Taking Complexity into Account: A Structured Literature Review on Multi-component Systems in the Context of Predictive Maintenance

Decision support for multi-component systems: visualizing interdependencies for predictive maintenance

“…MCS view in the context of PdM helps to improve predictive results, as well as understanding and interpretation of the predictions, thus, enhancing maintenance scheduling and decision support [9,4]. Current research has shown that interdependencies between components exist and have a strong impact on the deterioration process, as well as on the economical aspect [14,4,9].…”

“…Initially, this works focuses on a single component system (e.g., system-level or a component) providing acceptable results. However, the complexity of systems in digitized manufacturing increases due to mass customization and shorter product life cycles in digitized manufacturing [2], or the presence of different variants and configurations within one machine [4]. This leads to the need for new solutions to handle the complexity and its challenges properly.…”

“…MCS view in context of PdM seems promising in this regard. However, in the past, the lack of data and the degree of complexity present within MCSs prevented sufficient research on the implementation and application of MCS perspective [4]. Nevertheless, with the increase of the information basis (growing availability of cheap and powerful sensor technology [21]) and huge advances in AI and data processing capabilities [22], research in MCSs seems feasible.…”

“…This is mainly as a result of mass customization, more complex production processes, and shorter product life cycles in digitized manufacturing [2]. Traditional PdM approaches consider even very complex industrial systems on a single component level (e.g., system-level or a component) without paying enough attention to the interdependencies of components [4]. MCS seem promising in this regard as they explicitly consider interdependencies between system components and thus promise more accurate and understandable models [5,4].…”

“…In this case, a type of interdependencies (i.e., interactions) is deterioration effect between components. Modeling and interpreting interdependencies provide many advantages, such as more accurate results, higher interpretability, higher acceptance rate of these models, and finally improved maintenance scheduling strategies [5,7,4]. Until now some initial research attempting to improve the decision-making process for PdM in the context of MCSs is conducted [7,5].…”