Rule-Based Detection of Process Conformance Violations in Application Lifecycle Management

Abstract

Software engineering processes are the basis for the development of quality software products within time and within budget. In this paper we present an approach to detect process conformance violations that reveal deviations between planned and executed software engineering processes. The approach is based on process rules that complement the process documentation. A framework for defining, executing and evaluating these rules has been implemented as extension to an Application Lifecycle Management (ALM) solution. The framework has been applied in context of introducing new processes and practices in an industrial environment. Over the timespan of more than a year, process conformance has been continuously evaluated as part of the nightly build. We were able to demonstrate that the results can be used to identify hot spots in process conformance calling for immediate action, to reveal long-term issues that motivate process improvement measures and, furthermore, that the continuous feedback provided by the approach has a positive impact on process quality. Finally, the paper documents useful lessons we have learned during the application of our approach in a real-world setting.