Quality Assessment of Business Process Models Based on Thresholds

Process improvement is recognized as the main benefit of process modelling initiatives. Quality considerations are important when conducting a process modelling project. While the early stage of business process design might not be the most expensive ones, they tend to have the highest impact on the benefits and costs of the implemented business processes. In this context, quality assurance of the models has become a significant objective. An important step towards improved quality assurance is a precise assessment of quality. We analyze quality from the perspective of understandability and modifiability, which are both sub-characteristics of usability and maintainability, respectively. Several initiatives about business process metrics were published [2]. The significance of these metrics relies on a thorough empirical validation of their connection with quality attributes. There are, to date, still rather few initiatives to investigate the connection between structural process model metrics and quality characteristics , so we detect a gap in this area which needs more empirical research. Experimental data is obtained by a total of six experiments: three to evaluate understanda-bility and three to evaluate modifiability. In accordance with the previously identified issues, the purpose of this paper is to contribute to the maturity of measuring business process models. The aim of our empirical research approach is to validate the connections between an extensive set of metrics (number of nodes, diameter, density, coefficient of connectivity, average gateway degree, maximum gateway degree, separability, sequentiality, depth, gateway mismatch, gateway heterogeneity, cyclicity and concurrency) and the ease with which business process models can be understood (understandability) and modified (modifiability). A correlation analysis and a regression estimation were applied in order to test the connection between the metrics and both the under-standability and modifiability of the models. After the selection of the most suitable metrics for understandability and modifiability, we extracted threshold values in order to evaluate the measurement results. Such thresholds are an important aid to support the modeller of a business process. The statistical analyses suggest rejecting the null hypothesis (H0: There is no correlation between structural metrics and understandability and modifiability), since the structural metrics apparently seem to be closely connected with under-standability and modifiability. For understandability these include Number of