posi-
tively, for instance, by optimizing processes, allocating resources, scheduling, and
planning maintenance actions. The systems afford organizations to achieve enhanced
efficiency and productivity. Although the current research varies in scope and depth
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
across different sectors, the widespread applicability of PAS indicates its capacity
to drive innovation and streamline operations across a wide range of contexts. It is
important to note that this section did not detail all affordances, specifically under-
represented industries such as tourism, media, and finance. Please refer to Table6
and the example papers for more information on these sectors and their respective
affordances.
Even though the PAS in our literature sample are diverse, they share the com-
monality of prescribing the best course of action in a specific decision environment
and situation. Given the abovementioned affordances, we derived three overarching
affordance effects of PAS to improve decision-making processes: (1) improvement,
(2) scheduling, and (3) resource allocation, which we detail in the following.
Improvement (1) is centered on enhancing and optimizing the current state of
an object or the decision environment to achieve an improved state. This affordance
involves conducting a comprehensive analysis and adjusting processes, products, or
operations to make them more effective, efficient, and aligned with predefined goals.
For example, product design optimization focuses on continuously refining products
based on user feedback and market trends (e.g., Dey etal. 2019; Jank etal. 2019).
Likewise, within the healthcare sector, patient treatment planning and improvement
entail tailoring treatments to individual patient needs while constantly updating
these plans based on patient responses and emerging medical insights (e.g., Rider
etal. 2021; Zheng et al. 2021). Similarly, maintenance optimization aims to opti-
mize equipment performance and longevity in industrial settings, ultimately reduc-
ing downtime and increasing efficiency (e.g., Liu etal. 2019; Consilvio etal. 2019).
Scheduling (2) entails strategically organizing and coordinating tasks and actions
in the decision environment over time, creating and managing a timeline of activi-
ties that aligns with an organization’s objectives, resource availability, and external
factors. Effective scheduling and planning improve operational flow and resource
utilization, reducing bottlenecks and inefficiencies. For example, in healthcare,
patient scheduling is critical for maximizing medical facilities and staff use (e.g.,
Belciug and Gorunescu 2016; Srinivas and Ravindran 2018), while in the energy
sector, planning energy distribution is crucial for balancing supply with consumer
demand (Goyal etal. 2016).
Resource allocation (3) involves strategically distributing resources like work-
force, materials, and finances to areas most needed and will be most effective. This
process requires a thorough understanding of resources’ availability, potential, limi-
tations, and different organizations’ objectives and needs. For example, in logis-
tics, capacity and cargo management ensures optimal use of transport and storage
resources (Rizzo etal. 2020; Gutierrez-Franco etal. 2021). In urban planning, effi-
cient allocation of resources for waste collection and management is vital for main-
taining cleanliness and public health (Vargas etal. 2022).
The affordances are not mutually exclusive but can overlap depending on the
specific system, application use case, or complex organizational settings. For
instance, in manufacturing, the improvement of a product design (Improvement) is
closely linked to the planning of production schedules (Scheduling) and the allo-
cation of manufacturing resources (Resource Allocation). Similarly, in healthcare,
patient treatment plans (Improvement) need to be integrated with patient scheduling
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
(Scheduling and Planning) and the allocation of medical staff and equipment
(Resource Allocation). Understanding the interplay and overlap of these affordances
is crucial for effective management and decision-making. Organizations can holisti-
cally approach problem-solving and optimization by recognizing how they comple-
ment each other, leading to more comprehensive PAS.
5 Directions forfuture research
Drawing on the synthesis and conceptualization in the previous sections, we discuss
the main observations and deriving aspects that remain open to pave the ground for
future research. We will discuss possible research directions from technical, social,
and overarching perspectives. Table7 highlights the research agenda, key observa-
tions, and illustrative research directions or questions.
5.1 Technical perspectives
The preponderance of research within the technical subsystem is understandable, as
PAS are fundamentally rooted in technology, and a significant portion of investiga-
tions in this area stem from disciplines closely tied to technological advancements.
Our SLR has uncovered numerous PAS components that various authors in our sam-
ple have extensively researched and well-addressed, establishing a solid foundation
in the field. Despite this wealth of knowledge, we have identified specific gaps that
warrant further investigation.
For instance, action mechanisms have been relatively underexplored in the exist-
ing literature. We contend that these components are crucial for designing an effec-
tive PAS, as they drive the interaction between the prescriptive agent and the human
decision-maker. Further, ancillary features with a focus on seamlessly integrating
technology components within the social structure and the broader organizational
landscape are lacking in current research.
In addition to the identified gaps, mathematical programming and, to a lesser
extent, bio-inspired optimization algorithms have been well-established in pre-
scriptive analytics. Recently, there has been a surge of interest in incorporating ML
techniques to integrate predictions or probabilities as precursors to subsequent opti-
mization processes. ML, particularly deep learning, is widely researched for manag-
ing extensive and high-dimensional datasets (Janiesch etal. 2021). However, chal-
lenges surrounding interpretability and explainability have hindered its adoption
among decision-makers. While explainable and interpretable ML offer promising
solutions (Zschech etal. 2022; Herm etal. 2022; Wanner etal. 2022), integrating
these approaches into a PAS remains mainly open. Today, ML predominantly con-
tributes to descriptive and predictive analytics, but increased transparency and trust,
which are vital, especially for high-stakes decision-making, remain open. However,
interpretable ML would be a valuable extension to existing prescriptive analytics
approaches (Shollo etal. 2022).
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
Table 7 Research agenda with six key observations
Perspectives Key observations Illustrative research directions or questions
Technological Perspectives (1) Action mechanisms in prescriptive analytics are underresearched • Research on the design requirements of action, tracking, and adapta-
tion mechanisms in PAS
• Research on the design requirements of workflow interface for
human-AI interaction
(2) Interpretable ML and RL are possible as extensions to optimiza-
tion models
• Interpretable ML as an extension to PAS and its unique require-
ments
• RL and its unique characteristics, concepts, and requirements within
the PAS framework
Social perspectives (3) The human perspective is underrepresented in PAS-based
decision-making
• Research on how to increase the trust of the decision-maker regard-
ing the decision output of the technology components
• Do explainability, accountability, fairness, and bias require PAS-
specific research, or is it generalizable through AI/ML research?
(4) Links to and the embedment in the broader organizational land-
scape or structure are missing
• How can a PAS effectively contribute to driving fact-based decision-
making in organizations?
• How can a PAS be embedded across different organizational con-
texts and structures?
General Perspectives (5) Executive, adaptive, and self-governing archetypes are under-
explored, as well as the human-agent interface that facilitates the
interaction
• Research on case-agnostic blueprints with corresponding design
principles and options for the archetypes
• What interface forms can facilitate and improve the interaction
between humans and agents in a PAS?
• How can a PAS perceive and actualize affordances?
(6) The review identified diverse affordances in the sense of action
or action possibilities, but research along more nuanced affordance
theories could help design more effective PAS
• Affordance-actualization: investigate how decision outputs from
a PAS are implemented within organizations and the effects on
decision-making processes
• Affordance-network approach: examine how organizations achieve
more significant outcomes by connecting a series of more immedi-
ate, concrete decision outcomes
• Affordance-trajectories: the trajectory along which decision outputs
from PAS travel in organizations
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
Further, RL has also garnered attention for its potential in PAS owing to its
dynamic, adaptive, and iterative nature and its aptitude for addressing well-formal-
ized decision-making problems (Greene etal. 2022). Researchers have already dem-
onstrated the effectiveness of RL in the context of adaptive and self-governing PAS
archetypes. However, our understanding is limited by the scarcity of literature on
this topic. Consequently, further research is needed to explore RL’s unique charac-
teristics, concepts, and requirements within the PAS framework, ultimately contrib-
uting to a more robust understanding of its potential applications and benefits and an
alternative to more traditional optimization techniques such as linear programming.
Finally, As a recent innovation in AI, foundation models – particularly large lan-
guage models – have catalyzed a significant paradigm shift in the development of
AI systems. This transformation has already profoundly impacted existing IT ser-
vices and ecosystems while simultaneously enabling the creation of novel applica-
tions (Feuerriegel etal. 2024; Schneider etal. 2024). Through our review, we have
observed that the application of foundation models in prescriptive analytics remains
unexplored. However, we posit that this domain holds substantial potential. Future
research should thus investigate leveraging the advanced capabilities of founda-
tionmodels to enhance intelligent decision-making systems.
5.2 Social perspectives
As previously mentioned, the social subsystem (including the human decision-
maker) has been relatively underresearched, which is somewhat understandable con-
sidering the technological origins of prescriptive analytics. However, we argue that
increased attention to this perspective is essential for a more comprehensive under-
standing of the field, especially for the IS community.
Current research in this area often takes a case-specific approach, with insuffi-
cient consideration of the broader organizational landscape and how systems inte-
grate into the larger picture. The outputs generated by these decision-making pro-
cesses can have varying contextual implications, depending on factors such as the
business unit or hierarchical level, ultimately influencing strategic and operational
decision-making (e.g., Appelbaum etal. 2017). Consequently, we posit that a PAS
will be most valuable if it is organization-wide, encompassing all decision pro-
cesses and business functions, avoiding siloed structures, and made available (e.g.,
as-a-service) to all decision-makers tailored to their unique environments. From a
technological standpoint, achieving this vision requires infrastructural features such
as standardized data integration, interoperability, distributed computing, and effec-
tive API design (Lepenioti etal. 2020; Vieira etal. 2020; Verbraeken etal. 2021).
However, further research is needed to identify the specific requirements unique to
a PAS. Additionally, given that analytics is not solely a technology-driven concept
and necessitates a cultural shift towards evidence-based decision-making, it is worth
exploring whether a PAS should be a mere result of this shift or function as a driver
or initiative to move organizations toward more fact-based decision processes.
Beyond the organizational setting, the human decision-maker’s perspective war-
rants further research. Specifically, ML for predictive analytics has been extensively
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
studied regarding explainability, interpretability, accountability, fairness, and bias
(e.g., Meske etal. 2022; Nadeem etal. 2022; Kraus etal. 2023). Due to the inher-
ent differences between various analytics methods and outputs, particularly those
focused on prescriptive analytics, we argue that additional research is needed, spe-
cifically in the context of PAS, such as the decision-maker’s trust in decision outputs
(Caro and de Tejada Cuenca 2023).
This line of inquiry will facilitate a deeper understanding of the challenges and
opportunities associated with PAS. It will also foster their responsible development
and deployment within organizations, ensuring that they align with ethical standards
and contribute positively to decision-making processes.
5.3 General perspectives
Our synthesis identified four archetypes within PAS. Despite this progress, most
PAS in our sample are predominantly advisory, while the executive, adaptive, and
self-governing archetypes remain underexplored. This imbalance suggests a sig-
nificant disconnect between problem environments and PAS, as these systems are
often disjointed from the last two phases of the decision-making process. In real-
world scenarios, many environments are constantly in flux due to natural changes
or actions. This dynamic nature is frequently overlooked in current PAS research.
However, action execution, adaptation, and learning mechanisms hold great poten-
tial, as they can help reduce information loss across iterations and improve decision-
making processes over time while minimizing reliance on subjective or judgmental
human experiences (Sturm etal. 2021). The BA community must address this gap
by understanding the requirements of such systems and developing case-agnostic
blueprints with corresponding design principles and options.
Furthermore, delegation mechanisms warrant increased attention, represent-
ing the initial steps toward hybrid intelligence systems and the symbiosis between
agents and humans in decision-making (Dellermann etal. 2019; Peng etal. 2022).
By focusing on these underexplored areas, researchers can contribute to a more
comprehensive understanding of PAS, ultimately fostering the development of sys-
tems that effectively integrate advanced technology and human expertise in organi-
zational decision-making processes.
In our review, we utilized the affordance theory to examine how individuals
or organizations with specific objectives can leverage technology at a basic level.
Given the nature of the papers in our sample, we focused on affordance effects.
We argue that understanding the balance between technology’s enabling and con-
straining aspects is crucial for designing effective PAS for organizational decision-
making. Affordance theory offers a promising perspective for future research in
analytics-driven organizational decision-making. One area to explore is the process
of affordance-actualization, which involves understanding how the potential benefits
of technology are actualized in the form of organizational outcomes (Strong etal.
2014). In the context of our work, this would involve investigating how decision out-
puts from a PAS are implemented within organizations and the effects on decision-
making processes themselves.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
Another direction is applying the affordance-network approach, which examines
how organizations achieve more significant outcomes by connecting a series of more
immediate, concrete decision outcomes within a network of interrelated affordances
(Burton-Jones and Volkoff 2017). Furthermore, exploring the trajectory along which
affordances travel can provide valuable insights into the processes and conditions
that shape the perception and actualization of affordances in organizational deci-
sion-making (Thapa and Sein 2018). This line of inquiry can help elucidate how the
potential benefits of a PAS are transformed into tangible outcomes in practice. By
considering these aspects of affordance theory, researchers and practitioners could
develop more effective and comprehensive PAS for decision-making in organiza-
tional settings.
6 Concluding remarks
Our research contributes to the knowledge of BA, specifically PAS, as the most
sophisticated maturity level to support organizational decision-making with the
highest potential business value. In this context, we conducted an SLR on the state
of PAS research, emphasizing the IS artifact and GST as a theoretical framework.
We reviewed 262 relevant contributions, enabling a holistic view of the field and its
relevance to the broader BA community in research and practice, with three main
contributions.
Our first contribution is the development of a concept matrix comprising 23
distinct constituent components, revealing fundamental technical design elements.
Based on GST and considering important ancillary features, this updated under-
standing gives researchers a starting point to study the relationships between ele-
ments and their effective integration into organizational decision processes.
Second, we analyzed the meta-level of PAS, revealing the differing roles of pre-
scriptive agents and human decision-makers from a decision-theoretic perspective
and highlighting their synergy, delegation, and adaptability. Our conceptualization
led us to derive four archetypes, each providing varying levels of support in the deci-
sion-making process. From a practical perspective, our findings can serve as initial
blueprints or guiding principles on what system modules and features to consider
when designing a PAS for specific purposes.
Third, we identified various technology affordance effects of PAS across various
sectors, unveiling the purpose and benefits of employing such systems. Owing to the
extensive chronology of leveraging optimization algorithms within industries like
manufacturing and transportation, a data-driven approach to prescriptive analytics
has experienced considerable exploration in these sectors. Meanwhile, although spe-
cific fields exhibit a reduced degree of investigation, a comprehensive assessment
revealed a heterogeneous array of research endeavors spanning multiple industries.
Finally, our paper reveals six key findings or observations, enabling us to derive var-
ious research directions and implications for the BA and IS community.
As with most research endeavors, our study comes with certain limitations. Our
goal was to establish the status quo and create a shared understanding of the exist-
ing body of knowledge on PAS. To achieve this, we relied on concept matrices as
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
a qualitative analysis tool, which is never complete and serves as an initial founda-
tion for more comprehensive research and contextualization. Furthermore, we focus
our SLR on prescriptive analytics, resulting in an underrepresentation of the socio-
technical lens in our literature sample due to the current research’s technological and
algorithmic focus. However, even though the papers did not explicitly mention the
social subsystem or interactions, we could make educated assumptions based on,
for example, architectural overviews presented, which enabled us to derive system
archetypes, such as how different components synergize with the decision-maker.
Additionally, it is imperative to highlight that our primary objective was to
demystify the PAS landscape in research from an overarching IS perspective.
We endeavored to conceptualize and categorize the aspects within this domain,
acknowledging the inherent challenges in attributing specific concepts to distinct
groups, especially for prescriptive analytics techniques. While we recognize that the
clarity of separation between these elements is not always absolute, owing to the
multifaceted nature and numerous influencing factors, we believe our work provides
a substantial foundation. This framework is particularly beneficial for newcomers in
the field, whether designing systems or conducting research. It offers a solid ground-
ing, enabling a deeper and more nuanced understanding of the PAS space. We assert
that this is essential for anyone aspiring to navigate, contribute to, or innovate within
this ever-evolving and complex field. However, additional relevant PAS perspec-
tives may still be explored, for example, within a deeper algorithmic review scope.
For instance, a potential area of investigation could be the differentiation between
sequential (predict-then-optimize) and simultaneous (predict-and-optimize) PAS.
Here, the latter could potentially lead to improved decision outcomes, as it proposes
learning a predictive model by directly minimizing the cost of the downstream deci-
sion-making task (Vanderschueren etal. 2022; Zhang et al. 2022). Taking this as
an example, future reviews with an algorithmic focus would be valuable avenues of
inquiry, especially given the recent AI model innovations, which may significantly
impact how future PAS are designed.
Further, we note that "prescription" or "prescriptive analytics" may not be used
as frequently in every research discipline. Some contributions may only refer to an
affordance (e.g., scheduling, routing) or the technique (e.g., optimization, linear
programming) in their title, keyword, or abstract, possibly excluding these with our
search string. Although prescriptive analytics originated in the BA domain (Holsap-
ple etal. 2014; Delen and Zolbanin 2018), our review revealed that it is a well-estab-
lished concept in various research disciplines today, with special issues dedicated
to the topic in widely respected journals (e.g., Giesecke etal. 2022). It is clear that
today, prescriptive analytics is an overarching concept, describing a task objective
or even a decision-making paradigm nested in a socio-technical system instead of
a specific technique or algorithm to be employed, such as linear programming or a
specific ML technique. This trend is also reflected in our sample’s growing number
of publications after 2018, enabling us to establish a representative view of the cur-
rent research state with many contemporary examples from the literature. However,
our literature-centric approach must be assessed for practical relevance by incorpo-
rating real-world PAS applications and engaging with practitioners with hands-on
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
experience in the field. This perspective will ensure a well-rounded understanding of
the topic and its implications and help bridge the gap between theory and practice.
In summary, our research provides a comprehensive understanding of the current
state of PAS and highlights areas for future research and development. By explor-
ing these opportunities, researchers and practitioners can collaborate to create more
effective and efficient PAS, ultimately driving better decision-making and business
value in organizations.
Appendix
Appendix A: Detailed search syntax
Database Search string
Web of science TS = ((prescriptive) AND (model OR machine learning OR optimization OR evolution-
ary OR expert system OR heuristics OR simulation OR artificial intelligence OR
analytics))
Scopus TITLE-ABS-KEY (prescriptive AND (model OR machine AND learning OR optimiza-
tion OR evolutionary OR expert AND systems OR heuristics OR simulation OR
artificial AND intelligence OR analytics))
AISeL abstract:( prescriptive AND (model OR machine learning OR optimization OR evolu-
tionary OR expert system OR heuristics OR simulation OR artificial intelligence OR
analytics)) OR title:( prescriptive AND (model OR machine learning OR optimiza-
tion OR evolutionary OR expert system OR heuristics OR simulation OR artificial
intelligence OR analytics))
ACM DL [Keywords: prescriptive] AND [[Keywords: analytics] OR [Keywords: model] OR
[Keywords: machine learning] OR [Keywords: optimization] OR [Keywords: evolu-
tionary] OR [Keywords: expert system] OR [Keywords: heuristics] OR [Keywords:
simulation] OR [Keywords: artificial intelligence]][Title: prescriptive]
[[Title: analytics] OR [Title: model] OR [Title: machine learning] OR [Title: optimiza-
tion] OR [Title: evolutionary] OR [Title: expert system] OR [Title: heuristics] OR
[Title: simulation] OR [Title: artificial intelligence]]
IEEE explore ((prescriptive) AND (model OR machine learning OR optimization OR evolutionary
OR expert system OR heuristics OR simulation OR artificial intelligence OR analyt-
ics))
Appendix B: Literature sample
ID Year Article title
1 2020 A data analytic framework for physical fatigue management using wearable sensors
2 2022 A deficiency of prescriptive analytics-No perfect predicted value or predicted distribution
exists
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
ID Year Article title
3 2022 A dynamic predict, then optimize preventive maintenance approach using operational
intervention data
4 2019 A Dynamic Prescriptive Maintenance Model Considering System Aging and Degradation
5 2020 A Formative Usability Study to Improve Prescriptive Systems for Bioinformatics Big Data
6 2021 A Framework for Pandemic Prediction Using Big Data Analytics
7 2022 A Fuzzy Prescriptive Analytics Approach to Power Generation Capacity Planning
8 2016 A hybrid genetic algorithm-queuing multi-compartment model for optimizing inpatient bed
occupancy and associated costs
9 2023 A lower approximation based integrated decision analysis framework for a blockchain-based
supply chain
10 2021 A machine learning approach to enable bulk orders of critical spare-parts in the shipping
industry
11 2021 A model-based reinforcement learning approach for maintenance optimization of degrading
systems in a large state space
12 2020 A Modular Edge-/Cloud-Solution for Automated Error Detection of Industrial Hairpin
Weldings using Convolutional Neural Networks
13 2022 A prescriptive analytics approach to employee selection
14 2021 A prescriptive analytics framework for efficient E-commerce order delivery
15 2021 A Prescriptive Analytics Method for Cost Reduction in Clinical Decision Making
16 2022 A prescriptive Dirichlet power allocation policy with deep reinforcement learning
17 2022 A prescriptive framework to support express delivery supply chain expansions in highly
urbanized environments
18 2021 A Prescriptive Intelligent System for an Industrial Wastewater Treatment Process: Analyz-
ing pH as a First Approach
19 2022 A Prescriptive Machine Learning Method for Courier Scheduling on Crowdsourced Deliv-
ery Platforms
20 2017 A Proactive Event-driven Decision Model for Joint Equipment Predictive Maintenance and
Spare Parts Inventory Optimization
21 2017 A procedural approach for realizing prescriptive maintenance planning in manufacturing
industries
22 2019 A prognostic algorithm to prescribe improvement measures on throughput bottlenecks
23 2019 A reference architecture based on edge and cloud computing for smart manufacturing
24 2013 A reinforcement learning approach to autonomous decision-making in smart electricity
markets
25 2022 A Review of Predictive and Prescriptive Offshore Wind Farm Operation and Maintenance
26 2013 A specialty steel bar company uses analytics to determine available-to-promise dates
27 2021 A Two-Stage Data-Driven Spatiotemporal Analysis to Predict Failure Risk of Urban Sewer
Systems Leveraging Machine Learning Algorithms
28 2019 Aprescriptive analyticsapproach to markdown pricing for an e-commerce retailer
29 2021 Aprescriptive analyticsframework for optimal policy deployment using heterogeneous
treatment effects
30 2020 Asurveyon various applications ofprescriptive analytics
31 2013 Adaptive middleware for real-timeprescriptive analyticsin large scale powersystems
32 2019 An artificial intelligence decision support system for unconventional field development
design
33 2018 An asset-management oriented methodology for mine haul-fleet usage scheduling
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
ID Year Article title
34 2021 An Automated Tool to Support an Intelligence Learner Management System Using Learn-
ing Analytics and Machine Learning
35 2022 An Industry 4.0 Intelligent Decision Support System forAnalytical Laboratories
36 2015 An Information System for Sales Team Assignments Utilizing Predictive and Prescriptive
Analytics
37 2018 An Integration of Requirement Forecasting and Customer Segmentation Models towards
Prescriptive Analytics For Electrical Devices Production
38 2020 An Intelligence Learner Management System using Learning Analytics and Machine learn-
ing
39 2022 An Inverse Optimization Approach to Measuring Clinical Pathway Concordance
40 2017 Analysis and optimization based on reusable knowledge base of process performance
models
41 2015 Analysis and optimization in smart manufacturing based on a reusable knowledge base for
process performance models
42 2023 Analytical Problem Solving Based on Causal, Correlational and Deductive Models
43 2022 Analytics with stochastic optimisation: experimental results of demand uncertainty in
process industries
44 2017 Application of derivatives to nonlinear programming forprescriptive analytics
45 2016 Asset health management using predictive and prescriptive analytics for the electric power
grid
46 2023 Believing in Analytics: Managers’ Adherence to Price Recommendations from a DSS
47 2018 Big data on the shop-floor: sensor-based decision-support for manual processes
48 2021 Bootstrap robust prescriptive analytics
49 2010 Building Business Intelligence Applications Having Prescriptive and Predictive Capabilities
50 2021 Catalyzing a Culture of Care and Innovation Through Prescriptive Analytics and Impact
Prediction to Create Full-Cycle Learning
51 2021 Catch me if you scan: Data-driven prescriptive modeling for smart store environments
52 2014 Characterised and personalised predictive-prescriptive analyticsusing agent-based simula-
tion
53 2019 Chassis Leasing and Selection Policy for Port Operations
54 2020 Closing the loop: Real-time Error Detection and Correction in automotive production using
Edge-/Cloud-Architecture and a CNN
55 2021 Condition-based critical level policy for spare parts inventory management
56 2018 Constituent Elements for Prescriptive Analytics Systems
57 2019 Context-aware based restaurant recommender system: A prescriptive analytics
58 2022 Coupled Learning Enabled Stochastic Programming with Endogenous Uncertainty
59 2021 Course performance prediction and evolutionary optimization for undergraduate engineer-
ing program towards admission strategic planning
60 2020 Cyber-Physical-Social System for Parallel Driving: From Concept to Application
61 2021 Data Analytics based Prescriptive Analytics for Selection of Lean Manufacturing System
62 2014 Data analytics using simulation for smart manufacturing
63 2018 Data Analytics: The next dimension in molding sand control
64 2011 Data is Dead… Without What-If Models
65 2021 Data-Driven Collaborative Human-AI Decision Making
66 2019 Data-Driven Design Optimization for Industrial Products
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
ID Year Article title
67 2021 Data-Driven Methodology to Support Long-Lasting Logistics and Decision Making for
Urban Last-Mile Operations
68 2020 Data-Driven Prescriptive Maintenance: Failure Prediction Using Ensemble Support Vector
Classification for Optimal Process and Maintenance Scheduling
69 2021 Decision Support for Knowledge Intensive Processes Using RL Based Recommendations
70 2023 Defining content marketing and its influence on online user behavior: a data-driven prescrip-
tive analytics method
71 2022 Design and Development of We-CDSS Using Django Framework: Conducing Predictive
and Prescriptive Analytics for Coronary Artery Disease
72 2020 Design and Evaluation of a Process-aware Recommender System based on Prescriptive
Analytics
73 2015 Design and Implementation of the LogicBlox System
74 2022 Developing a prescriptive decision support system for shop floor control
75 2018 Differentially Private Prescriptive Analytics
76 2022 Dynamic Pricing for New Products Using a Utility-Based Generalization of the Bass Diffu-
sion Model
77 2020 Dynamic Thresholding Leading to Optimal Inventory Maintenance
78 2016 Early Predictions of Movie Success: The Who, What, and When of Profitability
79 2020 Effective reinforcementlearningthroughevolutionarysurrogate-assistedprescription
80 2020 Employees recruitment: A prescriptive analytics approach via machine learning and math-
ematical programming
81 2019 Evaluation of the Selected Philippine E-Government Websites’ Performance withPrescrip-
tiveAnalysis
82 2016 EventAction: Visual analytics for temporal event sequence recommendation
83 2020 Expert-in-the-loop prescriptive analytics using mobility intervention for epidemics
84 2022 Explainable Process Prescriptive Analytics
85 2017 Fast integrated reservoir modelling on the Gjøa field offshore Norway
86 2019 Fault Classification and Correction based on Convolutional Neural Networks exemplified by
laser welding of hairpin windings
87 2013 Five pillars of prescriptive analytics success
88 2016 Fleet asset capacity analysis and revenue management optimization using advanced pre-
scriptive analytics
89 2016 Fossil-free public transport: Prescriptive policy analysis for the Swedish bus fleets
90 2018 France’s Governmental Big Data Analytics: From Predictive to Prescriptive Using R
91 2023 From “prepare for the unknown” to “train for what’s coming”: A digital twin-driven and
cognitive training approach for the workforce of the future in smart factories
92 2021 From Prediction to Prescription: Evolutionary Optimization of Nonpharmaceutical Interven-
tions in the COVID-19 Pandemic
93 2020 From predictive to prescriptive analytics
94 2020 From predictive to prescriptive analytics: A data-driven multi-item newsvendor model
95 2020 From predictive to prescriptive process monitoring: Recommending the next best actions
instead of calculating the next most likely events
96 2015 From predictive uplift modeling to prescriptive uplift analytics: A practical approach to
treatment optimization while accounting for estimation risk
97 2023 Fundamental challenge and solution methods in prescriptive analytics for freight transporta-
tion
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
ID Year Article title
98 2020 HadoopSec 2.0: Prescriptive analytics-based multi-model sensitivity-aware constraints
centric block placement strategy for Hadoop
99 2020 Howprescriptive analyticsinfluences decision making in precision medicine
100 2021 Human-augmented prescriptive analytics with interactive multi-objective reinforcement
learning
101 2020 Hybrid Data-Driven and Physics-Based Modeling for Gas TurbinePrescriptive Analytics
102 2022 Hybrid Neuro-Genetic Machine Learning Models for the Engineering of Ring-spun Cotton
Yarns
103 2017 Identifying cost-effective waterflooding optimization opportunities in mature reservoirs
from data driven analytics
104 2017 Impact of Business Analytics and Enterprise Systems on Managerial Accounting
105 2020 Improving harvesting operations in an oil palm plantation
106 2022 Improving Prescriptive Maintenance by Incorporating Post-Prognostic Information Through
Chance Constraints
107 2022 Improving the tactical planning of solid waste collection with prescriptive analytics: a case
study
108 2022 Improving Variable Orderings of Approximate Decision Diagrams Using Reinforcement
Learning
109 2019 Innovative InterLabs System for Smart Learning Analytics in Engineering Education
110 2018 Integrated assortment planning and store-wide shelf space allocation: An optimization-
based approach
111 2018 Integrative Analytics for Detecting and Disrupting Transnational Interdependent Criminal
Smuggling, Money, and Money-Laundering Networks
112 2011 Intelligent call routing: Optimizing contact center throughput
113 2021 Intervention Support Program for Students at Risk of Dropping Out Using Fuzzy Logic-
Based Prescriptive Analytics
114 2023 Inventory Waste Management with Augmented Analytics for Finished Goods
115 2021 JANOS: An Integrated Predictive and Prescriptive Modeling Framework
116 2014 Key Performance Indicators for Facility Performance Assessment: Simulation of Core
Indicators
117 2022 Landscape Optimization for Prescribed Burns in Wildfire Mitigation Planning
118 2020 Layered Behavior Modeling via Combining Descriptive and Prescriptive Approaches: A
Case Study of Infantry Company Engagement
119 2019 Learning failure modes of soil slopes using monitoring data
120 2019 Learning to Match via Inverse Optimal Transport
121 2022 Linking Predictive and Prescriptive Analytics of Elderly and Frail Patient Hospital Services
122 2020 Location-based socialsimulationforprescriptiveanalyticsof disease spread
123 2020 Machine Learning for Predictive and Prescriptive Analytics of Operational Data in Smart
Manufacturing
124 2021 Managing the Training Process in Elite Sports: From Descriptive to Prescriptive Data
Analytics
125 2015 Marketing Strategy Support System for Small Businesses
126 2015 Media company uses analytics to schedule radio advertisement spots
127 2013 Model-based decision support for optimal brochure pricing: applying advanced analytics in
the tour operating industry
128 2020 Model-predictive safety optimal actions to detect and handle process operation hazards
129 2022 Network Analytics for Infrastructure Asset Management Systemic Risk Assessment
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
ID Year Article title
130 2015 Nurse-patient assignment models considering patient acuity metrics and nurses’ perceived
workload
131 2019 Offline-Online Approximate Dynamic Programming for Dynamic Vehicle Routing with
Stochastic Requests
132 2017 On the adoption and impact of predictive analytics for server incident reduction
133 2022 Operations (management) warp speed: Rapid deployment of hospital-focused predictive/
prescriptive analytics for the COVID-19 pandemic
134 2022 Optimal policy trees
135 2017 OptimizationBeyond Prediction:PrescriptivePriceOptimization
136 2018 Optimized assignment patterns in Mobile Edge Cloud networks
137 2020 Optimized Maintenance Decision-Making—A Simulation-SupportedPrescriptive Analyt-
icsApproach Based on Probabilistic Cost–Benefit Analysis
138 2022 Optimizing diesel fuel supply chain operations to mitigate power outages for hurricane
relief
139 2018 Optimizing outpatient appointment system using machine learning algorithms and schedul-
ing rules: A prescriptive analytics framework
140 2023 Optimizing the preventive maintenance frequency with causal machine
141 2015 People Skills: Building Analytics Decision Models That Managers Use-A Change Manage-
ment Perspective
142 2021 Personalized Multimorbidity Management for Patients with Type 2 Diabetes Using Rein-
forcement Learning of Electronic Health Records
143 2017 Petroleum Analytics Learning Machine’ for optimizing the Internet of Things of today’s
digital oil field-to-refinery petroleum system
144 2021 PI prob: A risk prediction and clinical guidance system for evaluating patients with recur-
rent infections
145 2022 Pitfalls and protocols of data science in manufacturing practice
146 2022 Predict, then schedule: Prescriptive analytics approach for machine learning-enabled
sequential clinical scheduling
147 2022 Predicting biodiesel properties and its optimal fatty acid profile via explainable machine
learning
148 2016 Predictive analytics model for healthcare planning and scheduling
149 2019 Predictive and Prescriptive Analytics for Performance Optimization: Framework and a Case
Study on a Large-Scale Enterprise System
150 2021 Predictive and prescriptive analytics in transportation geotechnics: Three case studies
151 2023 Predictive and Prescriptive Business Process Monitoring with Reinforcement Learning
152 2019 Predictive andprescriptive analyticsfor location selection of add‐on retail products
153 2017 Predictive andprescriptive analytics, machine learning and child welfare risk assessment:
The Broward County experience
154 2022 Predictive machine learning for prescriptive applications: A coupled training–validating
approach
155 2019 Predictive, prescriptive and detective analytics for smart manufacturing in the information
age
156 2020 Prescriptive Analytics Aids Completion Optimization in Unconventionals
157 2023 Prescriptive analytics applications in sustainable operations research: conceptual framework
and future research challenges
158 2015 Prescriptive Analytics Applied to Brace Treatment for AIS: A Pilot Demonstration
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
ID Year Article title
159 2015 Prescriptive Analytics Based Autonomic Networking for Urban Streams Services Provision-
ing
160 2023 Prescriptive analytics for a multi-shift staffing problem
161 2013 Prescriptive Analytics for Allocating Sales Teams to Opportunities
162 2016 Prescriptive analytics for big data
163 2019 Prescriptive analytics for completion optimization in unconventional resources
164 2017 Prescriptive analytics for FIFA World Cup lodging capacity planning
165 2022 Prescriptive Analytics for finding the optimal manufacturing practice based on the simula-
tion models of Lean Manufacturing and Total Quality Management
166 2022 Prescriptive Analytics for Flexible Capacity Management
167 2019 Prescriptive analytics for human resource planning in the professional services industry
168 2021 Prescriptive analytics for impulsive behaviour prevention using real-time biometrics
169 2018 Prescriptive Analytics for MEC Orchestration
170 2015 Prescriptive analytics for planning research-performance strategy
171 2014 Prescriptive analytics for recommendation-based business process optimization
172 2020 Prescriptive analytics for reducing 30-day hospital readmissions after general surgery
173 2020 Prescriptive Analytics for Swapping Aircraft Assignments at All Nippon Airways
174 2016 Prescriptive analytics for understanding of out-of-plane deformation in additive manufactur-
ing
175 2018 Prescriptive analytics in airline operations: arrival time prediction and cost index optimiza-
tion for short-haul flights
176 2021 Prescriptive Analytics in Internet of Things with Concentration on Deep Learning
177 2022 Prescriptive Analytics in Procurement: Reducing Process Costs
178 2021 Prescriptive analytics in public-sector decision-making: A framework and insights from
charging infrastructure planning
179 2013 Prescriptive Analytics System for Improving Research Power
180 2014 Prescriptive analytics system for scholar research performance enhancement
181 2018 Prescriptive analytics through constrained Bayesian optimization
182 2021 Prescriptive analytics with differential privacy
183 2021 Prescriptive Analyticsfor Dynamic Real Time Scheduling of Diffusion Furnaces
184 2021 Prescriptive analyticsfor flexible capacity management
185 2020 Prescriptive analyticsfor inventory management in health care
186 2020 Prescriptive Analyticsfor Real-Time Optimization of Deepwater Casing Exits
187 2021 Prescriptive Analyticsin Urban Policing Operations
188 2015 Prescriptive analyticsusing synthetic information
189 2019 Prescriptive analytics: a survey of emerging trends and technologies
190 2020 Prescriptive analytics: Literature review and research challenges
191 2022 Prescriptive block replacement policy for production degrading systems
192 2020 Prescriptive business process monitoring for recommending next best actions
193 2019 Prescriptive cluster-dependent support vector machines with an application to reducing
hospital readmissions
194 2016 Prescriptive Control of Business Processes
195 2020 Prescriptive data analytics to optimize casing exits
196 2019 Prescriptive Equipment Maintenance: A Framework
197 2022 Prescriptive Healthcare Analytics: A Tutorial on Discrete Optimization and Simulation
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
ID Year Article title
198 2014 Prescriptive information fusion
199 2020 Prescriptive Learning for Air-Cargo Revenue Management
200 2019 Prescriptive Maintenance of Railway Infrastructure: From Data Analytics to Decision Sup-
port
201 2022 Prescriptive maintenance technique for photovoltaic systems
202 2020 Prescriptive Modelling System Design for an Armature Multi-coil Rewinding Cobot
Machine
203 2020 Prescriptive Process Analytics with Deep Learning and Explainable Artificial Intelligence
204 2022 Prescriptive process monitoring: Quo vadis?
205 2023 Prescriptive selection of machine learning hyperparameters with applications in power
markets: Retailer’s optimal trading
206 2022 Prescriptive Trees for Integrated Forecasting and Optimization Applied in Trading of
Renewable Energy
207 2018 Prescriptiveanalyticssystemfor long-range aircraft conflict detection and resolution
208 2017 Prescstream: A framework for streaming soft real-time predictive and prescriptive analytics
209 2020 Price Investment using Prescriptive Analytics and Optimization in Retail
210 2019 PriMa: a prescriptive maintenance model for cyber-physical production systems
211 2022 PROAD (Process Advisor): A health monitoring framework for centrifugal pumps
212 2021 Probation Status Prediction and Optimization for Undergraduate Engineering Students
213 2018 Product Portfolio Design Using Prescriptive Analytics
214 2019 Rapid spatio-temporal flood prediction and uncertainty quantification using a deep learning
method
215 2016 Realtime Predictive and Prescriptive Analytics with Real-time Data and Simulation
216 2022 Reducing the Total Product Cost at the Product Design Stage
217 2020 Replenishment and denomination mix of automated teller machines with dynamic forecast
demands
218 2020 Requirements for Prescriptive Recommender Systems Extending the Lifetime of EV Bat-
teries
219 2014 Research Advising System Based on Prescriptive Analytics
220 2018 Rh-rt: A Data Analytics Framework for Reducing Wait Time at Emergency Departments
and Centres for Urgent Care
221 2023 Rollout-based routing strategies with embedded prediction: A fish trawling application
222 2019 Route-cost-assignment with joint user and operator behavior as a many-to-one stable match-
ing assignment game
223 2018 Rules engine and complex event processor in the context of internet of things for precision
agriculture
224 2019 SAFE: A Comprehensive Data Visualization System
225 2021 Seat Assignments With Physical Distancing in Single-Destination Public Transit Settings
226 2022 Selecting advanced analytics in manufacturing: a decision support model
227 2020 Sensor-Driven Learning of Time-Dependent Parameters for Prescriptive Analytics
228 2015 Service-Delivery Modeling and Optimization
229 2020 Simulation as a decision-making tool in a business analytics environment
230 2021 Smart “Predict, then Optimize”
231 2017 Smart Maintenance Decision Support Systems (SMDSS) based on corporate big data
analytics
232 2019 Smart Manufacturing with Prescriptive Analytics
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
ID Year Article title
233 2022 Smart urban transport and logistics: A business analytics perspective
234 2016 Social media optimization: Identifying an optimal strategy for increasing network size on
Facebook
235 2017 Software estimation—towards prescriptive analytics
236 2021 SolveDB + : SQL-based prescriptive analytics
237 2022 Solving anInstance ofaRouting Problem Through Reinforcement Learning andHigh
Performance Computing
238 2014 Sonora: A Prescriptive Model for Message Authoring on Twitter
239 2022 Spare parts supply with incoming quality control and inspection errors in condition based
maintenance
240 2022 Stochastic dynamic vehicle routing in the light of prescriptive analytics: A review
241 2021 Stock market predictor usingprescriptive analytics
242 2014 System Thinking: Crafting Scenarios forPrescriptive Analytics
243 2023 The Analytics of Bed Shortages: Coherent Metric, Prediction, and Optimization
244 2017 The green fleet optimization model for a low-carbon economy: A prescriptive analytics
245 2023 The Impact of Dashboard Feedback Type on Learning Effectiveness, Focusing on Learner
Differences
246 2021 The Methodology of Hybrid Modelling for Gas Turbine Subsystems Prescriptive Analytics
247 2022 The role of optimization in some recent advances in data-driven decision-making
248 2021 To imprison or not to imprison: an analytics model for drug courts
249 2019 Topical Prescriptive Analytics System for Automatic Recommendation of Convergence
Technology
250 2019 Towards an automated optimization-as-a-service concept
251 2022 Uncertainty-bounded reinforcement learning for revenue optimization in air cargo: a pre-
scriptive learning approach
252 2023 University admission process: a prescriptive analytics approach
253 2019 Unleashing Analytics to Reduce Costs and Improve Quality in Wastewater Treatment
254 2018 Using Bayesian belief network and time-series model to conduct prescriptive and predictive
analytics for computer industries
255 2019 Using Prescriptive Data Analytics to Reduce Grading Bias and Foster Student Success
256 2020 Usingprescriptive analyticsfor the determination of optimal crop yield
257 2019 Usingprescriptive analyticsto support the continuous improvement process
258 2020 Verizon Uses Advanced Analytics to Rationalize Its Tail Spend Suppliers
259 2022 Virtual Material Quality Investigation System
260 2019 Visual PROMETHEE: Developments of the PROMETHEE & GAIA multicriteria decision
aid methods
261 2021 Wartime industrial logistics information integration: Framework and application in optimiz-
ing deployment and formation of military logistics platforms
262 2019 What to do when decision-makers deviate from model recommendations? Empirical evi-
dence from hydropower industry
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
Appendix C: Identied constituent components inliterature sample
Constituent components References from the literature sample discussing or men-
tioning the concepts
Quantity
Decision variables [2], [3], [4], [7], [8], [9], [10], [11], [13], [14], [15], [17],
[18], [19], [20], [25], [26], [27], [28], [29], [30], [33],
[36], [39], [40], [42], [43], [44], [45], [48], [53], [56],
[58], [59], [65], [66], [67], [68], [69], [73], [74], [75],
[76], [79], [80], [83], [84], [87], [90], [91], [92], [93],
[94], [96], [97], [100], [102], [103], [105], [106], [107],
[108], [110], [112], [115], [117], [120], [123], [126],
[127], [128], [129], [130], [131], [135], [136], [137],
[138], [139], [140], [145], [146], [147], [150], [152],
[154], [156], [160], [161], [163], [164], [165], [166],
[167], [169], [170], [173], [175], [176], [177], [178],
[181], [182], [183], [184], [185], [186], [187], [189],
[190], [191], [193], [194], [195], [198], [199], [200],
[201], [202], [205], [206], [207], [209], [212], [216],
[217], [221], [222], [225], [226], [230], [234], [236],
[237], [239], [240], [242], [243], [244], [247], [250],
[252], [253], [258], [259], [261], [262]
147
Objectives [2], [3], [4], [7], [8], [9], [10], [11], [13], [14], [15], [17],
[18], [19], [20], [24], [25], [26], [27], [28], [29], [30],
[33], [36], [39], [40], [42], [43], [44], [45], [48], [49],
[53], [56], [58], [59], [60], [65], [66], [67], [68], [69],
[73], [74], [75], [76], [79], [80], [83], [84], [87], [90],
[91], [92], [93], [94], [96], [97], [100], [102], [103],
[104], [105], [106], [107], [108], [110], [112], [115],
[117], [120], [123], [126], [127], [128], [129], [130],
[131], [135], [136], [137], [138], [139], [140], [143],
[145], [146], [147], [149], [150], [151], [152], [154],
[155], [156], [160], [161], [162], [163], [164], [166],
[167], [169], [170], [173], [175], [176], [177], [178],
[181], [182], [183], [184], [185], [186], [187], [189],
[190], [191], [193], [194], [195], [198], [199], [200],
[201], [202], [204], [205], [206], [207], [209], [212],
[213], [216], [217], [221], [222], [225], [226], [227],
[229], [230], [234], [236], [237], [239], [240], [242],
[243], [244], [247], [250], [252], [258], [259], [261],
[262]
158
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
Constituent components References from the literature sample discussing or men-
tioning the concepts
Quantity
Constraints [2], [3], [4], [7], [8], [13], [14], [15], [17], [19], [25], [26],
[27], [28], [29], [30], [33], [36], [39], [40], [42], [43],
[44], [45], [48], [49], [53], [56], [58], [65], [66], [67],
[68], [69], [72], [73], [74], [75], [76], [80], [84], [87],
[90], [91], [92], [93], [94], [96], [97], [100], [102], [103],
[104], [105], [106], [107], [108], [110], [112], [115],
[117], [120], [123], [126], [127], [128], [129], [130],
[135], [136], [137], [138], [139], [140], [145], [146],
[147], [152], [154], [155], [156], [160], [161], [162],
[163], [164], [166], [167], [169], [170], [173], [175],
[176], [177], [178], [181], [182], [183], [184], [185],
[186], [187], [189], [190], [191], [195], [198], [199],
[200], [201], [202], [205], [206], [207], [209], [216],
[217], [221], [222], [225], [226], [227], [230], [234],
[236], [239], [240], [242], [243], [244], [247], [250],
[253], [258], [259], [261], [262]
137
Current state [1], [3], [4], [6], [9], [10], [11], [15], [16], [17], [20], [23],
[24], [25], [26], [27], [30], [34], [35], [36], [37], [38],
[40], [41], [42], [45], [50], [52], [53], [56], [59], [60],
[65], [66], [67], [68], [69], [70], [78], [80], [82], [90],
[93], [94], [96], [97], [99], [100], [103], [104], [105],
[107], [108], [109], [110], [111], [112], [113], [114],
[117], [118], [123], [125], [127], [129], [130], [132],
[133], [136], [137], [138], [139], [140], [145], [146],
[147], [152], [155], [162], [163], [166], [169], [173],
[176], [178], [179], [180], [182], [184], [185], [186],
[189], [190], [191], [192], [194], [195], [196], [200],
[202], [205], [207], [209], [210], [211], [212], [216],
[217], [218], [219], [220], [221], [223], [224], [226],
[228], [229], [231], [233], [235], [236], [240], [241],
[243], [247], [248], [249], [253], [257], [258], [259],
[261], [262]
133
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
Constituent components References from the literature sample discussing or men-
tioning the concepts
Quantity
Probabilities [1], [2], [3], [6], [7], [9], [10], [11], [12], [13], [14], [15],
[17], [18], [19], [20], [22], [24], [25], [27], [28], [29],
[30], [31], [32], [34], [35], [36], [37], [38], [39], [40],
[41], [42], [45], [47], [48], [49], [50], [51], [52], [53],
[54], [55], [56], [57], [58], [59], [60], [63], [65], [66],
[67], [68], [69], [70], [71], [72], [73], [74], [75], [77],
[78], [79], [80], [82], [84], [86], [87], [88], [90], [91],
[92], [93], [94], [95], [96], [98], [99], [100], [101], [102],
[103], [104], [105], [106], [108], [109], [110], [111],
[112], [114], [115], [119], [120], [121], [123], [125],
[127], [128], [129], [131], [132], [133], [134], [135],
[137], [138], [139], [140], [143], [144], [145], [146],
[147], [148], [149], [151], [152], [153], [154], [155],
[156], [158], [160], [161], [162], [163], [164], [166],
[167], [168], [171], [172], [175], [176], [177], [178],
[181], [182], [184], [185], [186], [187], [189], [190],
[191], [192], [193], [194], [195], [196], [198], [199],
[202], [203], [204], [205], [206], [207], [208], [209],
[210], [211], [212], [213], [214], [215], [216], [217],
[218], [220], [221], [223], [224], [226], [228], [229],
[230], [231], [233], [234], [235], [236], [238], [239],
[240], [241], [246], [247], [248], [251], [252], [253],
[254], [256], [257], [258], [259], [261], [262]
201
Mathematical program [2], [3], [4], [7], [9], [10], [13], [14], [15], [16], [17], [19],
[20], [25], [27], [28], [29], [30], [32], [33], [35], [36],
[39], [40], [41], [42], [43], [44], [45], [47], [48], [49],
[51], [53], [56], [58], [65], [67], [68], [73], [74], [75],
[76], [80], [81], [83], [84], [87], [88], [89], [93], [94],
[96], [97], [99], [100], [101], [103], [104], [105], [106],
[107], [110], [112], [115], [117], [120], [121], [124],
[126], [127], [129], [130], [131], [135], [136], [138],
[140], [146], [149], [150], [152], [154], [155], [156],
[157], [160], [161], [162], [163], [164], [166], [167],
[169], [170], [173], [174], [175], [176], [177], [178],
[181], [182], [183], [184], [185], [186], [187], [189],
[190], [191], [194], [195], [197], [198], [199], [200],
[201], [204], [205], [206], [207], [209], [210], [216],
[217], [218], [220], [221], [222], [225], [226], [228],
[229], [230], [233], [236], [239], [240], [243], [247],
[250], [251], [252], [253], [258], [260], [261], [262]
149
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
Constituent components References from the literature sample discussing or men-
tioning the concepts
Quantity
Machine learning [1], [2], [6], [9], [10], [11], [12], [13], [14], [15], [16],
[17], [18], [19], [20], [23], [24], [25], [26], [27], [29],
[30], [32], [34], [35], [36], [37], [40], [41], [42], [45],
[48], [50], [51], [53], [54], [56], [57], [58], [59], [60],
[64], [65], [66], [67], [68], [69], [70], [71], [72], [73],
[74], [77], [78], [79], [80], [83], [84], [86], [88], [90],
[92], [93], [94], [95], [96], [98], [99], [100], [101], [102],
[103], [104], [105], [106], [108], [109], [112], [114],
[115], [120], [121], [123], [127], [129], [132], [133],
[134], [135], [137], [139], [140], [142], [143], [145],
[146], [147], [148], [149], [150], [151], [152], [153],
[154], [155], [156], [157], [159], [160], [161], [162],
[163], [166], [168], [169], [171], [172], [175], [176],
[182], [184], [185], [186], [187], [189], [190], [192],
[193], [195], [196], [198], [199], [202], [203], [204],
[205], [206], [208], [209], [210], [211], [212], [213],
[214], [216], [218], [220], [221], [224], [226], [229],
[230], [233], [234], [236], [237], [238], [240], [241],
[246], [247], [248], [250], [251], [252], [253], [254],
[255], [256], [258], [259]
171
Evolutionary comp [8], [18], [25], [37], [59], [66], [67], [83], [90], [92], [102],
[104], [117], [128], [145], [147], [156], [176], [181],
[190], [212], [226], [229], [233], [234], [236], [259]
27
Simulation [17], [19], [21], [30], [40], [41], [52], [53], [56], [61],
[62], [64], [67], [72], [74], [83], [87], [88], [89], [91],
[92], [99], [101], [103], [104], [105], [107], [109], [116],
[117], [122], [132], [133], [135], [137], [139], [143],
[157], [158], [160], [162], [172], [176], [188], [189],
[190], [192], [197], [198], [199], [205], [210], [213],
[215], [216], [220], [224], [226], [228], [229], [233],
[238], [243], [250], [254], [259], [261]
67
Logic-based models [90], [113], [118], [165], [171], [190], [204], [211], [226],
[229], [231], [233]
12
Probabilistic models [3], [21], [25], [27], [50], [55], [73], [75], [90], [104],
[110], [118], [119], [129], [137], [144], [145], [151],
[158], [168], [177], [189], [190], [196], [198], [200],
[204], [207], [210], [216], [220], [221], [226], [227],
[228], [229], [233], [254]
38
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
Constituent components References from the literature sample discussing or men-
tioning the concepts
Quantity
Single decision [[1], [2], [3], [4], [7], [8], [9], [10], [11], [13], [14], [15],
[16], [17], [19], [20], [26], [27], [28], [29], [30], [36],
[37], [39], [40], [42], [43], [45], [49], [53], [54], [56],
[58], [65], [66], [67], [68], [69], [74], [76], [79], [80],
[84], [86], [89], [91], [92], [93], [94], [96], [98], [100],
[102], [103], [105], [106], [107], [108], [110], [112],
[114], [115], [117], [118], [119], [123], [126], [127],
[128], [129], [130], [131], [132], [133], [134], [135],
[136], [137], [138], [140], [142], [146], [147], [148],
[149], [154], [160], [161], [163], [164], [165], [167],
[168], [169], [171], [173], [174], [175], [177], [178],
[181], [183], [185], [187], [190], [192], [193], [194],
[195], [199], [200], [202], [203], [205], [206], [207],
[209], [211], [216], [217], [221], [222], [225], [226],
[227], [228], [230], [234], [237], [239], [240], [243],
[244], [247], [251], [252], [253], [258], [259], [262]
128
Multiple decisions [4], [40], [84], [91], [92], [105], [112], [114], [115], [134],
[136], [151], [169], [190], [209], [211], [213], [216],
[222], [231], [257]
21
execution [16], [21], [23], [24], [51], [54], [60], [67], [77], [82],
[91], [98], [112], [128], [136], [155], [159], [163], [192],
[196], [199], [202], [210], [223], [251]
25
Adaptation [4], [11], [12], [16], [21], [24], [32], [40], [51], [54], [60],
[65], [67], [77], [79], [82], [83], [92], [100], [114], [123],
[136], [159], [162], [194], [196], [198], [210], [218],
[220], [227], [234], [250], [262]
34
Integration [9], [12], [23], [26], [30], [31], [34], [35], [36], [38], [40],
[41], [49], [54], [59], [60], [64], [65], [67], [74], [77],
[100], [112], [114], [123], [125], [129], [132], [143],
[145], [159], [162], [163], [173], [189], [190], [194],
[196], [208], [210], [211], [223], [227], [228], [253],
[257], [258], [261]
48
Distributed computing [12], [23], [30], [31], [54], [114], [162], [189], [190],
[194], [196], [208], [209], [210], [223], [236]
16
Modulization [6], [12], [14], [21], [23], [25], [26], [30], [32], [34], [35],
[36], [38], [40], [41], [45], [47], [54], [56], [67], [68],
[71], [72], [74], [77], [79], [80], [91], [92], [98], [104],
[105], [109], [111], [112], [113], [114], [117], [118],
[123], [125], [128], [129], [132], [133], [136], [139],
[145], [146], [147], [149], [162], [163], [189], [190],
[194], [196], [208], [209], [210], [211], [214], [216],
[219], [220], [223], [224], [227], [228], [229], [231],
[233], [234], [249], [251], [252], [253], [258], [259],
[261]
80
Security- and privacy-preserving [34], [38], [59], [75], [82], [98], [159], [182], [190], [210] 10
Workflow interface [5], [6], [9], [12], [30], [32], [34], [36], [38], [40], [41],
[49], [65], [67], [73], [82], [91], [92], [100], [104], [114],
[115], [123], [163], [173], [189], [190], [194], [210],
[216], [218], [219], [220], [223], [225], [227], [228],
[236], [250], [251], [257], [258], [262]
43
Explainability [84], [147], [203] 3
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
Constituent components References from the literature sample discussing or men-
tioning the concepts
Quantity
Visualization [3], [5], [6], [9], [12], [21], [23], [34], [35], [36], [38],
[40], [41], [45], [46], [49], [56], [64], [65], [67], [70],
[71], [77], [79], [82], [84], [90], [91], [100], [103], [104],
[109], [114], [115], [117], [123], [129], [132], [143],
[145], [162], [163], [173], [179], [180], [189], [190],
[194], [195], [196], [200], [203], [207], [209], [210],
[211], [216], [219], [221], [224], [225], [228], [229],
[236], [241], [245], [256], [257], [258], [260], [261]
71
Extensibility [30], [40], [49], [73], [189], [208], [218], [236] 8
Appendix D: Identied technology aordances inliterature sample
Affordance (effect); frequency Example studies (IDs)
Maintenance planning for optimal maintenance
schedule; n = 24
[3], [4], [11], [12], [20], [21], [22], [54], [55], [68],
[77], [86], [101], [106], [137], [140], [171], [191],
[196], [210], [211], [231], [239], [246]
Production planning for optimal manufacturing
schedule; n = 13
[26], [37], [43], [47], [61], [74], [102], [123], [165],
[183], [226], [257], [259]
Product (portfolio) design optimization; n = 4 [66], [202], [213], [216]
Operations safety improvement and planning;
n = 1
[128]
Industrial worker training optimization; n = 1 [91]
Deformation control in additive manufacturing;
n = 1
[174]
Optimization of routing and scheduling; n = 15 [17], [19], [60], [67], [89], [97], [131], [159], [175],
[207], [222], [233], [237], [240], [244]
Capacity/cargo management and improvement;
n = 8
[53], [88], [166], [173], [184], [199], [225], [251]
Vehicle maintenance planning; n = 2 [10], [200]
Patient treatment planning and improvement; n = 8 [39], [99], [121], [142], [144], [158], [172], [193]
Patient scheduling; n = 8 [8], [130], [139], [146], [148], [197], [220], [243]
Pandemic/epidemic intervention planning; n = 5 [6], [83], [92], [122], [133]
Human health tracking and improvement; n = 4 [1], [71], [168]
Assortment and inventory planning (health); n = 1 [158]
Clinical investment management; n = 1 [15]
Optimizing power system/grid operations; n = 5 [7], [16], [25], [31], [262]
Disaster preparation/recovery planning; n = 3 [117], [138], [214]
Electricity brokerage optimization; n = 3 [24], [205], [206]
Maintenance planning (energy & environment);
n = 3
[27], [45], [201]
Wastewater treatment improvement; n = 2 [18], [253]
Waste collection and management planning; n = 2 [107], [114]
Optimization of deepwater casing exits; n = 2 [186], [195]
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
Affordance (effect); frequency Example studies (IDs)
Waterflooding process optimization; n = 1 [103]
Battery lifetime optimization; n = 1 [218]
Reservoir design planning; n = 1 [32]
Soil slope analysis; n = 1 [119]
Price optimization; n = 4 [46], [76], [135], [209]
Assortment and inventory planning; n = 3 [94], [110], [152]
Sales team assignments; n = 2 [36], [161]
Customer characterization; n = 1 [52]
Customer service recommendation; n = 1 [96]
Theft surveillance and automated checkout; n = 1 [51]
Academic performance improving; n = 6 [34], [38], [50], [109], [245], [255]
Dropout prevention planning; n = 2 [113], [212]
Admissions planning and selection; n = 2 [59], [252]
Maximize oil/gas recovery; n = 4 [85], [143], [156], [163]
Mining fleet scheduling; n = 1 [33]
Sand molding process improvement; n = 1 [63]
Laboratory task allocation and planning; n = 1 [35]
Biodiesel properties optimization; n = 1 [147]
Social media usage optimization; n = 2 [234], [238]
Network and computing resource orchestration;
n = 2
[136], [169]
Software development estimation; n = 1 [235]
Website performance analysis and optimization;
n = 1
[81]
Research advising; n = 6 [170], [179], [180], [219], [242], [249]
Harvesting operations planning and optimization,
n = 3
[90], [105], [223]
Crop yield optimization; n = 1 [256]
Fish trawling routing and optimization; n = 1 [221]
Law enforcement resource allocation and plan-
ning; n = 2
[111], [187]
Imprisonment decision planning and recommen-
dation; n = 1
[164]
Tournament lodging planning; n = 1 [248]
Sports event safety management and planning;
n = 1
[224]
Athlete training process improvement; n = 1 [124]
Infantry engagement planning; n = 1 [118]
Military logistics planning; n = 1 [261]
Markdown planning and price optimization; n = 1 [28]
Oder delivery scheduling; n = 1 [14]
Teller machine replenishment planning and
allocation; n = 1
[217]
Stock purchase recommendations; n = 1 [241]
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
Affordance (effect); frequency Example studies (IDs)
Radio advertising scheduling; n = 1 [126]
Movie planning and profit-maximizing; n = 1 [78]
Project staffing planning and allocation; n = 2 [167], [228]
Child welfare assessment; n = 1 [153]
Infrastructure planning and optimization; n = 1 [178]
Optimal tour pricing; n = 1 [127]
Restaurant recommendations; n = 1 [57]
Prescriptive process management; n = 8 [40], [84], [95], [151], [192], [194], [203], [204]
Employee recruiting and staffing; n = 3 [13], [80], [160]
Procurement and supplier management; n = 2 [177], [258]
Marketing management; n = 2 [70], [125]
Facility/asset management; n = 2 [116], [129]
Managerial Accounting; n = 1 [104]
Call center routing; n = 1 [112]
Server incident management and prevention; n = 1 [132]
Appendix E: Identied system archetypes
System archetypes References from the literature sample
Advisory PAS [1], [2], [3], [5], [6], [7], [8], [9], [10], [13], [14], [15], [17], [18], [19],
[20], [22], [26], [27], [28], [29], [31], [33], [34], [35], [36], [37], [38],
[39], [41], [42], [43], [44], [45], [47], [48], [49], [52], [53], [55], [57],
[58], [59], [61], [62], [63], [64], [66], [68], [69], [70], [71], [72], [73],
[74], [75], [76], [78], [80], [81], [84], [85], [86], [88], [89], [90], [93],
[94], [95], [96], [97], [99], [101], [102], [103], [104], [105], [106], [107],
[109], [110], [111], [113], [115], [116], [117], [118], [119], [120], [121],
[122], [124], [125], [126], [127], [129], [130], [131], [132], [133], [134],
[135], [137], [138], [139], [140], [142], [143], [144], [146], [147], [148],
[149], [151], [152], [153], [154], [156], [158], [160], [161], [164], [165],
[166], [167], [168], [169], [170], [171], [172], [173], [174], [175], [177],
[178], [179], [180], [181], [182], [183], [184], [185], [186], [187], [188],
[191], [193], [195], [197], [200], [201], [203], [205], [206], [207], [208],
[209], [211], [212], [213], [214], [215], [216], [217], [219], [221], [222],
[224], [225], [226], [228], [229], [230], [231], [236], [237], [238], [239],
[240], [241], [242], [243], [244], [246], [247], [248], [249], [252], [253],
[254], [255], [256], [257], [258], [259], [260], [261]
Executive PAS [23], [91], [98], [112], [128], [155], [163], [192], [199], [202], [223], [251]
Adaptive PAS [4], [11], [12], [32], [40], [65], [79], [83], [92], [100], [114], [123], [162],
[194], [198], [218], [220], [227], [234], [250], [262]
Self-governing PAS [16], [21] [24], [51], [54], [60], [67], [77], [82], [136], [159], [196], [210]
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
Appendix F: Decision processing techniques
In the following, we give an overview of exemplary technique subcategories used in
prescriptive analytics, following the survey of Lepenioti etal. (2020).
Decision processing Example techniques
Mathematical programming Mixed integer programming, linear programming, binary quadratic pro-
gramming, non-linear programming, stochastic optimization, conditional
stochastic optimization, constrained Bayesian optimization, fuzzy linear
programming, dynamic programming
Machine learning Various clustering algorithms, reinforcement learning, Boltzmann
machine, (deep) artificial neural networks
Evolutionary computation Genetic algorithms, evolutionary optimization, greedy algorithms, particle
swarm optimization
Simulation Simulation over random forest, risk assessments, stochastic simulations,
what-if scenarios
Logic-based models Association rules, decision rules, criteria-based rules, fuzzy rules,
distributed rules, benchmark rules, desirability functions, graph-based
recommendations
Probabilistic models Markov decision processes, hidden Markov models, Markov chains
Appendix G: Overview ofdata properties
Concepts References from the literature sample discussing the properties
of data
Data type Structured [1], [3], [4], [6], [8], [9], [14], [15], [17], [21], [22], [28], [30],
[32], [34], [35], [36], [37], [38], [40], [42], [45], [47], [49],
[51], [52], [53], [59], [63], [64], [65], [66], [67], [68], [69],
[74], [77], [78], [80], [82], [85], [87], [88], [90], [92], [93],
[94], [99], [100], [101], [104], [109], [110], [112], [114],
[115], [120], [123], [125], [126], [128], [129], [130], [132],
[133], [136], [137], [139], [140], [142], [143], [147], [151],
[152], [153], [154], [156], [158], [162], [163], [164], [166],
[167], [168], [171], [172], [173], [175], [178], [182], [184],
[185], [187], [188], [189], [190], [191], [194], [195], [196],
[201], [202], [204], [206], [208], [209], [210], [211], [212],
[213], [216], [218], [220], [221], [223], [224], [225], [227],
[228], [229], [230], [232], [234], [235], [236], [241], [246],
[248], [251], [252], [253], [254], [257], [258], [259], [260],
[261]
Unstructured [1], [3], [12], [21], [30], [34], [42], [45], [47], [48], [52], [54],
[64], [66], [67], [69], [70], [78], [80], [85], [86], [87], [90],
[92], [93], [99], [104], [109], [113], [117], [118], [125], [129],
[130], [132], [137], [143], [147], [151], [153], [163], [167],
[179], [182], [187], [188], [189], [190], [195], [202], [204],
[209], [210], [211], [212], [214], [220], [223], [224], [227],
[234], [235], [249], [258], [261]
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
Concepts References from the literature sample discussing the properties
of data
Data velocity Real-time/streaming [1], [4], [12], [16], [20], [21], [22], [23], [24], [25], [26], [30],
[31], [35], [40], [45], [47], [49], [53], [54], [55], [56], [60],
[65], [67], [69], [74], [90], [92], [100], [103], [104], [112],
[115], [123], [128], [129], [131], [137], [138], [142], [143],
[155], [159], [162], [168], [171], [172], [183], [186], [189],
[190], [192], [194], [195], [196], [199], [204], [207], [208],
[210], [214], [215], [220], [223], [224], [225], [227], [231],
[232], [233], [237], [239], [241], [246], [253], [261]
Historical/batches [1], [4], [6], [12], [14], [15], [17], [20], [21], [22], [25], [27],
[28], [30], [31], [32], [34], [35], [36], [37], [40], [45], [47],
[49], [52], [53], [56], [60], [63], [65], [67], [68], [69], [70],
[74], [78], [79], [88], [90], [92], [93], [94], [96], [97], [100],
[103], [104], [109], [110], [112], [114], [115], [120], [123],
[125], [127], [128], [129], [130], [132], [133], [135], [137],
[138], [139], [142], [143], [146], [147], [148], [149], [152],
[153], [155], [158], [162], [163], [164], [166], [167], [168],
[169], [171], [172], [175], [185], [186], [187], [189], [190],
[191], [192], [193], [194], [195], [196], [199], [200], [201],
[204], [207], [208], [209], [210], [211], [212], [213], [214],
[215], [216], [217], [219], [220], [221], [223], [224], [227],
[228], [229], [231], [233], [235], [236], [238], [241], [246],
[248], [251], [252], [253], [254], [258], [259], [261]
Data origin External [13], [14], [15], [17], [34], [45], [49], [52], [56], [67], [70], [78],
[88], [90], [92], [93], [95], [96], [104], [112], [115], [117],
[125], [126], [133], [137], [138], [139], [143], [145], [168],
[196], [203], [209], [210], [220], [223], [228], [233], [241],
[249], [252], [254], [257], [258], [259], [261]
Internal [4], [8], [12], [14], [15], [16], [17], [34], [35], [36], [45], [47],
[49], [52], [56], [67], [68], [74], [77], [88], [90], [93], [96],
[104], [109], [110], [112], [114], [115], [123], [125], [126],
[132], [133], [137], [139], [143], [145], [146], [152], [168],
[175], [185], [187], [191], [194], [195], [196], [203], [209],
[210], [211], [216], [220], [223], [228], [233], [253], [258],
[259], [261]
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
Concepts References from the literature sample discussing the properties
of data
Data generation Empirical [1], [3], [4], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17],
[18], [19], [20], [21], [22], [24], [27], [29], [30], [32], [34],
[36], [37], [39], [40], [45], [47], [49], [50], [51], [52], [53],
[54], [55], [56], [57], [58], [59], [60], [61], [63], [65], [66],
[67], [68], [71], [72], [74], [75], [76], [77], [78], [80], [81],
[82], [85], [86], [88], [90], [92], [93], [94], [95], [96], [97],
[98], [100], [101], [102], [103], [104], [105], [106], [108],
[109], [110], [112], [114], [115], [117], [120], [121], [123],
[125], [126], [128], [129], [130], [131], [132], [133], [135],
[136], [137], [138], [139], [140], [142], [143], [144], [146],
[147], [148], [149], [150], [151], [152], [153], [154], [156],
[158], [160], [161], [163], [164], [165], [166], [167], [168],
[169], [171], [172], [173], [175], [177], [178], [181], [182],
[184], [185], [186], [187], [190], [191], [192], [193], [194],
[195], [196], [199], [200], [201], [202], [203], [205], [206],
[207], [208], [209], [210], [211], [212], [213], [214], [216],
[217], [218], [219], [220], [221], [223], [224], [225], [226],
[227], [228], [230], [231], [232], [233], [234], [237], [239],
[240], [241], [243], [246], [248], [249], [251], [252], [253],
[254], [255], [256], [257], [258], [259], [260], [261], [262]
Synthetic [28], [30], [41], [43], [48], [53], [60], [62], [74], [75], [92],
[108], [116], [117], [120], [122], [127], [134], [136], [154],
[177], [181], [182], [188], [224], [230], [252]
Assumptions [1], [11], [20], [27], [32], [52], [56], [65], [66], [77], [80], [83],
[90], [92], [100], [104], [113], [116], [118], [123], [125],
[127], [130], [134], [137], [144], [156], [166], [190], [200],
[202], [218], [224], [228], [231], [239], [246], [255], [258],
[260]
Appendix H: Temporal trend analysis ofresults
0
10
20
30
40
50
60
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022/23
Decision Formulation
Decision Variables Objectives Constraints
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
0
10
20
30
40
50
60
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022/23
Decision Input
Current StateProbabilities
0
10
20
30
40
50
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022/23
Decision Output
Single Decision Multiple Decisions
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
0
5
10
15
20
25
30
35
40
45
50
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022/23
Decision Processing Techniques
Math. ProgrammingMachine LearningEvolutionary Comp.
Simulation Logic-based Models Probabilistic Models
0
2
4
6
8
10
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022/23
Action Mechanisms
Execution Mechanism Adaptation Mechanism
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
0
2
4
6
8
10
12
14
16
18
20
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022/23
Ancillary Features
IntegrationDistributed Computing
Modularization Security- and Privacy-Preserving
Workflow SupportExtensibility
VisualizationExplainability
0
10
20
30
40
50
60
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022/23
PAS Archetypes
Advisory PASExecutive PASAdaptive PAS Self-governing PAS
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
0
2
4
6
8
10
12
14
16
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022/23
Industries
ManufacturingTransport. & LogisticsHealth & Medtech
Energy & Environm. Retail & Trade Education
Chem. & Resources
Funding Open Access funding enabled and organized by Projekt DEAL.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License,
which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long
as you give appropriate credit to the original author(s) and the source, provide a link to the Creative
Commons licence, and indicate if changes were made. The images or other third party material in this
article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line
to the material. If material is not included in the article’s Creative Commons licence and your intended
use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permis-
sion directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/
licenses/by/4.0/.
References
Abdollahnejadbarough H, Mupparaju KS, Shah S, Golding CP, Leites AC, Popp TD, Shroyer E, Golany
YS, Robinson AG, Akgun V (2020) Verizon uses advanced analytics to rationalize its tail spend
suppliers. INFORMS J on Appl Anal 50(3):197–211. https:// doi. org/ 10. 1287/ inte. 2020. 1038
Ahmed I, Ahmad M, Jeon G, Piccialli F (2021) A Framework for pandemic prediction using big data
analytics. Big Data Res 25:100190. https:// doi. org/ 10. 1016/j. bdr. 2021. 100190
Aier S, Fischer C (2011) Criteria of progress for information systems design theories. Inf Syst E-Bus
Manage 9(1):133–172. https:// doi. org/ 10. 1007/ s10257- 010- 0130-8
Ali AR (2011) Intelligent call routing: optimizing contact center throughput. In: Proceedings of the Elev-
enth International Workshop on Multimedia Data Mining
Anderson C, Robey D (2017) Affordance potency: explaining the actualization of technology affordances.
Inf Organ 27(2):100–115. https:// doi. org/ 10. 1016/j. infoa ndorg. 2017. 03. 002
Anglou F-Z, Ponis S, Spanos A (2021) A machine learning approach to enable bulk orders of critical
spare-parts in the shipping industry. J Ind Eng 14(3):604. https:// doi. org/ 10. 3926/ jiem. 3446
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
Ansari F, Glawar R, Nemeth T (2019) PriMa: a prescriptive maintenance model for cyber-physical pro-
duction systems. Int J Comput Integr Manuf 32(4–5):482–503. https:// doi. org/ 10. 1080/ 09511 92X.
2019. 15712 36
Appelbaum D, Kogan A, Vasarhelyi M, Yan Z (2017) Impact of business analytics and enterprise systems
on managerial accounting. Int J of Account Inf Syst 25:29–44. https:// doi. org/ 10. 1016/j. accinf.
2017. 03. 003
Ayhan S, Costas P, Samet H (2018) Prescriptive analytics system for long-range aircraft conflict detec-
tion and resolution. In: Proceedings of the 26th ACM SIGSPATIAL International Conference on
Advances in Geographic Information Systems. pp 239–248
Baird A, Maruping LM (2021) The next generation of research on is use: a theoretical framework of
delegation to and from agentic IS artifacts. MISQ 45(1):315–341. https:// doi. org/ 10. 25300/ MISQ/
2021/ 15882
Ballings M, Van den Poel D, Bogaert M (2016) Social media optimization: identifying an optimal strat-
egy for increasing network size on Facebook. Omega 59:15–25. https:// doi. org/ 10. 1016/j. omega.
2015. 04. 017
Basdere M, Caniglia G, Collar C, Rozolis C, Chiampas G, Nishi M, Smilowitz K (2019) SAFE: a com-
prehensive data visualization system. INFORMS J on Appl Anal 49(4):249–261. https:// doi. org/ 10.
1287/ inte. 2019. 0989
Belciug S, Gorunescu F (2016) A hybrid genetic algorithm-queuing multi-compartment model for opti-
mizing inpatient bed occupancy and associated costs. Artif Intell in Med 68:59–69. https:// doi. org/
10. 1016/j. artmed. 2016. 03. 001
Bertsimas D, Kallus N (2020) From predictive to prescriptive analytics. Manage Sci 66(3):1025–1044.
https:// doi. org/ 10. 1287/ mnsc. 2018. 3253
Bhatt D, Naqvi S, Gunasekaran A, Dutta V (2023) Prescriptive analytics applications in sustainable oper-
ations research: conceptual framework and future research challenges. Ann Oper Res. https:// doi.
org/ 10. 1007/ s10479- 023- 05251-3
Bostrom RP, Heinen JS (1977) MIS Problems and Failures: a socio-technical perspective, part II: the
application of socio-technical theory. MIS Q 1(4):11. https:// doi. org/ 10. 2307/ 249019
Bousdekis A, Papageorgiou N, Magoutas B, Apostolou D, Mentzas G (2020) Sensor-driven learning of
time-dependent parameters for prescriptive analytics. IEEE Access 8:92383–92392. https:// doi. org/
10. 1109/ ACCESS. 2020. 29949 33
Brandt T, Dlugosch O, Abdelwahed A, van den Berg PL, Neumann D (2021) Prescriptive analytics in
urban policing operations. Manuf Serv Oper Manage 24(5):2463–2480. https:// doi. org/ 10. 1287/
msom. 2021. 1022
vom Brocke J, Simons A, Riemer K, Niehaves B, Plattfaut R, Cleven A (2015) Standing on the shoulders
of giants: challenges and recommendations of literature search in information systems research.
Commun Assoc Inf Syst 37:206–224. https:// doi. org/ 10. 17705/ 1CAIS. 03709
Burton-Jones A, Volkoff O (2017) How can we develop contextualized theories of effective use? a dem-
onstration in the context of community-care electronic health records. Inf Syst Res 28(3):468–489.
https:// doi. org/ 10. 1287/ isre. 2017. 0702
Caro F, de Tejada Cuenca AS (2023) Believing in analytics: managers’ adherence to price recommenda-
tions from a DSS. M&SOM 25(2):524–542. https:// doi. org/ 10. 1287/ msom. 2022. 1166
Ceselli A, Fiore M, Premoli M, Secci S (2019) Optimized assignment patterns in mobile edge cloud net-
works. Comput Oper Res 106:246–259. https:// doi. org/ 10. 1016/j. cor. 2018. 02. 022
Ceselli A, Fiore M, Furno A, Premoli M, Secci S, Stanica R (2018) Prescriptive Analytics for MEC
Orchestration. In: 2018 IFIP Networking Conference (IFIP Networking) and Workshops. IEEE,
Zurich, Switzerland, pp 1–9
Chatterjee S, Sarker S, Lee MJ, Xiao X, Elbanna A (2021) A possible conceptualization of the informa-
tion systems (IS) artifact: A general systems theory perspective. Inf Syst J 31(4):550–578. https://
doi. org/ 10. 1111/ isj. 12320
Chowdhary D, Khandelwal H (2018) Data Analytics: The next dimension in molding sand control. In:
The 73rd World Foundry Congress
Consilvio A, Sanetti P, Anguita D, Crovetto C, Dambra C, Oneto L, Papa F, Sacco N (2019) Prescriptive
Maintenance of Railway Infrastructure: From Data Analytics to Decision Support. In: 2019 6th
International Conference on Models and Technologies for Intelligent Transportation Systems (MT-
ITS). IEEE, Cracow, Poland, pp 1–10
Cooper HM (1988) Organizing knowledge syntheses: a taxonomy of literature reviews. Knowl in Soc
1(1):104–126. https:// doi. org/ 10. 1007/ BF031 77550
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
Darioshi R, Lahav E (2021) The impact of technology on the human decision-making process. Human
Behav and Emerg Tech 3(3):391–400. https:// doi. org/ 10. 1002/ hbe2. 257
de Jesus CG, Ledda MKC (2021) Intervention Support Program for Students at Risk of Dropping Out
Using Fuzzy Logic-Based Prescriptive Analytics. In: 2021 IEEE 17th International Colloquium on
Signal Processing & Its Applications (CSPA). IEEE, Langkawi, Malaysia, pp 144–149
Delen D, Ram S (2018) Research challenges and opportunities in business analytics. J Bus Anal 1(1):2–
12. https:// doi. org/ 10. 1080/ 25732 34X. 2018. 15073 24
Delen D, Zolbanin HM (2018) The analytics paradigm in business research. J Bus Res 90:186–195.
https:// doi. org/ 10. 1016/j. jbusr es. 2018. 05. 013
Dellermann D, Ebel P, Söllner M, Leimeister JM (2019) Hybrid intelligence. Bus Inf Syst Eng 61(5):637–
643. https:// doi. org/ 10. 1007/ s12599- 019- 00595-2
Dey S, Gupta N, Pathak S, Kela DH, Datta S (2019) Data-Driven Design Optimization for Industrial
Products. In: Datta S, Davim JP (eds) Optimization in Industry. Springer International Publishing,
Cham, pp 253–267
Effah J, Amankwah-Sarfo F, Boateng R (2021) Affordances and constraints processes of smart service
systems: insights from the case of seaport security in Ghana. Int J Inf Manage 58:102204. https://
doi. org/ 10. 1016/j. ijinf omgt. 2020. 102204
Eider M, Berl A (2020) Requirements for Prescriptive Recommender Systems Extending the Lifetime of
EV Batteries. In: 2020 10th International Conference on Advanced Computer Information Tech-
nologies (ACIT). IEEE, Deggendorf, Germany, pp 412–417
Elmachtoub AN, Grigas P (2022) Smart “predict, then optimize.” Manage Sci 68(1):9–26. https:// doi. org/
10. 1287/ mnsc. 2020. 3922
Fang X, Gao Y, Jen-Hwa HuP (2021) A prescriptive analytics method for cost reduction in clinical deci-
sion making. MIS Q 45(1):83–115. https:// doi. org/ 10. 25300/ MISQ/ 2021/ 14372
Feuerriegel S, Hartmann J, Janiesch C, Zschech P (2024) Generative AI. Bus Inf Syst Eng 66(1):111–
126. https:// doi. org/ 10. 1007/ s12599- 023- 00834-7
Flamand T, Ghoniem A, Haouari M, Maddah B (2018) Integrated assortment planning and store-wide
shelf space allocation: an optimization-based approach. Omega 81:134–149. https:// doi. org/ 10.
1016/j. omega. 2017. 10. 006
Fox H, Pillai AC, Friedrich D, Collu M, Dawood T, Johanning L (2022) A review of predictive and
prescriptive offshore wind farm operation and maintenance. Energies 15(2):504. https:// doi. org/ 10.
3390/ en150 20504
Frazzetto D, Nielsen TD, Pedersen TB, Šikšnys L (2019) Prescriptive analytics: a survey of emerging
trends and technologies. VLDB J 28(4):575–595. https:// doi. org/ 10. 1007/ s00778- 019- 00539-y
Galli L, Levato T, Schoen F, Tigli L (2021) Prescriptive analytics for inventory management in health
care. J Oper Res Soc 72(10):2211–2224. https:// doi. org/ 10. 1080/ 01605 682. 2020. 17761 67
Gibson JJ (1986) The Ecological Approach to Visual Perception. Lawrence Erlbaum Asso, Hillsdale, NJ
Giesecke K, Liberali G, Nazerzadeh H, Shanthikumar JG, Teo CP (2022) Introduction to the special sec-
tion on data-driven prescriptive analytics. Manage Sci 68(3):1591–1594. https:// doi. org/ 10. 1287/
mnsc. 2021. 4296
Gordon CAK, Burnak B, Onel M, Pistikopoulos EN (2020) Data-driven prescriptive maintenance: fail-
ure prediction using ensemble support vector classification for optimal process and maintenance
scheduling. Ind Eng Chem Res 59(44):19607–19622. https:// doi. org/ 10. 1021/ acs. iecr. 0c032 41
Goyal A, Aprilia E, Janssen G, Kim Y, Kumar T, Mueller R, Phan D, Raman A, Schuddebeurs J, Xiong J,
Zhang R (2016) Asset health management using predictive and prescriptive analytics for the elec-
tric power grid. IBM J Res Dev 60(1):4:1-4:14
Greene T, Shmueli G, Ray S (2022) Taking the Person Seriously: Ethically-aware IS Research in the Era
of Reinforcement Learning-based Personalization. J Assoc Inf 77:Available at: https:// aisel. aisnet.
org/ jais_ prepr ints/ 77
Gutierrez-Franco E, Mejia-Argueta C, Rabelo L (2021) Data-driven methodology to support long-lasting
logistics and decision making for urban last-mile operations. Sustainability 13(11):6230. https://
doi. org/ 10. 3390/ su131 16230
Harikumar H, Rana S, Gupta S, Nguyen T, Kaimal R, Venkatesh S (2022) Prescriptive analytics with dif-
ferential privacy. Int J Data Sci Anal 13(2):123–138. https:// doi. org/ 10. 1007/ s41060- 021- 00286-w
Hauser M, Flath CM, Thiesse F (2021) Catch me if you scan: data-driven prescriptive modeling for smart
store environments. Eur J Oper Res 294(3):860–873. https:// doi. org/ 10. 1016/j. ejor. 2020. 12. 047
Herm L-V, Steinbach T, Wanner J, Janiesch C (2022) A nascent design theory for explainable intelligent
systems. Electron Mark 32(4):2185–2205. https:// doi. org/ 10. 1007/ s12525- 022- 00606-3
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
Hinsen S, Hofmann P, Jöhnk J, Urbach N (2022) How Can Organizations Design Purposeful Human-AI
Interactions: A Practical Perspective From Existing Use Cases and Interviews. In: Hawaii Interna-
tional Conference on System Sciences (HICSS)
Holsapple C, Lee-Post A, Pakath R (2014) A unified foundation for business analytics. Decis Support
Syst 64:130–141. https:// doi. org/ 10. 1016/j. dss. 2014. 05. 013
Hu R, Fang F, Pain CC, Navon IM (2019) Rapid spatio-temporal flood prediction and uncertainty quan-
tification using a deep learning method. J Hydrol 575:911–920. https:// doi. org/ 10. 1016/j. jhydr ol.
2019. 05. 087
Islam S, Mouratidis H, Mahmud H (2021) An Automated Tool to Support an Intelligence Learner Man-
agement System Using Learning Analytics and Machine Learning. In: Maglogiannis I, Macintyre
J, Iliadis L (eds) Artificial intelligence applications and innovations. Springer International Pub-
lishing, Cham, pp 494–504
Ito S, Fujimaki R (2017) Optimization Beyond Prediction: Prescriptive Price Optimization. In: Proceed-
ings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Min-
ing. ACM, Halifax NS Canada, pp 1833–1841
Janiesch C, Zschech P, Heinrich K (2021) Machine learning and deep learning. Electron Mark 31(3):685–
695. https:// doi. org/ 10. 1007/ s12525- 021- 00475-2
Jank M-H, Dölle C, Schuh G (2019) Product Portfolio Design Using Prescriptive Analytics. In: Schmitt
R, Schuh G (eds) Advances in production research. Springer International Publishing, Cham, pp
584–593
Jensen MH, Persson JS, Nielsen PA (2023) Measuring benefits from big data analytics projects:
an action research study. Inf Syst E-Bus Manage 21(2):323–352. https:// doi. org/ 10. 1007/
s10257- 022- 00620-0
Jeong D-H, Joo H-S (2019) Topical prescriptive analytics system for automatic recommendation of
convergence technology. Biotechnol Bioprocess Eng 24(6):893–906. https:// doi. org/ 10. 1007/
s12257- 019- 0305-1
Jin Y, Qin SJ, Huang Q (2016) Prescriptive analytics for understanding of out-of-plane deformation in
additive manufacturing. In: 2016 IEEE International Conference on Automation Science and Engi-
neering (CASE). IEEE, Fort Worth, TX, USA, pp 786–791
Jozefowiez N, Semet F, Talbi E-G (2008) Multi-objective vehicle routing problems. Eur J Oper Res
189(2):293–309. https:// doi. org/ 10. 1016/j. ejor. 2007. 05. 055
Kast FE, Rosenzweig JE (1972) General system theory: applications for organization and management.
Acad Manage J 15(4):447–465. https:// doi. org/ 10. 2307/ 255141
Kawas B, Squillante MS, Subramanian D, Varshney KR (2013) Prescriptive Analytics for Allocating
Sales Teams to Opportunities. In: IEEE 13th International Conference on Data Mining Workshops.
IEEE, pp 211–218
Kiaghadi M, Hoseinpour P (2023) University admission process: a prescriptive analytics approach. Artif
Intell Rev 56(1):233–256. https:// doi. org/ 10. 1007/ s10462- 022- 10171-y
Kim J-S, Jin H, Züfle A (2020) Expert-in-the-Loop Prescriptive Analytics using Mobility Intervention for
Epidemics. In: International Workshop on Prescriptive Analytics for the Physical World
Knabke T, Olbrich S (2018) Building novel capabilities to enable business intelligence agility: results
from a quantitative study. Inf Syst E-Bus Manage 16(3):493–546. https:// doi. org/ 10. 1007/
s10257- 017- 0361-z
Konstantakopoulos GD, Gayialis SP, Kechagias EP (2022) Vehicle routing problem and related algo-
rithms for logistics distribution: a literature review and classification. Oper Res Int J 22(3):2033–
2062. https:// doi. org/ 10. 1007/ s12351- 020- 00600-7
Kraus M, Tschernutter D, Weinzierl S, Zschech P (2023) Interpretable generalized additive neural net-
works. Eur J Oper Res. https:// doi. org/ 10. 1016/j. ejor. 2023. 06. 032
Krumeich J, Werth D, Loos P (2016) Prescriptive control of business processes: new potentials through
predictive analytics of big data in the process manufacturing industry. Bus Inf Syst Eng 58(4):261–
280. https:// doi. org/ 10. 1007/ s12599- 015- 0412-2
Kubrak K, Milani F, Nolte A, Dumas M (2022) Prescriptive process monitoring: Quo vadis? PeerJ Com-
put Sci 8:e1097. https:// doi. org/ 10. 7717/ peerj- cs. 1097
Kumari M, Kulkarni MS (2022) Developing a prescriptive decision support system for shop floor control.
Ind Manage Data Syst 122(8):1853–1881. https:// doi. org/ 10. 1108/ IMDS- 09- 2021- 0584
Käki A, Kemppainen K, Liesiö J (2019) What to do when decision-makers deviate from model recom-
mendations? Empirical evidence from hydropower industry. Eur J Oper Res 278(3):869–882.
https:// doi. org/ 10. 1016/j. ejor. 2019. 04. 021
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
Lash MT, Zhao K (2016) Early predictions of movie success: the who, what, and when of profitability. J
Manag Inf Syst 33(3):874–903. https:// doi. org/ 10. 1080/ 07421 222. 2016. 12439 69
Lavy S, Garcia JA, Scinto P, Dixit MK (2014) Key performance indicators for facility performance
assessment: simulation of core indicators. Constr Manag Econ 32(12):1183–1204. https:// doi. org/
10. 1080/ 01446 193. 2014. 970208
Leidner DE, Gonzalez E, Koch H (2018) An affordance perspective of enterprise social media and organ-
izational socialization. J Strateg Inf Syst 27(2):117–138. https:// doi. org/ 10. 1016/j. jsis. 2018. 03. 003
Lepenioti K, Bousdekis A, Apostolou D, Mentzas G (2020) Prescriptive analytics: literature review and
research challenges. Int J Inf Manage 50:57–70. https:// doi. org/ 10. 1016/j. ijinf omgt. 2019. 04. 003
Lepenioti K, Bousdekis A, Apostolou D, Mentzas G (2021) Human-augmented prescriptive analytics
with interactive multi-objective reinforcement learning. IEEE Access 9:100677–100693. https://
doi. org/ 10. 1109/ ACCESS. 2021. 30966 62
Levasseur RE (2015) People skills: building analytics decision models that managers use: a change man-
agement perspective. Interfaces 45(4):363–364. https:// doi. org/ 10. 1287/ inte. 2015. 0798
Leyer M, Oberländer A, Dootson P, Kowalkiewicz M (2020) Patterns of decision-making processes with
AI involved Decision-making with artificial intelligence: Towards a novel conceptualization of pat-
terns. In: Pacific Asia Conference on Information Systems (PACIS)
Li X, Zhang L, Xiao T, Zhang S, Chen C (2019) Learning failure modes of soil slopes using monitoring
data. Probab Eng Mech 56:50–57. https:// doi. org/ 10. 1016/j. probe ngmech. 2019. 04. 002
Li X, Zhang W, Zhao X, Pu W, Chen P, Liu F (2021) Wartime industrial logistics information integration:
framework and application in optimizing deployment and formation of military logistics platforms.
J Ind Inf Integr 22:100201. https:// doi. org/ 10. 1016/j. jii. 2021. 100201
Liu B, Lin J, Zhang L, Kumar U (2019) A dynamic prescriptive maintenance model considering sys-
tem aging and degradation. IEEE Access 7:94931–94943. https:// doi. org/ 10. 1109/ ACCESS. 2019.
29285 87
Lo VSY, Pachamanova DA (2015) From predictive uplift modeling to prescriptive uplift analytics: a
practical approach to treatment optimization while accounting for estimation risk. J Market Anal
3(2):79–95. https:// doi. org/ 10. 1057/ jma. 2015.5
Longo F, Padovano A, De Felice F, Petrillo A, Elbasheer M (2023) From “prepare for the unknown”
to “train for what’s coming”: A digital twin-driven and cognitive training approach for the work-
force of the future in smart factories. J Ind Inf Integr 32:100437. https:// doi. org/ 10. 1016/j. jii. 2023.
100437
Majchrzak A, Markus ML (2013) Technology Affordances and Constraints in Management Information
Systems (MIS). In: Encyclopedia of Management Theory. Sage Publications
Markus ML, Silver M (2008) A foundation for the study of IT effects: a new look at Desanctis and
Poole’s concepts of structural features and spirit. J Assoc Inf 9(10):609–632. https:// doi. org/ 10.
17705/ 1jais. 00176
Mazon-Olivo B, Hernández-Rojas D, Maza-Salinas J, Pan A (2018) Rules engine and complex event
processor in the context of internet of things for precision agriculture. Comput Electron Agric
154:347–360. https:// doi. org/ 10. 1016/j. compag. 2018. 09. 013
McFowland E III, Gangarapu S, Bapna R (2021) A prescriptive analytics framework for optimal policy
deployment using heterogeneous treatment effects. MIS Q 45(4):1807–1832
Mehdiyev N, Fettke P (2020) PRESCRIPTIVE PROCESS ANALYTICS WITH DEEP LEARNING
AND EXPLAINABLE ARTIFICIAL INTELLIGENCE. In: Proceedings of the 28th European
Conference on Information Systems (ECIS)
Meske C, Bunde E, Schneider J, Gersch M (2022) Explainable artificial intelligence: objectives, stake-
holders, and future research opportunities. Inf Syst Manag 39(1):53–63. https:// doi. org/ 10. 1080/
10580 530. 2020. 18494 65
Mettler T, Sprenger M, Winter R (2017) Service robots in hospitals: new perspectives on niche evo-
lution and technology affordances. Eur J Inf Syst 26(5):451–468. https:// doi. org/ 10. 1057/
s41303- 017- 0046-1
Miikkulainen R, Francon O, Meyerson E, Qiu X, Sargent D, Canzani E, Hodjat B (2021) From prediction
to prescription: evolutionary optimization of nonpharmaceutical interventions in the COVID-19
pandemic. IEEE Trans Evol Computat 25(2):386–401. https:// doi. org/ 10. 1109/ TEVC. 2021. 30632
17
Mikalef P, Pappas IO, Krogstie J, Giannakos M (2018) Big data analytics capabilities: a systematic litera-
ture review and research agenda. Inf Syst E-Bus Manage 16(3):547–578. https:// doi. org/ 10. 1007/
s10257- 017- 0362-y
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
Mikalef P, Pappas IO, Krogstie J, Pavlou PA (2020) Big data and business analytics: a research agenda
for realizing business value. Inf Manag 57(1):103237. https:// doi. org/ 10. 1016/j. im. 2019. 103237
Moore JF, Carvalho A, Davis GA, Abulhassan Y, Megahed FM (2021) Seat assignments with physical
distancing in single-destination public transit settings. IEEE Access 9:42985–42993. https:// doi.
org/ 10. 1109/ ACCESS. 2021. 30652 98
Mortenson MJ, Doherty NF, Robinson S (2015) Operational research from taylorism to terabytes: a
research agenda for the analytics age. Eur J Oper Res 241(3):583–595. https:// doi. org/ 10. 1016/j.
ejor. 2014. 08. 029
Nadeem A, Marjanovic O, Abedin B (2022) Gender bias in AI-based decision-making systems: a sys-
tematic literature review. AJIS. https:// doi. org/ 10. 3127/ ajis. v26i0. 3835
Nakousi C, Pascual R, Anani A, Kristjanpoller F, Lillo P (2018) An asset-management oriented meth-
odology for mine haul-fleet usage scheduling. Reliab Eng Syst Saf 180:336–344. https:// doi. org/
10. 1016/j. ress. 2018. 07. 034
Niehaus F, Wiesche M (2021) A Socio-Technical Perspective on Organizational Interaction with AI:
A Literature Review. In: European Conference on Information Systems (ECIS)
Notz PM (2020) Explainable Subgradient Tree Boosting for Prescriptive Analytics in Operations
Management. SSRN Journal https:// ssrn. com/ abstr act= 35676 65. https:// doi. org/ 10. 2139/ ssrn.
35676 65
Oesterreich TD, Anton E, Teuteberg F, Dwivedi YK (2022) The role of the social and technical fac-
tors in creating business value from big data analytics: a meta-analysis. J Bus Res 153:128–149.
https:// doi. org/ 10. 1016/j. jbusr es. 2022. 08. 028
Orlikowski WJ, Iacono CS (2001) Research commentary: desperately seeking the “IT” in IT research:
a call to theorizing the IT artifact. Inf Syst Res 12(2):121–134. https:// doi. org/ 10. 1287/ isre.
12.2. 121. 9700
Pappas IO, Mikalef P, Giannakos MN, Krogstie J, Lekakos G (2018) Big data and business analytics
ecosystems: paving the way towards digital transformation and sustainable societies. Inf Syst
E-Bus Manage 16(3):479–491. https:// doi. org/ 10. 1007/ s10257- 018- 0377-z
Paré G, Trudel M-C, Jaana M, Kitsiou S (2015) Synthesizing information systems knowledge: a typol-
ogy of literature reviews. Inf Manag 52(2):183–199. https:// doi. org/ 10. 1016/j. im. 2014. 08. 008
Peng C, Van Doorn J, Eggers F, Wieringa JE (2022) The effect of required warmth on consumer
acceptance of artificial intelligence in service: the moderating role of AI-human collaboration.
Int J Inf Manage 66:102533. https:// doi. org/ 10. 1016/j. ijinf omgt. 2022. 102533
Pereira FD, Fonseca SC, Oliveira EHT, Cristea AI, Bellhauser H, Rodrigues L, Oliveira DBF, Isotani
S, Carvalho LSG (2021) Explaining individual and collective programming students’ behavior
by interpreting a black-box predictive model. IEEE Access 9:117097–117119. https:// doi. org/
10. 1109/ ACCESS. 2021. 31059 56
Perugini D, Perugini M (2014) Characterised and personalised predictive-prescriptive analytics using
agent-based simulation. IJDATS 6(3):209. https:// doi. org/ 10. 1504/ IJDATS. 2014. 063059
Pessach D, Singer G, Avrahami D, Chalutz Ben-Gal H, Shmueli E, Ben-Gal I (2020) Employees
recruitment: a prescriptive analytics approach via machine learning and mathematical program-
ming. Decis Support Syst 134:113290. https:// doi. org/ 10. 1016/j. dss. 2020. 113290
Peters M, Ketter W, Saar-Tsechansky M, Collins J (2013) A reinforcement learning approach to auton-
omous decision-making in smart electricity markets. Mach Learn 92(1):5–39. https:// doi. org/
10. 1007/ s10994- 013- 5340-0
Poornima S, Pushpalatha M (2020) A survey on various applications of prescriptive analytics. Int J
Intell Syst 1:76–84. https:// doi. org/ 10. 1016/j. ijin. 2020. 07. 001
Pospieszny P (2017) Software estimation: towards prescriptive analytics. In: Proceedings of the 27th
International Workshop on Software Measurement and 12th International Conference on Soft-
ware Process and Product Measurement. ACM, Gothenburg Sweden, pp 221–226
Pozzi G, Pigni F, Vitari C (2014) Affordance Theory in the IS Discipline: a Review and Synthesis of
the Literature. In: Americas Conference on Information Systems (AMCIS)
Raeesi Vanani I, Majidian S (2021) Prescriptive Analytics in Internet of Things with Concentration on
Deep Learning. In: García Márquez FP, Lev B (eds) Introduction to internet of things in man-
agement science and operations research. Springer International Publishing, Cham, pp 31–54
Raychaudhuri SJ, Manjunath S, Srinivasan CP, Swathi N, Sushma S, Nitin Bhushan KN, Narendra
Babu C (2021) Prescriptive analytics for impulsive behaviour prevention using real-time biom-
etrics. Prog Artif Intell 10(2):99–112. https:// doi. org/ 10. 1007/ s13748- 020- 00229-9
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
Ren C, Dong J, Ding H, Wang W (2006) Linking Strategic Objectives to Operations: Towards a More
Effective Supply Chain Decision Making. In: Proceedings of the 2006 Winter Simulation Con-
ference. pp 1422–1430
Revathy P, Mukesh R (2020) HadoopSec 2.0: Prescriptive analytics-based multi-model sensitivity-
aware constraints centric block placement strategy for Hadoop. IFS 39(6):8477–8486
Rider NL, Cahill G, Motazedi T, Wei L, Kurian A, Noroski LM, Seeborg FO, Chinn IK, Roberts
K (2021) PI prob: a risk prediction and clinical guidance system for evaluating patients with
recurrent infections. PLoS ONE 16(2):e0237285. https:// doi. org/ 10. 1371/ journ al. pone. 02372 85
Rizzo SG, Chen Y, Pang L, Lucas J, Kaoudi Z, Quiane J, Chawla S (2020) Prescriptive Learning for
Air-Cargo Revenue Management. In: IEEE International Conference on Data Mining (ICDM). pp
462–471
Rzepka C, Berger B (2018) User Interaction with AI-enabled Systems: A Systematic Review of IS
Research. In: International Conference on Information Systems (ICIS)
Salvio KBV, Palaoag TD (2019) Evaluation of the Selected Philippine E-Government Websites’ Perfor-
mance with Prescriptive Analysis. In: Proceedings of the 2019 5th International Conference on
Computing and Artificial Intelligence - ICCAI ‘19. ACM Press, Bali, Indonesia, pp 129–137
Santos LR, Rosati AG (2015) the evolutionary roots of human decision making. Annu Rev Psychol
66(1):321–347. https:// doi. org/ 10. 1146/ annur ev- psych- 010814- 015310
Sarker S, Chatterjee S, Xiao X, Elbanna A (2019) The sociotechnical axis of cohesion for the is disci-
pline: its historical legacy and its continued relevance. MIS Q 43(3):695–719. https:// doi. org/ 10.
25300/ MISQ/ 2019/ 13747
Schneider J, Meske C, Kuss P (2024) Foundation models: a new paradigm for artificial intelligence. Bus
Inf Syst Eng 66(2):221–231. https:// doi. org/ 10. 1007/ s12599- 024- 00851-0
Schoenfeld AH (2010) How we think. Routledge
Schoormann T, Strobel G, Möller F, Petrik D, Zschech P (2023) Artificial intelligence for sustainability:
a systematic review of information systems literature. CAIS 52:199–237. https:// doi. org/ 10. 17705/
1CAIS. 05209
Sedighi Maman Z, Chen Y-J, Baghdadi A, Lombardo S, Cavuoto LA, Megahed FM (2020) A data
analytic framework for physical fatigue management using wearable sensors. Expert Syst Appl
155:113405. https:// doi. org/ 10. 1016/j. eswa. 2020. 113405
Shiau W-L, Chen H, Wang Z, Dwivedi YK (2023) Exploring core knowledge in business intelligence
research. Internet Res 33(3):1179–1201. https:// doi. org/ 10. 1108/ INTR- 04- 2021- 0231
Shollo A, Hopf K, Thiess T, Müller O (2022) Shifting ML value creation mechanisms: a process model
of ML value creation. J Strateg Inf Syst 31(3):101734. https:// doi. org/ 10. 1016/j. jsis. 2022. 101734
Silva AJ, Cortez P (2022) An Industry 4.0 Intelligent Decision Support System for Analytical Laborato-
ries. In: Maglogiannis I, Iliadis L, Macintyre J, Cortez P (eds) Artificial intelligence applications
and innovations. Springer International Publishing, Cham, pp 159–169
Simon HA (1960) The new science of management decision. Harper & Brothers, New York
Slovic P, Fischhoff B, Lichtenstein S (1977) Behavioral decision theory. Annu Rev Psychol 28(1):1–39.
https:// doi. org/ 10. 1146/ annur ev. ps. 28. 020177. 000245
Soeffker N, Ulmer MW, Mattfeld DC (2022) Stochastic dynamic vehicle routing in the light of prescrip-
tive analytics: a review. Eur J Oper Res 298(3):801–820. https:// doi. org/ 10. 1016/j. ejor. 2021. 07. 014
Song S, Jeong D-H, Kim J, Hwang M, Gim J, Jung H (2014) Research Advising System Based on Pre-
scriptive Analytics. In: Park JJ, Pan Y, Kim C-S, Yang Y (eds) Future information technology.
Springer, Berlin Heidelberg, Berlin, Heidelberg, pp 569–574
Soroush M, Masooleh LS, Seider WD, Oktem U, Arbogast JE (2020) Model-predictive safety optimal
actions to detect and handle process operation hazards. AIChE J. https:// doi. org/ 10. 1002/ aic. 16932
Srinivas S, Ravindran AR (2018) Optimizing outpatient appointment system using machine learning
algorithms and scheduling rules: a prescriptive analytics framework. Expert Syst Appl 102:245–
261. https:// doi. org/ 10. 1016/j. eswa. 2018. 02. 022
Stefani K, Zschech P (2018) Constituent Elements for Prescriptive Analytics Systems. In: Twenty-Sixth
European Conference on Information Systems
Stein N, Meller J, Flath CM (2018) Big data on the shop-floor: sensor-based decision-support for manual
processes. J Bus Econ 88(5):593–616. https:// doi. org/ 10. 1007/ s11573- 017- 0890-4
Straub D, Welpe I (2014) Decision-Making Under Risk: A Normative and Behavioral Perspective. In:
Klüppelberg C, Straub D, Welpe IM (eds) risk: a multidisciplinary introduction. Springer Interna-
tional Publishing, Cham, pp 63–93
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
C.Wissuchek, P.Zschech
Strong D, Volkoff O, Johnson S, Pelletier L, Tulu B, Bar-On I, Trudel J, Garber L (2014) A theory of
organization-EHR affordance actualization. J Assoc Inf 15(2):53–85. https:// doi. org/ 10. 17705/
1jais. 00353
Sturm T, Gerlacha J, Pumplun L, Mesbah N, Peters F, Tauchert C, Nan N, Buxmann P (2021) Coordi-
nating human and machine learning for effective organization learning. MIS Q 45(3):1581–1602.
https:// doi. org/ 10. 25300/ MISQ/ 2021/ 16543
Suvarna M, Jahirul MI, Aaron-Yeap WH, Augustine CV, Umesh A, Rasul MG, Günay ME, Yildirim R,
Janaun J (2022) Predicting biodiesel properties and its optimal fatty acid profile via explainable
machine learning. Renew Energ 189:245–258. https:// doi. org/ 10. 1016/j. renene. 2022. 02. 124
Svenson O (1992) Differentiation and consolidation theory of human decision making: a frame of refer-
ence for the study of pre-and post-decision processes. Acta Psychol 80(1–3):143–168. https:// doi.
org/ 10. 1016/ 0001- 6918(92) 90044-E
Swaminathan JM (2018) Big data analytics for rapid, impactful, sustained, and efficient (RISE) humani-
tarian operations. Prod Oper Manag 27(9):1696–1700. https:// doi. org/ 10. 1111/ poms. 12840
Tamimi N, Samani S, Minaei M, Harirchi F (2019) An Artificial Intelligence Decision Support Sys-
tem for Unconventional Field Development Design. In: Proceedings of the 7th Unconventional
Resources Technology Conference
Tektaş B, Turan HH, Kasap N, Çebi F, Delen D (2022) A fuzzy prescriptive analytics approach to power
generation capacity planning. Energies 15(9):3176. https:// doi. org/ 10. 3390/ en150 93176
Thapa D, Sein MK (2018) Trajectory of affordances: insights from a case of telemedicine in Nepal. Info
Systems J 28(5):796–817. https:// doi. org/ 10. 1111/ isj. 12160
Trunk A, Birkel H, Hartmann E (2020) On the current state of combining human and artificial intel-
ligence for strategic organizational decision making. Bus Res 13(3):875–919. https:// doi. org/ 10.
1007/ s40685- 020- 00133-x
Uskov VL, Bakken JP, Shah A, Hancher N, McPartlin C, Gayke K (2019) Innovative InterLabs System
for Smart Learning Analytics in Engineering Education. 2019 IEEE global engineering education
conference (EDUCON). IEEE, Dubai, United Arab Emirates, pp 1363–1369
Vanderschueren T, Verdonck T, Baesens B, Verbeke W (2022) Predict-then-optimize or predict-and-opti-
mize? an empirical evaluation of cost-sensitive learning strategies. Inf Sci 594:400–415. https://
doi. org/ 10. 1016/j. ins. 2022. 02. 021
Vargas AP, Díaz D, Jaramillo S, Rangel F, Villa D, Villegas JG (2022) Improving the tactical planning of
solid waste collection with prescriptive analytics: a case study. Prod 32:e20210037. https:// doi. org/
10. 1590/ 0103- 6513. 20210 037
Vater J, Harscheidt L, Knoll A (2019) A Reference Architecture Based on Edge and Cloud Computing
for Smart Manufacturing. In: 28th International Conference on Computer Communication and Net-
works (ICCCN)
Vater J, Schlaak P, Knoll A (2020) A Modular Edge-/Cloud-Solution for Automated Error Detection of
Industrial Hairpin Weldings using Convolutional Neural Networks. In: IEEE 44th Annual Comput-
ers, Software, and Applications Conference (COMPSAC). pp 505–510
Verbraeken J, Wolting M, Katzy J, Kloppenburg J, Verbelen T, Rellermeyer JS (2021) A survey on dis-
tributed machine learning. ACM Comput Surv 53(2):1–33. https:// doi. org/ 10. 1145/ 33774 54
Vieira AAC, Dias LMS, Santos MY, Pereira GAB, Oliveira JA (2020) Supply chain data integration: a
literature review. J Ind Inf Integr 19:100161. https:// doi. org/ 10. 1016/j. jii. 2020. 100161
Vimala Rani M, Mathirajan M (2021) Prescriptive Analytics for Dynamic Real Time Scheduling of Dif-
fusion Furnaces. In: Srinivas S, Rajendran S, Ziegler H (eds) Supply Chain Management in manu-
facturing and service systems. Springer International Publishing, Cham, pp 241–278
Wang C-H, Cheng H-Y, Deng Y-T (2018) Using bayesian belief network and time-series model to con-
duct prescriptive and predictive analytics for computer industries. Comput Ind Eng 115:486–494.
https:// doi. org/ 10. 1016/j. cie. 2017. 12. 003
Wang N, Ho K, Pavlou G, Howarth M (2008) An overview of routing optimization for internet traffic
engineering. IEEE Commun Surv Tutorials 10(1):36–56. https:// doi. org/ 10. 1109/ COMST. 2008.
44836 69
Wanner J, Herm L-V, Heinrich K, Janiesch C (2022) The effect of transparency and trust on intelligent
system acceptance: evidence from a user-based study. Electron Mark 32(4):2079–2102. https:// doi.
org/ 10. 1007/ s12525- 022- 00593-5
Wanner J, Wissuchek C, Welsch G, Janiesch C (2023) A taxonomy and archetypes of business analyt-
ics in smart manufacturing. Data Base Adv Inf Syst 54(1):11–45. https:// doi. org/ 10. 1145/ 35835 81.
35835 84
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
Prescriptive analytics systems revised: asystematic…
Webster J, Watson RT (2002) Analyzing the Past to Prepare for the Future: Writing a Literature Review.
MIS Q 26(2):xiii–xxiii
Xylia M, Ibrahim O, Silveira S (2016) Fossil-free public transport: Prescriptive policy analysis for the
Swedish bus fleets. 2016 13th International Conference on the European Energy Market (EEM).
IEEE, Porto, Portugal, pp 1–5
Yang H, Duque D, Morton DP (2022) Optimizing diesel fuel supply chain operations to mitigate power
outages for hurricane relief. IISE Transactions 54(10):936–949. https:// doi. org/ 10. 1080/ 24725 854.
2021. 20214 61
Yanta S, Thammaboosadee S, Chanyagorn P, Chuckpaiwong R (2021) Probation Status Prediction and
Optimization for Undergraduate Engineering Students. 2021 13th International Conference on
Knowledge and Smart Technology (KST). IEEE, Bangsaen, Chonburi, Thailand, pp 191–196
Zadorojniy A, Wasserkrug S, Zeltyn S, Lipets V (2019) Unleashing analytics to reduce costs and improve
quality in wastewater treatment. INFORMS J on Appl Anal 49(4):262–268. https:// doi. org/ 10.
1287/ inte. 2019. 0990
Zhang P, Zhu X, Xie M (2021) A model-based reinforcement learning approach for maintenance optimi-
zation of degrading systems in a large state space. Comput Ind Eng 161:107622. https:// doi. org/ 10.
1016/j. cie. 2021. 107622
Zhang B, Ong YJ, Nakamura T (2022) SimPO: Simultaneous Prediction and Optimization. In: 2022
IEEE International Conference on Services Computing (SCC). IEEE, pp 120–122
Zheng H, Ryzhov IO, Xie W, Zhong J (2021) Personalized multimorbidity management for patients with
type 2 diabetes using reinforcement learning of electronic health records. Drugs 81(4):471–482.
https:// doi. org/ 10. 1007/ s40265- 020- 01435-4
Zschech P, Weinzierl S, Hambauer N, Zilker S, Kraus M (2022) GAM(e) changer or not? An evaluation
of interpretable machine learning models based on additive model constraints. In: Proceedings of
the 30th European Conference on Information Systems (ECIS)
vom Brocke J, Simons A, Niehaves B, Riemer K, Plattfat R, Cleven A (2009) RECONSTRUCTING
THE GIANT: ON THE IMPORTANCE OF RIGOUR IN DOCUMENTING THE LITERATURE
SEARCH PROCESS. In: European Conference on Information Systems (ECIS)
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps
and institutional affiliations.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center
GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers
and authorised users (“Users”), for small-scale personal, non-commercial use provided that all
copyright, trade and service marks and other proprietary notices are maintained. By accessing,
sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of
use (“Terms”). For these purposes, Springer Nature considers academic use (by researchers and
students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and
conditions, a relevant site licence or a personal subscription. These Terms will prevail over any
conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription (to
the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of
the Creative Commons license used will apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may
also use these personal data internally within ResearchGate and Springer Nature and as agreed share
it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not otherwise
disclose your personal data outside the ResearchGate or the Springer Nature group of companies
unless we have your permission as detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial
use, it is important to note that Users may not:
use such content for the purpose of providing other users with access on a regular or large scale
basis or as a means to circumvent access control;
use such content where to do so would be considered a criminal or statutory offence in any
jurisdiction, or gives rise to civil liability, or is otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association
unless explicitly agreed to by Springer Nature in writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a
systematic database of Springer Nature journal content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a
product or service that creates revenue, royalties, rent or income from our content or its inclusion as
part of a paid for service or for other commercial gain. Springer Nature journal content cannot be
used for inter-library loans and librarians may not upload Springer Nature journal content on a large
scale into their, or any other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not
obligated to publish any information or content on this website and may remove it or features or
functionality at our sole discretion, at any time with or without notice. Springer Nature may revoke
this licence to you at any time and remove access to any copies of the Springer Nature journal content
which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or
guarantees to Users, either express or implied with respect to the Springer nature journal content and
all parties disclaim and waive any implied warranties or warranties imposed by law, including
merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published
by Springer Nature that may be licensed from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a
regular basis or in any other manner not expressly permitted by these Terms, please contact Springer
Nature at
onlineservice@springernature.com
