Comparing the Applicability of Complexity Measurements for Simulink Models during Integration Testing – An Industrial Case Study

Context: Simulink models are used during softwareintegration testing in the automotive domain on hardware in theloop (HIL) rigs. As the amount of software in the car is increasingcontinuously, the number of Simulink models for control logicand plant models is increasing at the same time.Objective: The study aims for investigating the applicabilityof three approaches for evaluating model complexity in anindustrial setting. Additionally, insights on the understanding ofmaintainability in industry are gathered.Method: Simulink models from two vehicle projects at a Germanpremium car manufacturer are evaluated by applying thefollowing three approaches: Assessing a model’s (a) size, (b)structure, and (c) signal routing. Afterwards, an interview studyis conducted followed by an on-site workshop in order to validatethe findings.Results: The measurements of 65 models resulted in comparabledata for the three measurement approaches. Together with theinterview studies, conclusions were made on how well eachapproach reflects the experts’ opinions. Additionally, it waspossible to get insights on maintainability in an industrial setting.Conclusion: By analyzing the results, differences between thethree measurement approaches were revealed. The interviewsshowed that the expert opinion tends to favor the results of thesimple size measurements over the measurement including thesignal routing