Understanding the occurrence of errors in process models based on metrics

  • Authors:

  • Jan Mendling;Gustaf Neumann;Wil Van Der Aalst

  • Affiliations:

  • BPM Cluster, Faculty of Information Technology, Queensland University of Technology, Australia;Institute of Information Systems and New Media, Vienna University of Economics and Business Administration, Austria;Department of Computer Science, Eindhoven University of Technology, The Netherlands

  • Venue:
  • OTM'07 Proceedings of the 2007 OTM Confederated international conference on On the move to meaningful internet systems: CoopIS, DOA, ODBASE, GADA, and IS - Volume Part I
  • Year:
  • 2007

Quantified Score

Hi-index 0.02

Visualization

Abstract

Business process models play an important role for the management, design, and improvement of process organizations and process-aware information systems. Despite the extensive application of process modeling in practice, there are hardly empirical results available on quality aspects of process models. This paper aims to advance the understanding of this matter by analyzing the connection between formal errors (such as deadlocks) and a set of metrics that capture various structural and behavioral aspects of a process model. In particular, we discuss the theoretical connection between errors and metrics, and provide a comprehensive validation based on an extensive sample of EPC process models from practice. Furthermore, we investigate the capability of the metrics to predict errors in a second independent sample of models. The high explanatory power of the metrics has considerable consequences for the design of future modeling guidelines and modeling tools.