Quality assurance is one of the most critical activities in business process models which are obtained by reverse engineering, e.g., from existing information systems. Companies must deal with several quality faults in business process models such as irrelevant elements, fine-grain granularity or incompleteness, which affect understandability and modifiability of business process models. Hence, business process refactoring techniques are often used to improve these features, which change the internal structure of business process models while its external behavior is preserved. Unfortunately, different refactoring operators do not fulfill commutative property among them. For this reason, this paper addresses the challenge of establishing the best order in which to apply all the different refactoring operators and, therefore, to achieve the highest quality improvement. The research methodology consists of conducting a real-life case study to assess the influence of the refactoring operator's order in the understandability and modifiability of business process models. The case study demonstrates that there is a clear influence in these quality features in terms of the size and separability of the business process models under study.

No abstract

Modernization of legacy information systems is usually triggered by the need of introducing new business requirements or due to the technology obsolescence. During modernization software projects, there exists a lot of business knowledge that was embedded in source code owing to progressive maintenance, which is not present anywhere else. In order to preserve embedded business knowledge this paper presents MARBLE, a tool to recover business processes from legacy information systems. MARBLE provides an extensible framework, thus it is developed as an Eclipse™ plugin to ensure its future extension. So far, MARBLE supports Java-based system and focuses mainly on legacy source code. To facilitate its adoption in the industry, MARBLE has been applied in some real-life modernization projects, which demonstrated that MARBLE is suitable to retrieve business processes and facilitated its continuous improvement to reach an appropriate maturity level.

No abstract