Mathematical model of condition-based preventive maintenance of a complex technical system

Abstract

A new approach to mathematical modeling of complex technical systems according to their condition is being developed. Connections between subsystems of a complex system are considered to be arbitrary in terms of reliability. Due to wear, failures of subsystems can happen at random moments of time. Failures of some subsystems can lead to the entire system failure. The purpose of the simulation is to maintain the level of reliability and operability of a complex technical system at an optimal level for an unlimited time interval by means of regular preventive maintenance and repair. Technical instructions and specifications, as well as statistical data, are used in modeling a priori characteristics of subsystems. That information is used to determine the reliability of a complex system and its condition. The mathematical model is built in terms of the Markov decision-making process. The chosen optimization method allows obtaining the best policy for choosing acceptable preventive maintenance policy and repairs at the planned time of inspections and moments of failures.