Johannes De Smedt

Lecture Notes in Business Information Processing

The act of retrieving process models from event-based data logs can offer valuable information to business owners. Many approaches have been proposed for this purpose, mining for either a procedural or declarative outcome. A blended approach that combines both process model paradigms exists and offers a great way to deal with process environments which consist of different layers of flexibility. In this paper, it will be shown how to check such models for correctness, and how this checking can contribute to retrieving the models as well. The approach is based on intersecting both parts of the model and provides an effective way to check (i) whether the behavior is aligned, and (ii) where the model can be improved according to errors that arise along the respective paradigms. To this end, we extend the functionality of Fusion Miner, a mixed-paradigm process miner, in a way to inspect which amount of flexibility is right for the event log. The procedure is demonstrated with an implemented model checker and verified on real-life event logs.

Fusion Miner

The research area of business process mining has vastly matured in recent years. Its main focus centers around the extraction and analysis of process models from event logs. A strong emphasis lies on the automatic discovery of models for which numerous algorithms have been proposed already. So far, most discovery algorithms were limited to the derivation of single-paradigm models, which contain either procedural or declarative constructs, targeting the mining of strict and flexible processes respectively. This paper proposes the first fully-automated mining technique to discover procedural workflows combined with Declare templates to capture processes that are difficult to mine with only a single paradigm, e.g., workflows with different layers of flexibility.This approach provides process analysts with new discovery capabilities, including the retrieval of better fitting and more precise models with high comprehensibility. The main contribution consists of the Fusion Miner algorithm, which has been implemented in the process mining framework ProM as a plug-in. The first approach to offer a way to analyze process-oriented data in a mixed-paradigm fashion, i.e., one that mixes declarative and procedural models in a single state space. The outcome yields fit and precise process models that better support practitioners in judging the quality of their as-is business processes.An overview of how mixed-paradigm process models with intertwined state spaces can improve over single-paradigm approaches.A preliminary approach towards conformance checking of mixed-paradigm models.Three elaborate examples which explain how mixed-paradigm models can excel at capturing logs with different layers of flexibility.

Mixed-Paradigm Process Modeling with Intertwined State Spaces

Business process modeling often deals with the trade-off between comprehensibility and flexibility. Many languages have been proposed to support different paradigms to tackle these characteristics. Well-known procedural, token-based languages such as Petri nets, BPMN, EPC, etc. have been used and extended to incorporate more flexible use cases, however the declarative workflow paradigm, most notably represented by the Declare framework, is still widely accepted for modeling flexible processes. A real trade-off exists between the readable, rather inflexible procedural models, and the highly-expressive but cognitively demanding declarative models containing a lot of implicit behavior. This paper investigates in detail the scenarios in which combining both approaches is useful, it provides a scoring table for Declare constructs to capture their intricacies and similarities compared to procedural ones, and offers a step-wise approach to construct mixed-paradigm models. Such models are especially useful in the case of environments with different layers of flexibility and go beyond using atomic subprocesses modeled according to either paradigm. The paper combines Petri nets and Declare to express the findings.

Towards a Holistic Discovery of Decisions in Process-Aware Information Systems

The interest of integrating decision analysis approaches with the automated discovery of processes from data has seen a vast surge over the past few years. Most notably the introduction of the Decision Model and Notation (DMN) standard by the Object Management Group has provided a suitable solution for filling the void of decision representation in business process modeling languages. Process discovery has already embraced DMN for so-called decision mining, however, the efforts are still limited to a control flow point of view, i.e., explaining routing (constructs) or decision points. This work, however, introduces an integrated way of capturing the decisions that are embedded in the process, which is not limited to local characteristics, but provides a decision model in the form of a decision diagram which encompasses the full process execution span. Therefore, a typology is proposed for classifying different activities that contribute to the decision dimension of the process. This enables the possibility for an in-depth analysis of every activity, deciding whether it entails a decision, and what its relation is to other activities. The findings are implemented and illustrated on the 2013 BPI Challenge log, an exemplary dataset originating from a decision-driven process.

Multi-paradigm Process Mining: Retrieving Better Models by Combining Rules and Sequences

Business process mining is a well-established field of research which focuses on the automatic retrieval and analysis of process flows. The discovery and representation of these models is based on techniques that come in all shapes and forms. Most notably, procedurally-based algorithms such as Heuristics Miner have been used successfully for this purpose. Also, declarative process model miners have been proposed, which give other insights into the model by generating rules that apply on the activities. This paper proposes an integrated approach to combining these paradigms to discover process models that contain best of both worlds to enrich insights into the event logs under scrutiny.

Challenges in Refactoring Processes to Include Decision Modelling

Until lately decisions were regularly modelled as a part of the process model, negatively affecting the maintainability, comprehensibility and flexibility of processes as well as decisions. The recent establishment of the Decision Model and Notation (DMN) standard provides an opportunity for shifting in favour of a separation of concerns between the decision and process model. However, this challenge of separation of concerns and subsequently consistent integration has received limited attention. This work discusses difficulties and challenges in separating the concerns and then integrating the two models. The most challenging scenario is when integrating decision models which entail the process holistically, rather than merely focusing the decisions on local decision points. In this work, we shed a light on the importance of the separation of concerns and identify inconsistencies that might arise when separating and integrating processes and decisions.

A Service-Oriented Architecture Design of Decision-Aware Information Systems: Decision as a Service

Separating the decision modelling concern from the processes modelling concern has gained significant support in literature over the past few years, as incorporating both concerns into a single model has shown to impair the scalability, maintainability, flexibility and understandability of both processes and decisions. Most notably the introduction of the Decision Model and Notation (DMN) standard by the Object Management Group has provided a suitable solution for externalising decisions from processes. This paper introduces a systematic way of tackling the separation of the decision modelling concern from process modelling by providing a Decision as a Service (DaaS) layered Service-Oriented Architecture (SOA) which approaches decisions as externalised services that processes need to invoke on demand in order to obtain the decision outcome. Additionally, the benefits of the DaaS design on process-decision modelling are discussed in terms of scalability, maintainability, flexibility and understandability.

Decision mining in a broader context: an overview of the current landscape and future directions

The term Decision Mining has been put forward in literature to cover numerous applications in a diverse set of contexts. In the business process management community, it typically reflects the way processes and data required for decision purposes in those processes are blended into one model during discovery. However, the upcoming field of decision modeling and management requires the term to be repositioned in order to obtain a better understanding of the interplay of processes and decisions. In this paper, the different approaches that are currently available are delineated and a case is made for a new type of decision mining: one that separates the control flow and decision perspective in a less stringent form compared to existing approaches.

Modeling and Enacting Enterprise Decisions

For years, the capturing of business decisions in enterprise models has not been treated as a separate concern. Rather, decisions were included in business process models or in knowledge models and ontologies. This leaves the overall view of a decision and its interplay with other decision and data requirements dispersed and hard to maintain.

Improving Understandability of Declarative Process Models by Revealing Hidden Dependencies

Declarative process models have become a mature alternative to procedural ones. Instead of focusing on what has to happen, they rather follow an outside-in approach based on a rule base containing different types of constraints. The models are well-capable of representing flexible behavior, as everything that is not forbidden by the constraints in the model is possible during execution. These models, however, are more difficult to comprehend and require a higher mental effort of both the modeler and the reader. Since constraints can be added freely to the model, it is often overseen what impact the combination of them has. This is often referred to as hidden dependencies. This paper proposes a methodology to make these dependencies explicit for the declarative process modeling language Declare by considering a Declare model as a graph and relying on the constraints’ characteristics. Moreover, this paper also contributes by empirically confirming that a tool that can visualize hidden dependency information on top of a Declare model has a significant positive impact on the understandability of Declare models.