Intelligent maintenance and reliability optimization for mission-oriented systems under sustainability considerations

Modern life depends upon complex and highly interconnected production and distribution systems to provide energy, goods, and services. Some of these systems called mission-oriented systems (MOS) are typically required by design to run consecutive missions interspersed with scheduled maintenance breaks. MOS include systems such as production lines, military and civilian aircraft, offshore wind turbines, drones, and large petrochemical facilities. Monitoring the health and degradation level of these systems, predicting failure occurrences, and proactively maintaining these assets is crucial to meeting the required performance of the MOS. Selective maintenance (SM) is an innovative maintenance strategy applied to these MOS aiming to find an optimal list of maintenance actions to be performed on a subset of components to maximize the system reliability for the upcoming mission(s) or minimize costs and/or carbon emissions. In a context of increased competition and uncertainty, performance gains can ensure competitiveness, profitability, and increased consumer satisfaction. This research program focuses on modelling and optimizing the performance of MOS under uncertainty related to components’ degradation, operating conditions, labor, demand, fluctuating capacity and other disruptions in the context of Industry 5.0 and decarbonization. The three intertwined and complementary themes addressed in this proposal cover several of the 17 United Nations’ Sustainable Development Goals. The first theme deals with the development of advanced reliability and maintenance assurance decision support tools for large-scale mission-oriented systems (MOS) using novel blends of proportional hazard and machine learning models. Applications to offshore wind and tidal energy farms which hold great economic potential for Atlantic Canada will be investigated. The second theme deals with the development of integrated production, quality and condition-based maintenance optimization for stochastically deteriorating manufacturing MOS with human-centric considerations (Industry 5.0). The third theme of this research proposal aims to develop a predictive remanufacturing framework for MOS using novel models, innovative algorithms, and decision support tools to assist Extended Producer Responsibility (EPR) activities. The development and testing of our models and deep learning algorithms will be based on large datasets from our industrial partners. We will develop cutting-edge maintenance and reliability decision support tools, and train environmentally-conscious graduates. Novel optimization models and intelligent algorithms will be developed and used to support the operation of resilient and sustainable next-generation assets such as wind turbines and autonomous vehicles, and machinery in smart factories. This work will benefit the numerous Canadian companies and organizations with operations in the manufacturing, distribution, remanufacturing, and maintenance industries.