A method for economic evaluation of predictive maintenance technologies by integrating system dynamics and evolutionary game modelling

Predictive maintenance technologies can be employed for failure prediction and system health management. Nevertheless, the additional cost involved in establishing the predictive maintenance system can be an obstacle to its widespread application. The decision on the predictive maintenance technology adoption can be made through the computation of the return on investment. To investigate the mechanisms of dynamic game between stakeholders involved in predictive maintenance, we establish the SD-EGT model from the perspective of systems engineering. This paper aims to propose an integrated method for the economic evaluation of predictive maintenance technologies by considering the incremental costs and benefits associated with its deployment. As an exemplary case, we take the Lithium-ion batteries whose failures have led to unexpected safety accidents. Firstly, we construct a quantitative relationship model between the failure modes and the predictive benefits of Lithium-ion battery systems to quantify the incremental benefits. Then, we establish a cost-benefit analysis (CBA) model by using system dynamics (SD) to make decisions about cost-effectiveness. Secondly, to optimize the cost investment strategy for the predictive maintenance technology, we develop an enterprise-government evolutionary game model, considering the information asymmetry between players. Eventually, we conduct a sensitivity analysis of the static subsidy strategy. The proposed methodology is serviceable to optimize the decision-making of predictive maintenance technology investment, which is a difficult yet very important task in industrial practice.