Measuring Scalable Business Process Model Complexity Based on Basic Control Structure

The business process model is a representation of business activities illustrated through diagram notations. This model is composed of repeated specific patterns called basic control structure. Each basic control structure has a level of complexity. The metrics for formulating existing complexity are very diverse, but can only define complexity partially and are less sensitive to small changes in the structure of the business process model. In this paper, we propose a formula of complexity metric that can indicate small changes in structure, type of branching logic, number of branches, loops, and depth. We call it the Yaqin complexity formula. To get the Yaqin complexity formula, we carried out several activities. These activities are identifying the metrics involved using the Goal-Question-Metric (GQM) method, then formulating the metric complexity, the next activity is testing the formula with several business process models and analyzing the test results, and then validating the Yaqin complexity formula using the Weyuker's properties framework. The Yaqin complexity formula, which involves seven parameters, is proven to be more comprehensive than other complexity formulas that involve less than seven parameters. The Yaqin complexity formula also proved to be more sensitive to other complexity formulas, where 7 out of 8 cases affected the Yaqin complexity metric. The validation results state that the Yaqin complexity formula meets 8 of 9 Weyuker's properties. Thus we have succeeded in formulating the Yaqin complexity, which is more comprehensive in involving parameters and more sensitive in detecting small changes in the structure of the business process model.