Quality assessment procedure for IEC 61131-3-based control software for machine and plant manufacturers

Control software realizes an increasing proportion of system functionality in automated production systems (aPS). This requires the reuse of high-quality control software to remain competitive in the global market and deliver evolvable, maintainable and reconfigurable systems. While means for software quality assessment such as static code analysis and software metrics have proven to be beneficial in the computer science domain, they are still not commonly used in the aPS domain. Moreover, the current support for their application to control software is insufficient. This thesis aims to bridge this gap by proposing a quality assessment procedure for control software, including interview guiding questions, an analysis checklist, and means for visualizing and documenting analysis results. It supports control software developers to integrate static analysis means from computer science into their development workflow to assess the software quality and identify potential for its improvement. The procedure enables tailoring the assessment to company-specific boundary conditions by considering the rationale behind the code, deliberate design decisions and domain knowledge. The presented procedure was successfully evaluated with a prototypical implementation, three industrial case studies and two industrial expert groups. The evaluation confirmed the applicability of the developed procedure to real-world projects in an industrial context. It further supports the identification of improvement potential and deriving recommendations for action for overcoming identified weaknesses.