Declarative approaches to business process modeling are regarded as well suited for highly volatile environments, as they enable a high degree of flexibility. However, problems in understanding and maintaining declarative process models often impede their adoption. Likewise, little research has been conducted into the understanding of declarative process models. This paper takes a first step toward addressing this fundamental question and reports on an empirical investigation consisting of an exploratory study and a follow-up study focusing on the system analysts' sense-making of declarative process models that are specified in Declare. For this purpose, we distributed real-world Declare models to the participating subjects and asked them to describe the illustrated process and to perform a series of sense-making tasks. The results of our studies indicate that two main strategies for reading Declare models exist: either considering the execution order of the activities in the process model, or orienting by the layout of the process model. In addition, the results indicate that single constraints can be handled well by most subjects, while combinations of constraints pose significant challenges. Moreover, the study revealed that aspects that are similar in both imperative and declarative process modeling languages at a graphical level, while having different semantics, cause considerable troubles. This research not only helps guiding the future development of tools for supporting system analysts, but also gives advice on the design of declarative process modeling notations and points out typical pitfalls to teachers and educators of future systems analysts.

Hierarchy has widely been recognized as a viable approach to deal with the complexity of conceptual models. For instance, in declarative business process models, hierarchy is realized by sub-processes. While technical implementations of declarative sub-processes exist, their application, semantics, and the resulting impact on understandability are less understood yet-this research gap is addressed in this work. More specifically, we discuss the semantics and the application of hierarchy and show how subprocesses enhance the expressiveness of declarative modeling languages. Then, we turn to the influence of hierarchy on the understandability of declarative process models. In particular, we present a cognitive-psychology-based framework that allows to assess the impact of hierarchy on the understandability of a declarative process model. To empirically Communicated by Dr. Selmin Nurcan. test the proposed framework, a combination of quantitative and qualitative research methods is followed. While statistical tests provide numerical evidence, think-aloud protocols give insights into the reasoning processes taking place when reading declarative process models.

Even though considerable progress regarding the technical perspective on modeling and supporting business processes has been achieved, it appears that the human perspective is still often left aside. In particular, we do not have an in-depth understanding of how process models are inspected by humans, what strategies are taken, what challenges arise, and what cognitive processes are involved. This paper contributes toward such an understanding and reports an exploratory study investigating how humans identify and classify quality issues in BPMN process models. Providing preliminary answers to initial research questions, we also indicate other research questions that can be investigated using this approach. Our qualitative analysis shows that humans adapt different strategies on how to identify quality issues. In addition, we observed several challenges appearing when humans inspect process models. Finally, we present different manners in which classification of quality issues was addressed.

No abstract

Declarative approaches to process modeling are regarded well suited for highly volatile environments as they provide a high degree of flexibility. However, problems in understanding and maintaining declarative process models impede their usage. To compensate for these shortcomings, Test Driven Modeling (TDM) has been proposed. This paper reports on an empirical investigation in which TDM is viewed from two different angles. First, the impact of TDM on communication is explored in a case study. Results indicate that domain experts are inclined to use test cases for communicating with the model builder (system analyst) and prefer them over the process model. The second part of the investigation, a controlled experiment, investigates the impact of TDM on process model maintenance. Data gathered in this experiment indicates that the adoption of test cases significantly lowers cognitive load and increases the perceived quality of changes.

Abstract. Declarative approaches to business process modeling are regarded as well suited for highly volatile environments, as they enable a high degree of flexibility. However, problems in understanding declarative process models often impede their adoption. Particularly, a study revealed that aspects that are present in both imperative and declarative process modeling languages at a graphical level-while having different semantics-cause considerable troubles. In this work we investigate whether a notation that does not contain graphical lookalikes, i.e., a textual notation, can help to avoid this problem. Even though a textual representation does not suffer from lookalikes, in our empirical study it performed worse in terms of error rate, duration and mental effort, as the textual representation forces the reader to mentally merge the textual information. Likewise, subjects themselves expressed that the graphical representation is easier to understand.

Abstract. Considerable progress regarding impact factors of process model understandability has been achieved. For example, it has been shown that layout features of process models have an effect on model understandability. Even so, it appears that our knowledge about the modeler's behavior regarding the layout of a model is very limited. In particular, research focuses on the end product or the outcome of the process modeling act rather than the act itself. This paper extends existing research by opening this black box and introducing an enhanced technique enabling the visual analysis of the modeler's behavior towards layout. We demonstrate examples showing that our approach provides valuable insights to better understand and support the creation of process models. Additionally, we sketch challenges impeding this support for future research.

No abstract