Making Sense of Declarative Process Models

This case study is examining reading strategies in declarative business process models. In particular, we elaborated declarative business process models with a ‘normal’ layout and with a horizontally and vertically mirrored layout. Participating subjects had then to describe the model and answer questions about particular aspects of the models. To analyse the answers, we use questionnaires and think-aloud technique.

Material

We showed the participating subjects two models, for both models there exists a variant with a normal and one with a mirrored layout. We alternated not only the layout variants of the declarative process models, but also the order of the two models, resulting in four experimental groups:

Group 1: process model 1 (normal layout), process model 2 (mirrored layout) [see questionnaire here]
Group 2: process model 1 (mirrored layout), process model 2 (normal layout) [see questionnaire here]
Group 3: process model 2 (mirrored layout), process model 1 (normal layout) [see questionnaire here]
Group 4: process model 2 (normal layout), process model 1 (mirrored layout) [see questionnaire here]

To support subjects during the study and to avoid that constraints were misunderstood, subjects received this handout and this introduction before the experiment, and this sheet with constraints during the experiment.

Model Characteristics

Process Model 1 (P5) Process model 2 (P6)
Activities 12 24
Constraints 18 25
Constraint types 8 7
Components 3 6
Domain Software Development Electronic Company

Data

The demographics of the participating subjects can be found here, the results of this study can be found here (ods format).

Publications

  • C. Haisjackl, I. Barba, S. Zugal, P. Soffer, I. Hadar, M. Reichert, J. Pinggera and B. Weber: Understanding Declare models: strategies, pitfalls, empirical results. Software & Systems Modeling 15(2):325–352, 2016.