Development of Smart Semiconductor Manufacturing: Operations Research and Data Science Perspectives

Abstract

With advances in information and telecommunication technologies and data-enabled decision making, smart manufacturing can be an essential component of sustainable development. In the era of the smart world, semiconductor industry is one of the few global industries that are in a growth mode to smartness, due to worldwide demand. The important opportunities that can boost the cost reduction of productivity and improve quality in wafer fabrication are based on the simulations of actual environment in Cyber-Physical Space and integrate them with decentralized decision-making systems. However, this integration faced the industry with novel unique challenges. The stream of the data from sensors, robots, and Cyber-Physical Space can aid to make the manufacturing smart. Therefore, it would be an increased need for modeling, optimization, and simulation for the value delivery from manufacturing data. This paper aims to review the success story of smart manufacturing in semiconductor industry with the focus on data-enabled decision making and optimization applications based on operations research and data science perspective. In addition, we will discuss future research directions and new challenges for this industry.

Full text

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2933167, IEEE Access
Date of publication xxxx 00, 0000, date of current version xxxx 00, 0000.
Digital Object Identifier 10.1109/ACCESS.2019.DOI
Development of Smart Semiconductor
Manufacturing: Operations Research
and Data Science Perspectives
MARZIEH KHAKIFIROOZ1, (Member, IEEE), MAHDI FATHI2, (Member, IEEE) and KAN WU 3,
(Member, IEEE)
1School of Science and Engineering, Tecnològico de Monterrey, Mèxico (mkhakifirooz@tec.mx)
2Department of Industrial and Systems Engineering, Mississippi State University, Starkville, MS, US (e-mail: fathi@ise.msstate.edu)
3Division of Systems and Engineering Management, Nanyang Technological University, Singapore (e-mail: wukan@ntu.edu.sg)
Corresponding author: Marzieh Khakifirooz (e-mail: mkhakifirooz@tec.mx).
ABSTRACT With advances in information and telecommunication technologies and data-enabled decision
making, smart manufacturing can be an essential component of sustainable development. In the era of
the smart world, semiconductor industry is one of the few global industries that are in a growth mode
to smartness, due to worldwide demand. The important opportunities that can boost the cost reduction of
productivity and improve quality in wafer fabrication are based on the simulations of actual environment
in Cyber-Physical Space and integrate them with decentralized decision-making systems. However, this
integration faced the industry with novel unique challenges. The stream of the data from sensors, robots, and
Cyber-Physical Space can aid to make the manufacturing smart. Therefore, it would be an increased need for
modeling, optimization, and simulation for the value delivery from manufacturing data. This paper aims to
review the success story of smart manufacturing in semiconductor industry with the focus on data-enabled
decision making and optimization applications based on operations research and data science perspective.
In addition, we will discuss future research directions and new challenges for this industry.
INDEX TERMS cloud computing, Cyber-Physical Space, data science, Industry 4.0, Internet of Things,
operations research, smart manufacturing, semiconductor industry.
I. INTRODUCTION
The importance of national manufacturing strategies such
as Advanced Manufacturing Partnership and Industry 4.0
have reemphasized the shifting standard of manufacturing
and production system, which led to the fourth industrial
generation.
The industrial revolution stream drives deployment of
novel concepts for smart factories, new generation of mon-
itoring and collaborating systems, or in general words, the
smart manufacturing system. Smart manufacturing system
is built upon the emerging advanced technologies including
Cyber-Physical Space (CPS), Internet of Things (IoT), cloud
and cognitive computing, big data analysis and information
and communication technology [1]. The first step toward
smart manufacturing is connectivity [2]. All the components
in the industry must be connected to a single network which
is being allowed by the CPS and IoT providing information
interchange and connectivity to attain a flexible and self-
adaptive production system.
On the other hands, as a part of the technology road-map
for semiconductors driven by Moore’s law system scaling
[3], there are more and more challenges by the poverty of
resources and emergence of information technology. There-
fore, the seamless interaction of smart manufacturing com-
ponents such as big-data, instant data, information tech-
nology (cloud, and multi-mode sensors), high-performance
computing, mobile computing, and autonomous sensing and
computing is necessary for driving “More Moore" (MM)
technologies [4].
The paradigm of smart manufacturing and the semicon-
ductor industry is a back-end loop design. Consider technolo-
gies enabling smart manufacturing can emerge the sensor
technology, network communication, advanced in data analy-
sis, and advanced in software and system as requirements for
industrial development. Any evolution in the aforementioned
components of smart manufacturing could affect directly on
VOLUME 00, 2019 1

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2933167, IEEE Access
Khakifirooz et al.: Smart Semiconductor Manufacturing: OR&DS Perspectives
FIGURE 1. Relationship between smart manufacturing and semiconductor industry
performance and quality enhancement, innovation, and smart
production. Thereupon, intelligent semiconductor devices are
vital solutions to this growth. In a back-end loop, this is the
smart manufacturing technology which helps semiconductor
industry to produce and perform smarter (see Fig. 1).
The operations control of manufacturing facilities of the
semiconductor is known as a tough task and is envisaged
as one of the most composite manufacturing environments.
One solution to deal with these difficulties is to choose the
manufacturing and process data to analyze and modeling
processes to empower factories in order to intensify an
enhanced knowledge of the challenges associated with the
production process and to grow visions which can develop
prevailing procedures. Hereupon, this is very important to
have enough understanding of the prevailing position of
research about decision making based data engineering tech-
nologies in the semiconductor industry and recognize fields
for future research to maintain the further technologies for
wafer manufacturing. Therefore, the contributions of this
study can be summarized as 1) detect gaps in the existing
works, 2) develop significant research ideas, 3) categorize
existing research struggles and form a layout that can deliver
different ideas related to the operations research and data
science (OR&DS) area in smart wafer manufacturing.
To the best of our knowledge, there is no such a com-
prehensive study among the existing literature that has been
covered all the aforementioned contributions of this study.
II. REVIEW METHOD
This paper provides a three-stage qualitative literature review
method (identification, classification, and evaluation) [5] on
the scientific progress of the fourth industrial revolution from
the OR&DS perspective for semiconductor manufacturing.
Most precisely, three research questions are given as follows:
1) Identification: what are the main challenges from the
OR&DS points of view, enabling the industrial revo-
lution in semiconductor manufacturing?
2) Classification: how are the OR&DS addressed the
scientific and technological challenges in smart semi-
conductor manufacturing?
3) Evaluation: what are the managerial suggestions from
the integrated information of reviewed papers to pre-
vail the unseen and future challenges in the path for-
ward to the implementation of smart semiconductor
manufacturing?
The study applied a two-step screening procedure to select
relevant studies. In the first place, the study carefully defined
the scope of the literature review by selecting the studies
which have used terms “semiconductor", “wafer," “integrated
circuit," or “chip" in their title or indexed keywords. The
study used the Scopus database as a search engine. The time-
frame of review is narrowed by the milestone of national
manufacturing strategies since 2011. From the search result,
only literature reported in English and published in decision
science field was included in the review process. In the sec-
ond step, all cited literature are cross-checked using Google
scholar search engine.
The study classified indexed keywords for further investi-
gation. The indexed keywords of each article are classified in
one group. The unrelated words to the OR&DS filed were
removed, and a unique title is selected for all words with
similar meaning. Then the decision support matrix is com-
posed based on the classification result to illustrate the link
among the keywords. Thereafter, the Mutually Exclusive-
Collectively Exhaustive (MECE) method [6] was applied for
feature extraction (select parent methods, and the most com-
patible technique with them). The steps of this classification
procedure are summarized in Fig. 2. After data screening and
key factor extraction, 47 keywords are selected by MECE
method and classified into six families. The classification
result is summarized in Table 1.
Selected literature varied in quality and quantity in differ-
ent fields. To ensure that the search result was reliable, those
studies that their methodology had high similarity with other
studies were eliminated while considering the priority for re-
cently published journal articles. The studies were evaluated
and classified according to several methodological criteria in
order to shortlist the qualify papers for further analysis of
2VOLUME 00, 2019

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2933167, IEEE Access
Khakifirooz et al.: Smart Semiconductor Manufacturing: OR&DS Perspectives
FIGURE 2. Key steps for classification the OR&DS indexed keywords.
TABLE 1. The MCME classification for OR&DS related keywords
Parent Keywords Family Member Keywords
Capacity Planning Demand Forecasting & Delivery; Supply Chains; Customer Satisfaction; Enterprise Resource Planning; Resource Allocation &
Facility Layout
Inventory Management Flexible Manufacturing Systems; Scheduling & Rescheduling; Dispatching; Virtual Reality; Investment & Profitability; Work-
In-Process; Random Process; Product Life Cycle; Cost Management; Maintenance; Cycle Time Reduction; Batch Processing;
Risk Management
Sustainability Assembly; Technology Transfer
Standardization Re-engineering; Productivity; Reliability; Bench-marking; Operation Management; Human Resource Management; Just In
Time Production; Machine Utilization; Product Test; Product Design; Data Reduction; System Diagnosis & Fault Detection;
Quality Management; Environmental Management; Sales, Marketing & Financial Management; Product Line Design
Production Planning Technology Management; Performance Measurement
Decision Support Systems & Decision Theory Yield Management & Enhancement; Control & Monitoring (online/offline); Strategic Planning
their main contribution as follows:
•Organize the type of research methods by Wieringa
et al. [7] (including: validation, evaluation, solution,
philosophical, opinion, experience)
•Classify the areas of manufacturing by Meziane et
al. [8] (including quality management, design, process
and planning, control, environment, health and safety,
maintenance and diagnosis, scheduling, and virtual
manufacturing)
•Categorize the form of contribution by keywording
method [9] (including: architecture, framework, the-
ory, methodology, model, platform, process, tool)
•Classify the type of analytic by Delen et al. [10]
(including: descriptive, predictive, and prescriptive)
III. ROAD-MAP OF OR&DS IN SEMICONDUCTOR
With regards to the information collected from the search
process, this section explores how OR&DS influenced on
semiconductor industry. The role of OR&DS in the smart
semiconductor industry is reviewed by answering some ad-
ditional research questions in this direction.
A. BY GROWING THE SMART MANUFACTURING, HOW
OR&DS RELATED RESEARCH FOUND THEIR WAY INTO
SEMICONDUCTOR MANUFACTURING INTELLIGENCE?
The historical review of the infrastructure of smart semicon-
ductor manufacturing aligns with the fourth industrial revo-
lution shows how decision-making process became mature
in this industry by adapting the OR&DS tools. The summary
shows that:
•Before 2011
Methods such as:
Data mining since the late 90s, AI since the late 80s, heuris-
tic algorithm since the early 90s, Machine Learning since
the late 80s, data development management since the late
80s, Fuzzy logic since the early 90s, optimization methods
such as linear-programming since the early 90s, non linear-
programming since 2000s and convex optimization since the
early 90s, data visualization since the late 90s, game theory
since the late 90s, queuing theory since 90s,
and concepts such as:
Advanced manufacturing since the late 80s, intelligence
manufacturing since the early 90s, Enterprise Resource
Planning (ERP) since the late 90s, Overall Equipment Ef-
ficiency (OEE) since the late 90s, Decision Support System
(DSS) since 90s, virtual manufacturing since the early 90s,
e-manufacturing since 2000s, and agent-based system since
early 2000s
have been appearing in literature to discover the challenges
in the semiconductor industry and moving forward to smart
manufacturing. A summary of important research publica-
tions is presented in Table 2.
•After 2011:
Despite the needs for moving forward the intelligent pro-
duction, the annual gathering and academic reports had a
vital role in leading the semiconductor industry toward the
smartness. Fig. 3 is summarized this progress.
B. WHAT KIND OF STUDIES IS BEING CARRIED OUT IN
THE FIELD OF OR&DS IN SEMICONDUCTOR
MANUFACTURING?
The main objective of this question is to focus on research in
terms of the philosophical point of view along with practical
assessments. The classification result according to the defini-
tion of areas of manufacturing by Meziane et al. is depicted
in Fig. 4. The result ratifies that there is an extensive gap in
VOLUME 00, 2019 3

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2933167, IEEE Access
Khakifirooz et al.: Smart Semiconductor Manufacturing: OR&DS Perspectives
TABLE 2. Summary of most distinguished researches have been done before 2011.
Methods & Concepts References
AI, Data mining & Machine Learning [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21]
Heuristic algorithm [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32],
Fuzzy logic [33], [34], [35], [36]
Optimization [37], [38], [39], [40], [41], [42], [43], [44]
Game theory [45], [46], [44]
Advanced/ Intelligence manufacturing [47], [33], [48] [49]
OEE [50], [51], [52], [53], [54]
fitting the manufacturing design for intelligent layout. The
intelligent layout design for manufacturing generally refers
to system engineering design, sensor allocation problems,
and design the software agent solutions merge with high-tech
computing technology or service-oriented computing. There
is also a lack of investigation on virtual manufacturing, simu-
lation the physical environment, e-manufacturing, and AR. In
addition, trends related to the environmental issues and health
and safety such as green industry and re-manufacturing are
demanding topics for smart manufacturing, which had less
attention in semiconductor industry yet.
To determine the gap of the research for smart IC industry,
we modified the classification by Meziane et al. for semi-
conductor manufacturing context. Fig. 5 illustrates the con-
tributions of each class for the smart semiconductor industry.
The scale of contribution defines such that the most relevant
topic granted with the score of 100. Due to dependency
among process steps in wafer fabrication, challenges are
spread along the production process such that single solution
cannot solve the problem. Therefore, the hybrid models are
a ubiquitous solution in semiconductor-related literature to
deal with an epidemic dimension of problems. The databases
of most common techniques in Fig. 4 are used as the basis
for Fig. 5. Fig. 5 demonstrates how the hybrid method is
associated with each other. Following decisions could be
extracted from Fig. 5:
Capacity planning:
•Enterprise resource planning is designed for increas-
ing or decreasing capacity at the production facilities
as well as planning when and whether to build new
facilities.
•Demand forecasting highly related to customer re-
quirements. The demands are unpredictable and can be
lost if the manufacturer does not have enough capacity
during a period of high demand.
•Capacity planning is a function of the hedge to meet
the needs of the semiconductor capacity supplier and
demander.
•In a semiconductor supply chain, the low demand
variability and the high process flexibility are affected
by capacity planning.
Sustainability:
•Design for sustainability includes full lifecycle con-
cepts, design for assembly (and disassembly), de-
sign for extended life, and design for reuse/re-
manufacture/recycling. [62]
•Performance measurement and technology alignment
are tightly correlated because of dynamic and progres-
sive shifts in deregulated markets.
Standardization:
•Standardization of environmental management sys-
tems is considered as a revolutionary force that will
transform both the ways managers think about envi-
ronmental functions and the relationship between man-
ufacturing and environmental regulators (as evidence
one can refer to ISO 14000 regulations).
•The relationship between standardization and data re-
duction could back to the data standardization for
reducing variable variation, or can indicate to the
influence of process standardization on reducing the
number of unnecessary variables.
•Undoubtedly, similar to the other manufacturing pro-
cess, semiconductor equipment quality can be im-
proved through bench-marking, standardization, and
automation [63].
•Just-in-time production (Kanban) is applicable for in-
dustries with less customization module and high stan-
dardization for producing products in small lots which
are exactly matched with the nature of semiconductor
device fabrications.
•Standardized processes can capture and institutionalize
existing knowledge within organizational routines that
help establish a common frame and working habits
among employees [64].
•Standardization is a remedy for increasingly hetero-
geneous consumer needs, product and process com-
plexities, plus reduction of scale economies. It can
mitigate the effects of process complexity and product
imperfection.
•standardization can enable the customization in prod-
uct design [65].
•Lack of standardization causes system fault and error.
•There is a cyclical nature between standardization
direction and customization direction such that the
future of semiconductor devices is standardization in
manufacturing but customized in the application.
Inventory management:
•According to the little’s law [66] there is a cyclic
relationship between the throughput, inventory (ma-
terial inventory, WIP inventory, and finished product
inventory) and cycle time, such that the high WIP is
required for high throughput with low cycle time.
4VOLUME 00, 2019

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2933167, IEEE Access
Khakifirooz et al.: Smart Semiconductor Manufacturing: OR&DS Perspectives
FIGURE 3. Road-map of the fourth industrial revolution for semiconductor manufacturing in the academic sector [55], [56], [57], [58], [59], [60], [61].
.
FIGURE 4. Class allotment of areas of manufacturing for smart semiconductor industry
•Schedules can be used to manage the inventory re-
quirements, and maintenance, where Flexible Manu-
facturing System consists of scheduling algorithm and
involves the inventory information. A dispatching al-
gorithm decides how to use factory resources upon the
availability of resources. Therefore, manufacturers can
manage the throughput, inventory, and consequently
profitability, risk, and cost, all together.
VOLUME 00, 2019 5

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2933167, IEEE Access
Khakifirooz et al.: Smart Semiconductor Manufacturing: OR&DS Perspectives
FIGURE 5. Contribution of most frequent topics among the literature since 2011 related to smart semiconductor industry based on classification in Table 3.
•Visual simulation and the modeling process based on
the virtual environment can facilitate the optimiza-
tion of workshop layout (i.e., inventory management,
scheduling, batch processing).
Production planning:
•Performance measurement is the key to improving per-
formance and is a prerequisite to improving production
planning.
•The general policy of sustainable development mecha-
nism is neglected essential details of how technology
can be transferred successfully. Though technology
can play an innovative role in improving sustainable
manufacturing.
Decision support system and decision theory:
•Supporting strategic decisions are more common in
research development for semiconductor manufactur-
ing. While the nature of strategic decisions is changing
significantly from a single organization’s strategies to
internal layers of manufacturing.
•knowledge management for supporting process diag-
nosis and decision-making is required to approve by
control and monitoring system as well as data acquisi-
tion.
•Yield management and enhancement normally are sup-
ported by DSS to epitomize the decision rules for
expert engineers.
The classification study for the type of research method
by Wieringa et al. [7] is illustrated in Fig. 6. Fig. 5 shows
how the type of research is branched over topics, and Fig.
6 shows the contribution of each type of research based on
philosophical points of view. For simplicity of comparison,
according to the definition of “experience" in [7], and since
this type of research has seldom happened in OR&DS filed,
we remove the experience from the list. Concluding from Fig.
5 and Fig. 6, digitizing the knowledge-based system has the
lowest contribution among the other research topic in current
statues which is required to have more inspection for advance
development of smart semiconductor industry.
6VOLUME 00, 2019

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2933167, IEEE Access
Khakifirooz et al.: Smart Semiconductor Manufacturing: OR&DS Perspectives
FIGURE 6. Partitioning the contribution of most common topic in smart
semiconductor based on type of research
.
C. WHICH AREAS OF SEMICONDUCTOR
MANUFACTURING ARE OR&DS TECHNIQUES BEING
APPLIED IN?
The objective of this question is to highlight the types of
inputs and outcomes. To categorize the literature according
to the form of their contributions [9] we divided the at-
tributes of contributions into two groups of variability based
on outcomes and results (including architecture, framework,
model, methodology), and variability on input information
(including theory, platform, process, tool). In this category,
the platform indicates the hardware or software components
which enables the applications to execute, and the framework
is the software solution for the problem. The process is the
approach to reach that solution. The theory is the guideline
or road-map for entering to the mathematical model. Sub-
sequently, the tool addresses to the utilities for proposing
the solution, and architecture is components which interact
together to achieve the solution. Fig. 7 illustrates the 2D
plot between each category. The result shows that there
is a vacancy for research on integrating the mathematical
model with software and hardware platforms. The theoretical
approaches for developing the smart semiconductor industry
plus the advantages of using high-tech computing technology
provide more spaces for further investigation.
D. WHAT KIND OF ANALYTICAL ANALYSIS IS BEING
USED IN THE AREA OF OR&DS IN SEMICONDUCTOR
MANUFACTURING?
The objective of this question is to discuss the analytical
approaches of OR&DS in the semiconductor industry. Ac-
cording to Delen et al. [10], the analytical analysis is clas-
sifying to descriptive, predictive, and prescriptive analysis
where the descriptive analysis enables the business reporting,
dashboards, data warehousing, and scorecards. Subsequently,
the predictive analysis facilities data mining, forecasting, text
mining, and Web or media mining and prescriptive analysis
empower the expert systems, decision models, optimization,
and simulation. Although we expect that the application
of descriptive analysis and Web mining or text mining in
semiconductor manufacturing is sporadic, we still considered
all aspects of analytical analysis. The level of interest of
each class of taxonomy presented in Fig. 8. Apparently,
for developing smart semiconductor industry, the descriptive
analysis will be an inevitable tool, basically for visualizing
the production process from the event-driven process.
IV. MANAGEMENT SUGGESTION
Despite challenges mentioned in the preceding sections, in
the following, some managerial suggestions are given for the
development of smart semiconductor manufacturing environ-
ment.
Digitalize knowledge-based DSS
In a smart manufacturing environment, sharing expert do-
main knowledge at the manager-operator and operator-
machine level is essential. Recommender systems and
opinion mining can support real-time, data-based deci-
sion making. Machine/user relationship mining and clus-
tering can increase the self-awareness, self-learning, and
self-maintenance of production systems. Finally, Reciprocal
Learning-Based DSS (RL-DSS) [67] can make repetitive
decisions and reduce the human decision making a load.
Routine decision tasks can be programmed, and learning
algorithms can enhance performance. Then decision-makers
can update their knowledge, and the improved system can
help to create better decisions. Therefore, research opportu-
nities in this domain include:
•Incorporating the behavior of human decision-makers
into solutions.
•Automating decisions made by humans.
•Highlighting the interface of information systems with
humans
Incorporate the dynamicity into the solutions
The dynamic nature of the semiconductor industry requires
a dynamic solution. Dynamic characteristics are inherent
features in all semiconductor devices and transistors. The
dynamic behavior of semiconductor devices refers to the
act when a device is connecting to a regulated source and
rapid changes of voltage and current occur. On the other
hand, to modeling the uncertainties concerning future charac-
teristics of semiconductor technology, a long-term dynamic
solution is required for endogenizing interactions between
decision structure and uncertainties. Therefore to optimize
the integrating the time horizon into one objective functions
and to link different time steps in the model by various
types of constraints, only dynamic constraints can solve the
complexity of the problems. Research propositions include:
•Developing stochastic and dynamic versions of solu-
tions and deterministic models [68].
•Anticipating the stochasticity in the models based
on dynamic programming, robust optimization, and
stochastic programming.
Design software-based solution with user-friendly in-
terface
VOLUME 00, 2019 7

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2933167, IEEE Access
Khakifirooz et al.: Smart Semiconductor Manufacturing: OR&DS Perspectives
FIGURE 7. Class allotment of form of contribution methods applied in semiconductor industry.
FIGURE 8. Type of analytical methods applied in semiconductor industry.
In this era of Industry 4.0, thanks to the integration of sensors
and Edge Computing solutions that allow collection and
access to online data, for customized development and im-
plementations of smart manufacturing, a complete, codeless
programming, and scalable wireless protocol software stack
are required to help companies for real-time monitoring,
predictive maintenance in less downtime, optimized indus-
trial performances, and power conservation. The software
is needed to be designed based on a user-friendly, modular
architecture, and consist of development boards, debugging
tools, and all other standard requirements. Research scopes
include:
•Considering the role of high-tech computing tech-
niques, including cloud computing techniques in
decision-making and parallel computing on Graphics
Processing Units (GPU).
•Knowing the restrictions of current packaged software
for semiconductor management, process, and produc-
tion.
•Proposing alternative software solutions including
service-oriented computing and software agents for
semiconductor planning and scheduling applications.
•Designing domain-specific solutions based on open-
source software. The selection of an appropriate sim-
ulation tool is often crucial for the success of the
project. General-purpose simulation tools often re-
quire much domain-specific customization. Therefore,
domain-specific simulation tools, like Factory Ex-
plorer or AutoSched AP, are often a better choice for
simulating wafer fabrication environment.
•Adapting the existing solution (i.e., SECS/GEM [69])
with IoT and cloud technologies and equipping them
with more intelligent decision rules.
Forming the hybrid configuration of OR&DS models
OR techniques are primarily applied to the decision-making
process. While there are many different ways to determine
how to make decisions, the most mainstream OR techniques
are focused on modeling decision problems in a mathemati-
cal programming framework. In these kinds of contexts, there
is typically a set of decision variables, constraints over these
variables, and an objective function dependent on decision
variables that are subjected to minimize or maximize. DS, on
the other hand, is mostly concerned with making inferences.
DS typically starting with a big pile of data and the purpose is
to infer something about data have not seen yet in the big pile.
The most common related research purposes are 1) which
solution yield the best results, 2) how the time-dependent
models can be extended for the future, 3) how a big pile
labeled data offer labels for new, unlabelled observations.
Therefore, making a hybrid solution by a combination of
OR&DS can fulfill all the needs for the trade-off between
data analysis and decision-making process. Research do-
mains include:
•Facilitating problems, and decision making based OR
perspective by data mining techniques.
•Implementing “Manufacturing Execution System"
(MES), “Enterprise Resource Planning" (ERP), and
8VOLUME 00, 2019

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2933167, IEEE Access
Khakifirooz et al.: Smart Semiconductor Manufacturing: OR&DS Perspectives
“Advanced Planning and Scheduling" (APS) for devel-
oping the integrated production planning and schedul-
ing solutions. Integrating the APS with ERP and MES
is a challenging issue to be considered.
•Decreasing the measurement uncertainty by merging
the hybrid methodology with state of the art statistical
inferences [70].
Simulation and data-driven solutions
As the scale and subsequently complexity of a production
process grow, the characterization of the process model,
which consists of physical elements becomes highly im-
portant. In particular, it is essential to employ a modeling
approach that can handle specification of scalable physical
models as size and complexity of data and system increases.
Research opportunities include:
•Simulating physical environment in order to compre-
hend the connections in real setting circumstance and
planning to find solution approaches in the risk-free
environment.
•Visualizing production planning processes by the use
of the event-driven process.
•Modeling and analyzing semiconductor challenges by
utilization of various simulation paradigms (i.e., agent-
based systems, hybrid models, reduced simulation
models, systems dynamics).
•Supporting the different aspect of decision-making in
the semiconductor by embedding the actual simulation
methods in the existing and forthcoming information
systems.
Process integration
As the costs of developing new product and restoring new
technology increase, a detailed simulation model represent-
ing the production operations, tools matching, scheduling,
and monitoring rules are needed for accurately planning
the capacity of these facilities and regulations. The main
challenge is a lengthy procedure of building, experimenting,
and analyzing a sufficiently detailed model for a new design.
The key to building accurate and computationally efficient
models is to decide on the details representing the equipment
capacity and advantage of applying high-tech technology.
On the other hand, the impact of energy consumption on
climate change and the rising cost of energy has become
a challenging issue for the semiconductor manufacturing
industry today. Regarding the new product design and use
of new technology, designing and deploying green and sus-
tainable manufacturing facilities is one of the key agenda
from International Technology Road-map of Semiconductor
(ITRS) for achieving the goals of ITRS, the product life-
cycle information from recovery organizations needs to take
into consideration to improve resource efficiency. Therefore,
possible research scopes include:
•Integrating decisions made by the different elements in
the system to avoid the ad hoc situation.
•Integrating the high-tech computing procedures to de-
rive the computationally tractable models, and to dis-
course, the diverse uncertainties come across in the
industry [71].
•Incorporating sustainability aspects into proposed so-
lutions and deterministic models.
•Integrating the product lifetime into account for de-
mand planning [72].
V. CONCLUSION AND FUTURE RESEARCH
DIRECTION
As a conclusion and future research direction, we attempted
to have a broader vision of the requirements for industrial
development and intelligence manufacturing of semiconduc-
tor products. These requirements are barely indicated in
literature with analytic context and are known as the new
obligations for the next step toward smart manufacturing.
Following we discuss some of the highlighted topics in this
chain.
A. SEMICONDUCTOR SUPPLY CHAIN MANAGEMENT
Semiconductor SC is growing exponentially and contributing
substantially to the global economy. This growth accom-
panies by continuous technology migration and minimizing
cost for different applications in green energy, communica-
tion, computers, automotive, medical, and electronics indus-
tries. [73]. There are some survey papers for Semiconduc-
tor SC with the scope of needs, practices and integration
issues such as 1) Research agenda framework for supply
network integration (questionnaire-based) [74]; 2) Decision
paradigms for SCM (questionnaire-based) [75]; 3) Successes
and opportunities in modelling and integrating planning,
scheduling, equipment configuration and fab capability as-
sessment [76], [77]; 4) E-markets and SC collaboration [78],
and 5) Strategic SC network design and SC simulation mod-
els [79]–[81].
According to [82] and [79], one future direction of semi-
conductor industry would be global SC simulation models
based on a marketing-operations perspective which leads
another research direction in the area of operations manage-
ment such as production planning and demand fulfillment,
inventory control, capacity and demand planning, and mar-
keting and sales models. Moreover, positioning the “Order
Penetration Points" (OPPs) in global semiconductor SC net-
works is another strategic competitive decision, especially for
novel product architectures with new options which can be
modeled with game theory (see [82], [83]).
B. SUSTAINABILITY AND RE-MANUFACTURING
Materials, products, and processes are becoming smarter,
sustainable, energy-aware, and innovation-driven. Sustain-
ability includes 1) Lower use of energy and materials, 2)
Greater environmental friendliness [84], and 3) Circular
economy and re-manufacturing [2]. Nowadays, the semi-
conductor industry has significantly and exponentially in-
creased the rate of e-waste in daily life [85], [86]. There is a
challenge for inventing efficient and pollution-free high-tech
recycling technologies for e-waste, which help to enhance the
VOLUME 00, 2019 9

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2933167, IEEE Access
Khakifirooz et al.: Smart Semiconductor Manufacturing: OR&DS Perspectives
comprehensive utilization of resources, and consequently, it
will develop the cyclic economy. There is a critical future
research direction on new recycling Electrostatic separation
which is simple and optimize energy consumption without
any wastewater discharge to recover the mixtures containing
conductors (copper), semiconductors (extrinsic silicon), and
nonconductors (woven glass-reinforced resin) in wafer fabri-
cation process [87].
C. GREEN SMART SEMICONDUCTOR
MANUFACTURING
Another future research stream would be data-driven de-
cision making and optimization applications in integrated
smart and green manufacturing. Some challenges in this area
would be: 1) Business Model Challenge: manufacturers face
threats from digital disruptors that are often quick to adapt
traditional products and exploit new opportunities through
the latest technology. 2) Data and Security Challenge: Smart
manufacturing is heavily reliant on technology and data,
which brings challenges of protecting data and ensuring
security. Smart manufacturing systems and the generated
data from that might also be targets for cyber attacks. 3)
Operations Challenges: Manufacturers need to be agile and
respond more quickly to update their technology. Connecting
different systems to get an end-to-end picture of the manufac-
turing process, supply chain, and product usage are a further
challenge [88].
Eventually, the fast-growing semiconductor manufactur-
ing requires a Knowledge Management Systems (KMS) in
order to support management DSS. This KMS will identify
and analyze research trend gaps and organize a future re-
search agenda for new product development [89].
REFERENCES
[1] L. Wang and G. Wang, “Big data in cyber-physical systems, digital
manufacturing and industry 4.0,” International Journal of Engineering and
Manufacturing (IJEM), vol. 6, no. 4, pp. 1–8, 2016.
[2] A. Kusiak et al., “Smart manufacturing must embrace big data,” Nature,
vol. 544, no. 7648, pp. 23–25, 2017.
[3] R. R. Schaller, “Moore’s law: past, present and future,” IEEE spectrum,
vol. 34, no. 6, pp. 52–59, 1997.
[4] W. Arden, M. Brillouët, P. Cogez, M. Graef, B. Huizing, and R. Mahnkopf,
“More-than-moore white paper,” Version, vol. 2, p. 14, 2010.
[5] B. Esmaeilian, S. Behdad, and B. Wang, “The evolution and future of
manufacturing: A review,” Journal of Manufacturing Systems, vol. 39, pp.
79–100, 2016.
[6] C.-Y. Lee and B.-S. Chen, “Mutually-exclusive-and-collectively-
exhaustive feature selection scheme,” Applied Soft Computing, vol. 68,
pp. 961–971, 2018.
[7] R. Wieringa, N. Maiden, N. Mead, and C. Rolland, “Requirements en-
gineering paper classification and evaluation criteria: a proposal and a
discussion,” Requirements Engineering, vol. 11, no. 1, pp. 102–107, 2006.
[8] F. Meziane, S. Vadera, K. Kobbacy, and N. Proudlove, “Intelligent systems
in manufacturing: current developments and future prospects,” Integrated
Manufacturing Systems, vol. 11, no. 4, pp. 218–238, 2000.
[9] K. Petersen, R. Feldt, S. Mujtaba, and M. Mattsson, “Systematic mapping
studies in software engineering.” in EASE, vol. 8, 2008, pp. 68–77.
[10] D. Delen and H. Demirkan, “Data, information and analytics as services,”
2013.
[11] D. Khera, M. Zaghoul, L. W. Linholm, and C. L. Wilson, “A neural
network approach for classifying test structure results,” in Microelectronic
Test Structures, International Conference on. IEEE, 1989, pp. 201–204.
[12] M. H. Hansen and D. A. James, “A computing environment for spatial data
analysis in the microelectronics industry,” Bell Labs Technical Journal,
vol. 2, no. 1, pp. 114–129, 1997.
[13] A. Famili, W.-M. Shen, R. Weber, and E. Simoudis, “Data preprocessing
and intelligent data analysis,” Intelligent data analysis, vol. 1, no. 1, pp.
3–23, 1997.
[14] A. Kusiak, “Decomposition in data mining: An industrial case study,”
IEEE transactions on electronics packaging manufacturing, vol. 23, no. 4,
pp. 345–353, 2000.
[15] M. Gardner and J. Bieker, “Data mining solves tough semiconductor
manufacturing problems,” in Proceedings of the sixth ACM SIGKDD
international conference on Knowledge discovery and data mining. ACM,
2000, pp. 376–383.
[16] A. Kusiak, “Feature transformation methods in data mining,” IEEE Trans-
actions on Electronics packaging manufacturing, vol. 24, no. 3, pp. 214–
221, 2001.
[17] D. Braha and A. Shmilovici, “Data mining for improving a cleaning pro-
cess in the semiconductor industry,” IEEE transactions on semiconductor
manufacturing, vol. 15, no. 1, pp. 91–101, 2002.
[18] A. Kusiak, “Rough set theory: a data mining tool for semiconductor man-
ufacturing,” IEEE transactions on electronics packaging manufacturing,
vol. 24, no. 1, pp. 44–50, 2001.
[19] P. Backus, M. Janakiram, S. Mowzoon, C. Runger, and A. Bhargava,
“Factory cycle-time prediction with a data-mining approach,” IEEE Trans-
actions on Semiconductor Manufacturing, vol. 19, no. 2, pp. 252–258,
2006.
[20] C.-F. Chien, W.-C. Wang, and J.-C. Cheng, “Data mining for yield
enhancement in semiconductor manufacturing and an empirical study,”
Expert Systems with Applications, vol. 33, no. 1, pp. 192–198, 2007.
[21] S.-C. Hsu and C.-F. Chien, “Hybrid data mining approach for pattern
extraction from wafer bin map to improve yield in semiconductor manu-
facturing,” International Journal of Production Economics, vol. 107, no. 1,
pp. 88–103, 2007.
[22] C. R. Glassey and W. W. Weng, “Dynamic batching heuristic for simul-
taneous processing,” IEEE transactions on semiconductor manufacturing,
vol. 4, no. 2, pp. 77–82, 1991.
[23] C.-Y. Lee, R. Uzsoy, and L. A. Martin-Vega, “Efficient algorithms
for scheduling semiconductor burn-in operations,” Operations Research,
vol. 40, no. 4, pp. 764–775, 1992.
[24] V. Chandru, C.-Y. Lee, and R. Uzsoy, “Minimizing total completion
time on batch processing machines,” international Journal of Production
Research, vol. 31, no. 9, pp. 2097–2121, 1993.
[25] D. S. Hochbaum and D. Landy, “Scheduling semiconductor burn-in oper-
ations to minimize total flowtime,” Operations research, vol. 45, no. 6, pp.
874–885, 1997.
[26] S. V. Mehta and R. Uzsoy, “Minimizing total tardiness on a batch pro-
cessing machine with incompatible job families,” IIE transactions, vol. 30,
no. 2, pp. 165–178, 1998.
[27] H. H. Xiong and M. Zhou, “Scheduling of semiconductor test facility via
petri nets and hybrid heuristic search,” IEEE Transactions on Semiconduc-
tor Manufacturing, vol. 11, no. 3, pp. 384–393, 1998.
[28] J. M. Swaminathan, “Tool capacity planning for semiconductor fabrication
facilities under demand uncertainty,” European Journal of Operational
Research, vol. 120, no. 3, pp. 545–558, 2000.
[29] S. J. Mason, J. W. Fowler, and W. Matthew Carlyle, “A modified shifting
bottleneck heuristic for minimizing total weighted tardiness in complex
job shops,” Journal of Scheduling, vol. 5, no. 3, pp. 247–262, 2002.
[30] H. Balasubramanian, L. Mönch, J. Fowler, and M. Pfund, “Genetic al-
gorithm based scheduling of parallel batch machines with incompatible
job families to minimize total weighted tardiness,” International Journal of
Production Research, vol. 42, no. 8, pp. 1621–1638, 2004.
[31] L. Mönch, H. Balasubramanian, J. W. Fowler, and M. E. Pfund, “Heuristic
scheduling of jobs on parallel batch machines with incompatible job
families and unequal ready times,” Computers & Operations Research,
vol. 32, no. 11, pp. 2731–2750, 2005.
[32] F. Pan and R. Nagi, “Robust supply chain design under uncertain demand
in agile manufacturing,” Computers & operations research, vol. 37, no. 4,
pp. 668–683, 2010.
[33] S.-S. Chen, “Intelligent control of semiconductor manufacturing pro-
cesses,” in Intelligent modeling, diagnosis and control of manufacturing
processes. World Scientific, 1992, pp. 211–230.
[34] P.-C. Chang, J.-C. Hieh, and T. W. Liao, “Evolving fuzzy rules for due-date
assignment problem in semiconductor manufacturing factory,” Journal of
Intelligent Manufacturing, vol. 16, no. 4-5, pp. 549–557, 2005.
10 VOLUME 00, 2019

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2933167, IEEE Access
Khakifirooz et al.: Smart Semiconductor Manufacturing: OR&DS Perspectives
[35] K. M. Tay and C. P. Lim, “Fuzzy fmea with a guided rules reduction system
for prioritization of failures,” International Journal of Quality & Reliability
Management, vol. 23, no. 8, pp. 1047–1066, 2006.
[36] H.-Y. Kang and A. H. Lee, “Priority mix planning for semiconductor
fabrication by fuzzy ahp ranking,” Expert Systems with Applications,
vol. 32, no. 2, pp. 560–570, 2007.
[37] P. Gyugyi, Y. M. Cho, G. Franklin, and T. Kailath, “Convex optimization
of wafer temperature trajectories for rapid thermal processing,” in Control
Applications, IEEE Conference on. IEEE, 1993, pp. 409–414.
[38] J. S. Kim and R. C. Leachman, “Decomposition method application to a
large scale linear programming wip projection model,” European journal
of operational research, vol. 74, no. 1, pp. 152–160, 1994.
[39] Y.-F. Hung and R. C. Leachman, “A production planning methodology
for semiconductor manufacturing based on iterative simulation and linear
programming calculations,” IEEE Transactions on Semiconductor manu-
facturing, vol. 9, no. 2, pp. 257–269, 1996.
[40] I. Chambouleyron, J. Martinez, A. Moretti, and M. Mulato, “Retrieval of
optical constants and thickness of thin films from transmission spectra,”
Applied Optics, vol. 36, no. 31, pp. 8238–8247, 1997.
[41] S. Bermon and S. J. Hood, “Capacity optimization planning system
(caps),” Interfaces, vol. 29, no. 5, pp. 31–50, 1999.
[42] S.-K. S. Fan, “Quality improvement of chemical-mechanical wafer pla-
narization process in semiconductor manufacturing using a combined
generalized linear modelling-non-linear programming approach,” Interna-
tional Journal of Production Research, vol. 38, no. 13, pp. 3011–3029,
2000.
[43] P.-C. Chang and T. W. Liao, “Combining som and fuzzy rule base for flow
time prediction in semiconductor manufacturing factory,” Applied Soft
Computing, vol. 6, no. 2, pp. 198–206, 2006.
[44] J. Asmundsson, R. L. Rardin, and R. Uzsoy, “Tractable nonlinear produc-
tion planning models for semiconductor wafer fabrication facilities,” IEEE
Transactions on Semiconductor Manufacturing, vol. 19, no. 1, pp. 95–111,
2006.
[45] A. Chatterjee, R. Nurani, S. Seshadri, and J. Shanthikumar, “Role of
yield management in fabless-foundry partnerships,” in Semiconductor
Manufacturing, IEEE International Symposium on. IEEE, 1997, pp. P31–
P34.
[46] S. Mallik and P. T. Harker, “Coordinating supply chains with competition:
Capacity allocation in semiconductor manufacturing,” European Journal
of Operational Research, vol. 159, no. 2, pp. 330–347, 2004.
[47] D. J. Miller and J. A. Fenimore, “The application of advanced manufactur-
ing technology in semiconductor fabrication,” in Electronic Manufacturing
Technology , Design-to-Manufacturing Transfer Cycle, IEEE International
Symposium on. IEEE, 1988, pp. 201–205.
[48] N. W. Hatch and D. C. Mowery, “Process innovation and learning by doing
in semiconductor manufacturing,” Management Science, vol. 44, no. 11-
part-1, pp. 1461–1477, 1998.
[49] J. F. Nijs, J. Szlufcik, J. Poortmans, S. Sivoththaman, and R. P. Mertens,
“Advanced manufacturing concepts for crystalline silicon solar cells,”
IEEE Transactions on Electron Devices, vol. 46, no. 10, pp. 1948–1969,
1999.
[50] R. C. Leachman, “Closed-loop measurement of equipment efficiency and
equipment capacity,” IEEE Transactions on Semiconductor manufactur-
ing, vol. 10, no. 1, pp. 84–97, 1997.
[51] S. H. Huang, J. P. Dismukes, J. Shi, Q. Su, G. Wang, M. A. Razzak,
and D. E. Robinson, “Manufacturing system modeling for productivity
improvement,” Journal of Manufacturing Systems, vol. 21, no. 4, pp. 249–
259, 2002.
[52] R. Oechsner, M. Pfeffer, L. Pfitzner, H. Binder, E. Müller, and T. Vonder-
strass, “From overall equipment efficiency (oee) to overall fab effective-
ness (ofe),” Materials science in semiconductor processing, vol. 5, no. 4-5,
pp. 333–339, 2002.
[53] J. Jacobs, L. Etman, E. Van Campen, and J. Rooda, “Characterization of
operational time variability using effective process times,” IEEE Transac-
tions on semiconductor manufacturing, vol. 16, no. 3, pp. 511–520, 2003.
[54] A. De Ron and J. Rooda, “Equipment effectiveness: Oee revisited,” IEEE
Transactions on Semiconductor Manufacturing, vol. 18, no. 1, pp. 190–
196, 2005.
[55] J. Morse, “Advanced manufacturing partnership: Where does nanomanu-
facturing fit, and where should investments be made?” NNN Newsletter,
vol. 4, no. 11, 2011.
[56] C.-Y. Hsu, C.-F. Chien, and P.-N. Chen, “Manufacturing intelligence for
early warning of key equipment excursion for advanced equipment control
in semiconductor manufacturing,” Journal of the Chinese Institute of
Industrial Engineers, vol. 29, no. 5, pp. 303–313, 2012.
[57] K. Dequeant, P. Vialletelle, P. Lemaire, and M.-L. Espinouse, “A literature
review on variability in semiconductor manufacturing: the next forward
leap to industry 4.0,” in Winter Simulation Conference. IEEE Press,
2016, pp. 2598–2609.
[58] J. Moyne, J. Samantaray, and M. Armacost, “Big data capabilities applied
to semiconductor manufacturing advanced process control,” IEEE Trans-
actions on Semiconductor Manufacturing, vol. 29, no. 4, pp. 283–291,
2016.
[59] T. J. Tang, A. Chung, A. Zhao, R. Kang, M. Zhang, K. Chien, J. Yang,
and J. Zhang, “An iot inspired semiconductor reliability test system inte-
grated with data-mining applications,” in Cloud Computing and Internet
of Things, International Conference on. IEEE, 2016, pp. 111–114.
[60] A. Weber, “The role of models in semiconductor smart manufacturing,” in
e-Manufacturing and Design Collaboration Symposium. IEEE, 2016, pp.
1–3.
[61] R. Herding and L. Mönch, “Designing an ontology for agent-based
planning and control tasks in semiconductor supply chains,” in OTM:
The Move to Meaningful Internet Systems, International Conference on.
Springer, 2016, pp. 65–75.
[62] S. Paramanathan, C. Farrukh, R. Phaal, and D. Probert, “Implementing
industrial sustainability: the research issues in technology management,”
R&D Management, vol. 34, no. 5, pp. 527–537, 2004.
[63] J. B. Silber, “Improving reuse of semiconductor equipment through
benchmarking, standardization, and automation,” Ph.D. dissertation, Mas-
sachusetts Institute of Technology, 2006.
[64] M. M. Crossan, H. W. Lane, and R. E. White, “An organizational learn-
ing framework: From intuition to institution,” Academy of management
review, vol. 24, no. 3, pp. 522–537, 1999.
[65] J. M. Swaminathan, “Enabling customization using standardized opera-
tions,” California Management Review, vol. 43, no. 3, pp. 125–135, 2001.
[66] J. D. Little and S. C. Graves, “Little’s law,” in Building intuition. Springer,
2008, pp. 81–100.
[67] D. J. Power, M. Khakifirooz, M. Fathi, and P. M. Pardalos, “How can deci-
sion support aid smart manufacturing initiatives?” http://dssresources.com,
2018.
[68] M. Khakifirooz, C.-F. Chien, and M. Fathi, “Compensating misalignment
using dynamic random-effect control system: A case of high-mixed wafer
fabrication,” IEEE Transactions on Automation Science and Engineering,
2019.
[69] Y.-C. Fan and J.-Y. J. Chang, “Equipment communication architecture
for smart manufacturing,” in Smart Manufacturing, Industrial & Logistics
Engineering (SMILE), 2018 IEEE International Conference on. IEEE,
2018, pp. 27–32.
[70] T. Paksoy, ˙
I. Karao˘
glan, H. Gökçen, P. M. Pardalos, and B. Tor˘
gul, “An
experimental research on closed loop supply chain management with
internet of things,” Journal of Economics Bibliography, vol. 3, no. 1S, pp.
1–20, 2016.
[71] C. Fang, X. Liu, P. M. Pardalos, and J. Pei, “Optimization for a three-stage
production system in the internet of things: procurement, production and
product recovery, and acquisition,” The International Journal of Advanced
Manufacturing Technology, vol. 83, no. 5-8, pp. 689–710, 2016.
[72] X. Liu, J. Pei, L. Liu, H. Cheng, M. Zhou, and P. M. Pardalos, “Life cycle
assessment in an iot environment,” in Optimization and Management in
Manufacturing Engineering. Springer, 2017, pp. 209–246.
[73] C.-F. Chien, S. Dauzère-Pérès, H. Ehm, J. W. Fowler, Z. Jiang, S. Krish-
naswamy, T.-E. Lee, L. Moench, and R. Uzsoy, “Modelling and analysis
of semiconductor manufacturing in a shrinking world: challenges and
successes,” European Journal of Industrial Engineering 4, vol. 5, no. 3,
pp. 254–271, 2011.
[74] T. Callarman, J. Fowler, E. Gel, M. Pfund, and D. Shunk, “Creating a re-
search agenda framework for semiconductor supply network integration,”
in Evolution of Supply Chain Management. Springer, 2004, pp. 161–187.
[75] Y. Sun, A. Feller, D. Shunk, J. Fowler, T. Callarman, and B. Duarte,
“Decision paradigms in the semiconductor supply chain: a survey and
analysis,” in Automation Science and Engineering, 2007. CASE 2007.
IEEE International Conference on. IEEE, 2007, pp. 106–110.
[76] K. Fordyce, R. J. Milne, C.-T. Wang, and H. Zisgen, “Modeling and
integration of planning, scheduling, and equipment configuration in semi-
conductor manufacturing: Part i. review of successes and opportunities,”
International Journal of Industrial Engineering: Theory, Applications and
Practice, vol. 22, no. 5, pp. 575–600, 2015.
VOLUME 00, 2019 11

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2019.2933167, IEEE Access
Khakifirooz et al.: Smart Semiconductor Manufacturing: OR&DS Perspectives
[77] ——, “Modeling and integration of planning, scheduling, and equipment
configuration in semiconductor manufacturing: Part ii. fab capability as-
sessment,” International Journal of Industrial Engineering: Theory, Appli-
cations and Practice, vol. 22, no. 5, pp. 601–617, 2015.
[78] B. K. Bahinipati and S. Deshmukh, “E-markets and supply chain collab-
oration: a literature-based review of contributions with specific reference
to the semiconductor industries,” Logistics Research, vol. 4, no. 1-2, pp.
19–38, 2012.
[79] L. Mönch, R. Uzsoy, and J. W. Fowler, “A survey of semiconductor
supply chain models part i: semiconductor supply chains, strategic network
design, and supply chain simulation,” International Journal of Production
Research, pp. 1–22, 2017.
[80] R. Uzsoy, J. W. Fowler, and L. Mönch, “A survey of semiconductor
supply chain models part ii: demand planning, inventory management, and
capacity planning,” International Journal of Production Research, pp. 1–
19, 2018.
[81] ——, “A survey of semiconductor supply chain models part ii: demand
planning, inventory management, and capacity planning,” International
Journal of Production Research, pp. 1–19, 2018.
[82] E. Teimoury and M. Fathi, “An integrated operations-marketing perspec-
tive for making decisions about order penetration point in multi-product
supply chain: a queuing approach,” International Journal of Production
Research, vol. 51, no. 18, pp. 5576–5596, 2013.
[83] E. Teimoury, M. Modarres, I. Khondabi, and M. Fathi, “A queuing ap-
proach for making decisions about order penetration point in multiechelon
supply chains,” The International Journal of Advanced Manufacturing
Technology, vol. 63, no. 1-4, pp. 359–371, 2012.
[84] P. Centobelli, R. Cerchione, and E. Esposito, “Environmental sustainabil-
ity in the service industry of transportation and logistics service providers:
Systematic literature review and research directions,” Transportation Re-
search Part D: Transport and Environment, vol. 53, pp. 454–470, 2017.
[85] M. Kwak, S. Behdad, Y. Zhao, H. Kim, and D. Thurston, “E-waste stream
analysis and design implications,” Journal of Mechanical Design, vol. 133,
no. 10, p. 101003, 2011.
[86] M. Sabbaghi, S. Behdad, and J. Zhuang, “Managing consumer behavior
toward on-time return of the waste electrical and electronic equipment: A
game theoretic approach,” International Journal of Production Economics,
vol. 182, pp. 545–563, 2016.
[87] M. Xue, G. Yan, J. Li, and Z. Xu, “Electrostatic separation for recycling
conductors, semiconductors, and nonconductors from electronic waste,”
Environmental science & technology, vol. 46, no. 19, pp. 10556–10 563,
2012.
[88] P. M. Pardalos, “Smart/green manufacturing: Data enabled
decision making and optimization applications-presentation;
http://www.ise.ufl.edu/cao/,” 2017.
[89] P. Centobelli, R. Cerchione, and E. Esposito, “Knowledge management
in startups: systematic literature review and future research agenda,”
Sustainability, vol. 9, no. 3, p. 361, 2017.
MARZIEH KHAKIFIROOZ has a Ph.D. in Indus-
trial Engineering and Engineering Management
and an MS degree in Industrial Statistics from the
National Tsing Hua University (NTHU), Hsinchu,
Taiwan. Currently, she is an assistant professor at
school of engineering, Tecnològico de Monterrey,
Mexico. Khakifirooz has outstanding practical ex-
perience from her various global consultancies for
high-tech industries. Her research interests include
the application of optimization in smart manufac-
turing, Industry 4.0, decision making, and machine teaching. She is an active
member of System Dynamic Society, Institute of Electrical and Electronics
Engineers (IEEE), and Institute of Industrial and Systems Engineers (IISE).
MAHDI FATHI is a Postdoctoral Associate at the
Department of Industrial and Systems Engineering
at Mississippi State University. He received his
BS and MS from the Department of Industrial
Engineering, Amirkabir University of Technology
(Tehran Polytechnic) and Ph.D. from Iran Uni-
versity of Science and Technology, Tehran, Iran
in 2006, 2008 and 2013, respectively. He is the
recipient of three postdoctoral fellowships and was
a visiting scholar at Center for Applied Optimiza-
tion, Dept. of Industrial and Systems Engineering-University of Florida
(USA) and Dept. of Electrical Engineering-National Tsing Hua University
in Taiwan. He worked at Optym as a senior systems engineer and at
A Model Of Reality Inc. as a system design engineer in the USA and
several other companies in different industry sectors. Prof. Fathi is an active
member of several societies and institutions and serves on the editorial
board of several journals. His research interests include Queuing Theory
and Its Applications; Stochastic Process; Optimization; Artificial Intelligent;
Uncertain Quantification; Smart Manufacturing & Industry 4.0; Reliability
with their applications in Health Care, Bio-medicine, Agriculture Energy.
KAN WU is currently an assistant professor in
the Division of Systems and Engineering Man-
agement, NTU. He received the BS degree from
National Tsinghua University in Taiwan, MS de-
gree from University of California at Berkeley in
1996, and Ph.D. degree from Georgia Institute
of Technology in 2009 (Major in ISyE). He has
over 10 years of experience in the semiconductor
industry, from consultants to managers. Before
joining NTU, he was the CTO and founding team
member of a startup company in the US. His Ph.D. dissertation was awarded
the third place for the IIE Pritsker Doctoral Dissertation Award in 2010.
His research interests are primarily in the performance evaluation of supply
chains and manufacturing systems.
12 VOLUME 00, 2019