Prof. PhD. Marlon Dumas (University of Tartu, Estonia)
Prof. PhD. John Krogstie (Faculty of Information Technology and Electrical Engineering, Norwegian University of Science and Technology Trondheim, Norway)
Prof. PhD. Jaak Tepandi (School of Information Technologies, Tallinn University of Technology Tallinn, Estonia)
A picture is worth a thousand words, but a few words can greatly enhance a picture. It is common to find textual and diagrammatic components complementing each other in enterprise models in general, and business process models in particular. Previous studies have considered the question of the relative understandability of diagrammatic versus textual representations of process models for different types of users. However, the effect of combining textual and diagrammatic components on the actual use and reuse of process models has received little attention. In this setting, this thesis approaches the questions of: (i) how do structured and unstructured components of process models co-exist with each other in practice; and (ii) what determines that a process model is used on a sustained basis within an organization? These two questions are first approached separately, and then jointly, via three complementary studies. The first study focuses on the co-existence of different types of process documentation structures within an organization. Specifically, this study proposes a framework, namely the Process Documentation Cube (PDC) for mapping and assessing business process documentation with the aim of identifying gaps and potential inconsistencies. The PDC framework is built on the principle that documentation should exist in an organization at different levels of detail, across different organizational areas and in different formats, ranging from unstructured text to structured artifacts such as tables and diagrams. Accordingly, the PDC framework combines three assessment dimensions: granularity, organizational area and structure. The suitability of PDC to support process documentation mapping and planning efforts is validated via a series of case studies. The second study focuses on the identification of factors that affect the sustained use of process models in an organization. First, the study puts forward a priori model of sustained process model use derived from existing factor models of business process modelling success and reuse. This model is packaged as an assessment instrument and applied to four organizations from different domains. Based on these case studies, a subset of factors and relationships are identified, which collectively explain differences in the observed sustained use of process models across the organizations in question. The third study bridges across the two above-mentioned questions. Specifically, this study addresses the question of how the mix of diagrammatic and textual components in business process models affects their sustained use. This question is approached by means of a case study in a telecommunications company where models with different mixtures of text and diagrams have been collected over time. The study shows that models where diagrams are used to capture ordering relations between activities at a granular level, while text is used at a more detailed level, are more likely to be used on a sustained basis. The latter finding emphasizes the potential benefits of considering different documentation formats not as alternative, but as complementary parts of a whole.