Optimization of preventive repair in a dynamic system of machines

Abstract

The determination of the number of preventive repair to make in a park of machines in a given period, in the majority of practical cases, is based on the knowledge of essential parameters obtained from the statistics such as: 1) Mean life duration of the equipment 2) Mean functioning duration until the first failure 3) Mean functioning duration between the failures 4) renewing intensity of the pa k of machines. However theses values cannot be taken as credible base data to be able to optimise the number of repair of machines in a dynamic exploitation system of a park of machines and obtain the wanted number of machines at the end of the planned period. The cited 4 parameters depend on 2 principal factors: • Evolution and importance of the park of machines (variable in time) • Exploitation methods in the park (repair quality and used means). The renewing intensity of the park of machines is a function of the demand and degradation intensity of machines. From another side, the repair duration depend on the functioning periodicity cited here, on their nature and their costs. These last influence, in a significant manner, the calculus results. It is then primordial to look for the solution of the problem by introducing the notion of parameters optimum values. In this paper is given a new approach of the problem, which studies the dependence of the cited parameters, of the exploitation time and of the importance of the park of machines always renewed. This has for effect to act directly on the costs due to this dependence that is: • The renewing costs (purchase of new machines and new repair means); • The exploitation and repair costs. This method permits to obtain optimal values: • Of the number of repairs; • Of the number of machines to purchase in the planned period; • Of the exploitation duration of every machine