Opportunistic Maintenance for Multi-Unit Series Systems Based on Gated Recurrent Units Prediction Model

Maintenance plays an important role in sustainable manufacturing as systematic and timely MRO processes can significantly reduce downtime and production losses. This paper presents a two-stage approach for an opportunistic maintenance methodology. Firstly, a real-time gated recurrent units (GRU) model is established to predict the reliability of facilities. Secondly, an ensemble evolutionary algorithm is proposed to identify the feasibilities to conduct different maintenance strategies. In addition to cost and resource availability, energy consumption is also considered as an optimization objective. An industrial case study is used to illustrate the improvements which can be achieved with the proposed methodology.