Software Process Analysis Methodology - A Methodology Based on Lessons Learned in Embracing Legacy Software

Maikel Leemans, Wil M. P. van der Aalst, Mark G. J. van den Brand, Ramon R. H. Schiffelers, Leonard Lensink

Abstract: Over the last decades, the complexity of high-tech systems, and the software systems controlling them, has increased considerably. In practice, it is hard to keep knowledge and documentation of these ever-evolving software systems up-to-date with their actual realization; we are dealing with legacy software. Clearly, this lack of knowledge, insight, and understanding is more and more becoming a critical issue. Process mining provides an interesting opportunity to improve understanding and analyze software behavior based on observations from the system on the run. However, a concrete software process analysis methodology was lacking. This paper 1) discusses a software process analysis case study at ASML, a large high-tech company, and, based on the lessons learned, 2) presents a concrete methodology for analyzing software processes. The presented methodology actively includes the system under analysis and is based on practical experiences in applying process mining on industrial-scale legacy software.