Technical Debt Identification and assessment in mechatronic systems applying indicators, patterns, and metrics - TDebituM

• Applicant: Professorin Dr.-Ing. Birgit Vogel-Heuser
• Subject Area: Software Engineering and Programming Languages ; Automation, Mechatronics, Control Systems, Intelligent Technical Systems, Robotics
• Term: from 2021 to 2024
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 451267453

Technical decisions that are expedient in the short-term, yet prove as disadvantageous and expensive in the long-term are often made, as their range of consequences, impact, and corrective measures are not considered at all or underestimated. This phenomenon is described as Technical Debt (TD) in the domain of software systems. The few research results on TD in mechatronics more precisely automated production systems already show in comparison to software and embedded systems that at first new additional TD types can be observed in all of the three mechatronic disciplines even in application software and its platforms, second TD types that result from the interdisciplinarity of mechatronics and third from the additional life cycle phases of mechatronic systems.TDebituM therefore systematically analysis TD in mechatronic systems for both automated production systems, more precisely machine and plant manufacturing, and mechatronic products in the automotive industry representing the two German key industries investigating, for example, the following research questions: - Which additional TD types and subtypes can be identified in comparison to software systems, in which phase of the mechatronic systems’ life cycle and with which frequencies, causes, and consequences do they occur?- Which indicators, patterns, and metrics can be identified for such mechatronic TD or TD in mechanical, electrical engineering as well as application software and its platforms?- Which source effect relations can be observed?Using expert interviews in German and selected international industrial companies, TD incidents in the field of mechatronic systems as mentioned above are collected. Statistical analysis methods are applied to examine the classified TD properties for occurrence, criticality, significance, and correlation. Identified static TD patterns and sequences are used to define TD metrics quantitatively. A concept developed in an iterative way with experts of different foci serves the visualization of, and interaction with TD and its communication in the company itself as well as in its supply chain. The validity of the results is proven by an evaluation of the results with individual experts as well as focus groups in at least 10% of the interviewed companies.TDebituM explores the deliberate and, as far as possible, the inadvertent TD incidents and extends the existing knowledge for software and embedded systems by considering the mechatronic disciplines, additional phases in the life cycle, interdisciplinary dependencies, and the resulting problem shifting. Early detection of potential TD incidents and a transparent presentation of the interrelationships, dimensions, and risks support the technical decision-making process along the life cycle of complex mechatronic systems and lay the foundation for all further TD management activities.

• DFG Programme: Research Grants