![The Petri net derived from the Finite State Automaton representing the Declare model mined out of BPIC 2013 log [33]. Labels are abbreviated: a = " Accepted " , c = " Completed " , u = " Unmatched " , q = " Queued " .](https://www.researchgate.net/profile/Johannes-Prescher-2/publication/268743962/figure/fig1/AS:392314607423488@1470546499015/The-Petri-net-derived-from-the-Finite-State-Automaton-representing-the-Declare-model_Q320.jpg)
Content uploaded by Johannes Prescher
Author content
All content in this area was uploaded by Johannes Prescher on Nov 25, 2014
Content may be subject to copyright.
A preview of the PDF is not available
From Declarative Processes to Imperative Models
Abstract and Figures
Nowadays organizations support their creation of value by explicitly defining the processes to be carried out. Processes are specifically discussed from the angle of simplicity, i.e., how compact and easy to understand they can be represented. In most cases, organizations rely on imperative models which, how-ever, become complex and cluttered when it comes to flexibility and optionality. As an alternative, declarative modeling reveals to be effective under such cir-cumstances. While both approaches are well known for themselves, there is still not a deep understanding of their semantic interoperability. With this work, we examine the latter and show how to obtain an imperative model out of a set of declarative constraints. To this aim, we devise an approach leading from a De-clare model to a behaviorally equivalent Petri net. Furthermore, we demonstrate that any declarative control flow can be represented by means of a Petri net for which the property of safety always holds true.
Figures - uploaded by Johannes Prescher
Author content
All figure content in this area was uploaded by Johannes Prescher
Content may be subject to copyright.
... Alternatively, conversions to the same execution language can be obtained. However, the conversion turns out to be often intractable [50], or resorts to approaches that are too language-specific [51,52]. The latter has caused no previous work to apply existing techniques, either procedural or declarative, to mixed-paradigm models. ...
... In [51], Declare constraints are translated into equivalent Petri nets with weighted, reset and inhibitor arcs. In [50,13], regular expressions (REs) are used to define the semantics of Declare templates. Since REs and Monadic Second ...
... Other mixed-paradigm solutions exist as well, such as BPMN with Declare (BPMN-D)[52]. Furthermore, conversions between both types of models have been introduced, e.g., from Declare to Petri Pre-print copy of the manuscript published by Elsevier identified by doi: 10.1016/j.is.2020.101685nets[50,51]. Nevertheless, both approaches can be converted to models that are compatible with the proposed conformance checking technique. ...
Mixed-paradigm process models integrate strengths of procedural and declarative representations like Petri nets and Declare. They are specifically interesting for process mining because they allow capturing complex behavior in a compact way. A key research challenge for the proliferation of mixed-paradigm models for process mining is the lack of corresponding conformance checking techniques. In this paper, we address this problem by devising the first approach that works with intertwined state spaces of mixed-paradigm models. More specifically, our approach uses an alignment-based replay to explore the state space and compute trace fitness in a procedural way. In every state, the declarative constraints are separately updated, such that violations disable the corresponding activities. Our technique provides for an efficient replay towards an optimal alignment by respecting all orthogonal Declare constraints. We have implemented our technique in ProM and demonstrate its performance in an evaluation with real-world event logs.
... Alternatively, conversions to the same execution language can be obtained. However, the conversion turns out to be often intractable [50], or resorts to approaches that are too language-specific [51,52]. The latter has caused no previous work to apply existing techniques, either procedural or declarative, to mixed-paradigm models. ...
... In [51], Declare constraints are translated into equivalent Petri nets with weighted, reset and inhibitor arcs. In [50,13], regular expressions (REs) are used to define the semantics of Declare templates. Since REs and Monadic Second ...
... Other mixed-paradigm solutions exist as well, such as BPMN with Declare (BPMN-D) [52]. Furthermore, conversions between both types of models have been introduced, e.g., from Declare to Petri nets [50,51]. Nevertheless, both approaches can be converted to models that are compatible with the proposed conformance checking technique. ...
Mixed-paradigm process models integrate strengths of procedural and declarative representations like Petri nets and Declare. They are specifically interesting for process mining because they allow capturing complex behaviour in a compact way. A key research challenge for the proliferation of mixed-paradigm models for process mining is the lack of corresponding conformance checking techniques. In this paper, we address this problem by devising the first approach that works with intertwined state spaces of mixed-paradigm models. More specifically, our approach uses an alignment-based replay to explore the state space and compute trace fitness in a procedural way. In every state, the declarative constraints are separately updated, such that violations disable the corresponding activities. Our technique provides for an efficient replay towards an optimal alignment by respecting all orthogonal Declare constraints. We have implemented our technique in ProM and demonstrate its performance in an evaluation with real-world event logs.
... We describe four examples that are frequently used for process mining, namely directly-follows 8 graphs, Petri nets, Declare constraints, and process trees [2]. It is well known that transformations exist between many of these representations [18,19,20] Directly-follows graphs. Event sequences can be modeled as a graph, where each node represents a type of event and arcs describe directly-follows relationships between these event types [2]. ...
... Event sequences can be described as a collection of rules by which 9 the underlying behavior is constrained. One such specific modeling language that uses LTL for describing event sequence data is Declare [23,20]. Table 2 shows a set of rules showing that constrain event sequences. ...
Event sequence data is increasingly available. Many business operations are supported by information systems that record transactions, events, state changes, message exchanges, and so forth. This observation is equally valid for various industries, including production, logistics, healthcare, financial services, education, to name but a few. The variety of application areas explains that techniques for event sequence data analysis have been developed rather independently in different fields of computer science. Most prominent are contributions from information visualization and from process mining. So far, the contributions from these two fields have neither been compared nor have they been mapped to an integrated framework. In this paper, we develop the Event Sequence Visualization framework (ESeVis) that gives due credit to the traditions of both fields. Our mapping study provides an integrated perspective on both fields and identifies potential for synergies for future research.
... • Random Conjunction formulas combine 19 common LTL f formulas from [17,38] (see the first 19 patterns in Table 2) as random conjunctions in the style of [39] in two sets of 500 formulas: • Pattern Formulas 7 scalable patterns (see the last 7 patterns in Table 2) inspired by [40] up to length 100. The total number of generated formulas is 700. ...
... Declare Patterns From [17,38] sat* Existence a a must be executed at least once sat* Absence 2 ¬ (a ∧ a) a can be executed at most once sat* Choice ...
Linear Temporal Logic over finite traces (LTLf) was proposed in 2013 and has attracted increasing interest around the AI community. Though the theoretic basis for LTLf has been thoroughly explored since that time, there are still few algorithmic tools that are able to provide an efficient reasoning strategy for LTLf. In this paper, we present a SAT-based framework for LTLf satisfiability checking, which is the foundation of LTLf reasoning. We use propositional SAT-solving techniques to construct a transition system, which is an automata-style structure, for an input LTLf formula; satisfiability checking is then reduced to a path-search problem over this transition system. Based on this framework, we further present CDLSC (Conflict-Driven LTLf Satisfiability Checking), a novel algorithm (heuristic) that leverages information produced by propositional SAT solvers, utilizing both satisfiability and unsatisfiability results. More specifically, the satisfiable results of the SAT solver are used to create new states of the transition system and the unsatisfiable results to accelerate the path search over the system. We evaluate all 5 off-the-shelf LTLf satisfiability algorithms against our new approach CDLSC. Based on a comprehensive evaluation over 4 different LTLf benchmark suits with a total amount of 9317 formulas, our time-cost analysis shows that 1) CDLSC performs best on checking unsatisfiable formulas by achieving approximately a 4X time speedup, compared to the second-best solution (K-LIVE [1]); 2) Although no approaches dominate checking satisfiable formulas, CDLSC performs best on 2 of the total 4 tested satisfiable benchmark suits; and 3) CDLSC gains the best overall performance when considering both satisfiable and unsatisfiable instances.
... Therefore, in formal terms, from case 3 onwards, the behavioral rule that the sepsis triage occurs after the general triage abruptly decreases in the number and share of observations. Several rule languages with a rich spectrum of behavioral constraints have been proposed [27], [28], [29]. In rule languages based on linear temporal logic such as DECLARE, we can formally state that the rule ALTERNATERESPONSEpER Triage, ER Sepsis Triageq drops in confidence. ...
Event sequence data is increasingly available in various application domains, such as business process management, software engineering, or medical pathways. Processes in these domains are typically represented as process diagrams or flow charts. So far, various techniques have been developed for automatically generating such diagrams from event sequence data. An open challenge is the visual analysis of drift phenomena when processes change over time. In this paper, we address this research gap. Our contribution is a system for fine-granular process drift detection and corresponding visualizations for event logs of executed business processes. We evaluated our system both on synthetic and real-world data. On synthetic logs, we achieved an average F-score of 0.96 and outperformed all the state-of-the-art methods. On real-world logs, we identified all types of process drifts in a comprehensive manner. Finally, we conducted a user study highlighting that our visualizations are easy to use and useful as perceived by process mining experts. In this way, our work contributes to research on process mining, event sequence analysis, and visualization of temporal data.
... Therefore, in formal terms, from case 3 onwards, the behavioral rule that the sepsis triage occurs after the general triage abruptly decreases in the number and share of observations. Several rule languages with a rich spectrum of behavioral constraints have been proposed [27], [28], [29]. In rule languages based on linear temporal logic such as DECLARE, we can formally state that the rule ALTERNATERESPONSEpER Triage, ER Sepsis Triageq drops in confidence. ...
Event sequence data is increasingly available in various application domains, such as business process management, software engineering, or medical pathways. Processes in these domains are typically represented as process diagrams or flow charts. So far, various techniques have been developed for automatically generating such diagrams from event sequence data. An open challenge is the visual analysis of drift phenomena when processes change over time. In this paper, we address this research gap. Our contribution is a system for fine-granular process drift detection and corresponding visualizations for event logs of executed business processes. We evaluated our system both on synthetic and real-world data. On synthetic logs, we achieved an average F-score of 0.96 and outperformed all the state-of-the-art methods. On real-world logs, we identified all types of process drifts in a comprehensive manner. Finally, we conducted a user study highlighting that our visualizations are easy to use and useful as perceived by process mining experts. In this way, our work contributes to research on process mining, event sequence analysis, and visualization of temporal data.
... The benefit of using Declare constraints is that they cover both unary and binary relations, meaning they can also cover the behaviour of a single activity. Furthermore, the constraints' automata can be converted into Petri nets if desired [42,43]. ...
The field of predictive process monitoring focuses on modelling future characteristics of running business process instances, typically by either predicting the outcome of particular objectives (e.g. completion (time), cost), or next-in-sequence prediction (e.g. what is the next activity to execute). This paper introduces Processes-As-Movies (PAM), a technique that provides a middle ground between these predictive monitoring. It does so by capturing declarative process constraints between activities in various windows of a process execution trace, which represent a declarative process model at subsequent stages of execution. This high-dimensional representation of a process model allows the application of predictive modelling on how such constraints appear and vanish throughout a process' execution. Various recurrent neural network topologies tailored to high-dimensional input are used to model the process model evolution with windows as time steps, including encoder-decoder long short-term memory networks, and convolutional long short-term memory networks. Results show that these topologies are very effective in terms of accuracy and precision to predict a process model's future state, which allows process owners to simultaneously verify what linear temporal logic rules hold in a predicted process window (objective-based), and verify what future execution traces are allowed by all the constraints together (trace-based).
... Another context in which one can investigate the applicability and adaptation of our approach is that of the models of behaviour expressing distributed systems invariants [42]. Finally, one can design new quality measures that relate arbitrary numbers of behaviours (not just behaviours of a specification and its execution log), e.g., to establish a basis for comparing results of various process querying methods [79], models of behaviour that summarise traces at varying levels of abstraction [43], and different behavioural representations [83]. ...
The behavioural comparison of systems is an important concern of software engineering research. For example, the areas of specification discovery and specification mining are concerned with measuring the consistency between a collection of execution traces and a program specification. This problem is also tackled in process mining with the help of measures that describe the quality of a process specification automatically discovered from execution logs. Though various measures have been proposed, it was recently demonstrated that they neither fulfil essential properties, such as monotonicity, nor can they handle infinite behaviour. In this article, we address this research problem by introducing a new framework for the definition of behavioural quotients. We prove that corresponding quotients guarantee desired properties that existing measures have failed to support. We demonstrate the application of the quotients for capturing precision and recall measures between a collection of recorded executions and a system specification. We use a prototypical implementation of these measures to contrast their monotonic assessment with measures that have been defined in prior research.
... There are some approaches to construct activity-centric process models from the data-centric process models. An approach to transforming declarative models into Petri Net models is presented in [8]. This approach maps the declarative constraints of the input model into regular expressions, transforms them into the Finite State Automaton and then derives a Petri Net. ...
In recent years, artifact-centric paradigm as a data-centric approach to business process modeling has gained momentum. Compared to the traditional activity-centric paradigm that focuses on process control-flow and treats data as simple black boxes that act as input and output to these activities, the artifact-centric paradigm provides equal support to both the control-flow and data. Most of the existing process modeling is activity-centric, although the artifact-centric modeling enables higher process flexibility and reusability. This is mainly due to the existence of numerous notations, tools and technologies that provide increased support to activity-centric process modeling and execution. Therefore, this paper proposes a trace-based approach to transform artifact-centric process models into activity-centric process models and to analyse the consistency of transformed and base models. A case study is utilized to demonstrate the feasibility of the proposed approach.
Artful processes are those processes in which the experience, intuition, and knowledge of the actors are the key factors in determining the decision making. They are typically carried out by the "knowledge workers," such as professors, managers, and researchers. They are often scarcely formalized or completely unknown a priori. Throughout this article, we discuss how we addressed the challenge of discovering declarative control flows in the context of artful processes. To this extent, we devised and implemented a two-phase algorithm, named MINERful. The first phase builds a knowledge base, where statistical information extracted from logs is represented. During the second phase, queries are evaluated on that knowledge base, in order to infer the constraints that constitute the discovered process. After outlining the overall approach and offering insight on the adopted process modeling language, we describe in detail our discovery technique. Thereupon, we analyze its performances, both from a theoretical and an experimental perspective. A user-driven evaluation of the quality of results is also reported on the basis of a real case study. Finally, a study on the fitness of discovered models with respect to synthetic and real logs is presented.
Petri nets efficiently model both data- and control-flow. Control-flow is either modeled explicitly as flow of a specific kind of data, or implicit based on the data-flow. Explicit modeling of control-flow is useful for well-known and highly structured processes, but may make modeling of abstract features of models, or processes which are highly dynamic, overly complex. Declarative modeling, such as is supported by Declare and DCR graphs, focus on control-flow, but does not specify it explicitly; instead specifications come in the form of constraints on the order or appearance of tasks. In this paper we propose a combination of the two, using colored Petri nets instead of plain Petri nets to provide full data support. The combined approach makes it possible to add a focus on data to declarative languages, and to remove focus from the explicit control-flow from Petri nets for dynamic or abstract processes. In addition to enriching both procedural processes in the form of Petri nets and declarative processes, we also support a flow from modeling only abstract data- and control-flow of a model towards a more explicit control-flow model if so desired. We define our combined approach, and provide considerations necessary for enactment. Our approach has been implemented in CPN Tools 4.
Artful processes are informal processes typically carried out by those people whose work is mental rather than physical (managers, professors, researchers, engineers, etc.), the so called "knowledge workers". In this paper we propose the MailOfMine approach to automatically build, on top of a collection of e-mail messages, a set of workflow models that represent the artful processes laying behind the knowledge workers activities.
Declarative approaches are particularly suitable for modeling highly flexible processes. They especially apply to artful processes, i.e., rapid informal processes that are typically carried out by those people whose work is mental rather than physical (managers, professors, researchers, engineers, etc.), the so called "knowledge workers". This paper describes MINERful++, a two-step algorithm for an efficient discovery of constraints that constitute declarative workflow models. As a first step, a knowledge base is built, with information about temporal statistics gathered from execution traces. Then, the statistical support of constraints is computed, by querying that knowledge base. MINERful++ is fast, modular, independent of the specific formalism adopted for representing constraints, based on a probabilistic approach and capable of eliminating the redundancy of subsumed constraints.
A linear bounded transducer (pushdown transducer) is a linear bounded automaton (pushdown automaton) with outputs. Answers to the following two problems are derived in the paper: (1) If S is a linear bounded transducer or a pushdown transducer, and L is context sensitive, context free, or regular; is S(L) some well known type of set? (2) Does there exist a decision procedure to determine for arbitrary sets L1 and L2, both context sensitive or both context free, whether there exists a linear bounded transducer (pushdown transducer) which maps (a) L1 onto L2 or (b) L1 into L2 nontrivially?
CPN Tools is a tool for modeling, simulating, and analyzing colored Petri nets. The latest iteration of the tool, CPN Tools 4, extends this with constraints known from declarative languages such as Declare and DCR Graphs. Furthermore, this version introduces an explicit process perspective, powerful extensibility allowing third parties to extend the tools capabilities, and a visualization perspective making it possible to make high-level visualizations of executions directly in the tool. In our demonstration, we show how it is possible to create models incorporating declarative and imperative constructs and how to use these models to generate simulation logs that can be directly imported into ProM. We show off the new process perspective on top of colored Petri nets, exemplify the use of the perspective to generate readable Java code directly from models, and show how the visualization perspective makes it possible to show the formal underlying model alongside an easier-tograsp for non-experts high-level visualization. Our intended audience comprise current users of CPN Tools interested in recent developments and practitioners interested in colored Petri nets and hybrid models. We expect to tailor each demonstration to the wishes of the audience.
In this paper we study when an LTL formula on finite traces (LTL/ formula) is insensitive to infiniteness, that is, it can be correctly handled as a formula on infinite traces under the assumption that at a certain point the infinite trace starts repeating an end event forever, trivializing all other propositions to false. This intuition has been put forward and (wrongly) assumed to hold in general in the literature. We define a necessary and sufficient condition to characterize whether an LTL/ formula is insensitive to infiniteness, which can be automatically checked by any LTL reasoner. Then, we show that typical LTL/ specification patterns used in process and service modeling in CS, as well as trajectory constraints in Planning and transition-based LTL/ specifications of action domains in KR, are indeed very often insensitive to infiniteness. This may help to explain why the assumption of interpreting LTL on finite and on infinite traces has been (wrongly) blurred. Possibly because of this blurring, virtually all literature detours to Buchi automata for constructing the NFA that accepts the traces satisfying an LTL/ formula. As a further contribution, we give a simple direct algorithm for computing such NFA. Copyright © 2014, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.
In this paper we look into the assumption of interpreting LTL over finite traces. In particular we show that LTLf, i.e., LTL under this assumption, is less expressive than what might appear at first sight, and that at essentially no computational cost one can make a significant increase in expressiveness while maintaining the same intuitiveness of LTLf. Indeed, we propose a logic, LDLf for Linear Dynamic Logic over finite traces, which borrows the syntax from Propositional Dynamic Logic (PDL), but is interpreted over finite traces. Satisfiability, validity and logical implication (as well as model checking) for LDLf are PSPACE-complete as for LTLf (and LTL).
Petrify is a tool for (1) manipulating concurrent specifications and (2) synthesis and optimization of asynchronous control circuits. Given a Petri Net (PN), a Signal Transition Graph (STG), or a Transition System (TS) it (1) generates another PN or STG which is simpler than the original description and (2) produces an optimized net-list of an asynchronous controller in the target gate library while preserving the specified input-output behavior. An ability of back-annotating to the specification level helps the designer to control the design process. For transforming a specification petrify performs a token flow analysis of the initial PN and produces a transition system (TS). In the initial TS, all transitions with the same label are considered as one event. The TS is then transformed and transitions relabeled to fulfill the conditions required to obtain a safe irredundant PN. For synthesis of an asynchronous circuit petrify performs state assignment by solving the Complete State Coding problem. State assignment is coupled with logic minimization and speed-independent technology mapping to a target library. The final net-list is guaranteed to be speed-independent, i.e., hazard-free under any distribution of gate delays and multiple input changes satisfying the initial specification. The tool has been used for synthesis of PNs and PNs composition, synthesis and re-synthesis of asynchronous controllers and can be also applied in areas related with the analysis of concurrent programs. This paper provides an overview of petrify and the theory behind its main functions.
