Provide Proactive maintenance model using Markov Decision Process Based on stochastic dynamic planning

The use of new approaches and technologies in the design of maintenance plans and policies, taking into account the changes and growth of technology, is one of the factors that are increasingly considered by experts today, and in addition to reducing costs Repair, reducing downtime of machines and increasing the useful life of machinery and equipment, create a competitive advantage in production systems. The purpose of this study is to provide a model for realizing a proactive maintenance approach. In this regard, the decision-making problem for "selecting maintenance policies and programs at the optimal time with the lowest cost" has been modeled. For conducting research, historical data related to the implementation of maintenance programs and risk assessment in the gas pipeline network have been used, and based on recursive induction in stochastic dynamic planning with Markov decision-making process in finite time, mathematical model is designed. In this research, to assign each of the specified policies and actions related to the maintenance program to the identified risks and defects, simulations and optimizations based on time and cost have been performed and sensitivity analysis and validation of the model are performed. The rate of improvement and the rate of optimization in the cost of implementing maintenance policies and the time of its implementation, indicate the efficiency of the proposed model.