Research On Availability And Maintenance Policy Under Imperfect Repair

Maintenance is a pivotal element which could factor into the development of industry, military technology, aviation and other fields, and the reliability of maintenance and replacement policy are the indices of the greatest concern among equipment users. Accordingly, it is always the focus of studies. So far imperfect maintenance has been gradually accepted; and how to describe the status of equipment not as good as new after maintenance as well as how to determine the availability and replacement policy of system is of great practical significance for equipment users.On the premise of imperfect maintenance environment, the present thesis examines the availability and replacement policy of equipment by use of Quasi-Renewal theories and methods, based on which a model on stationary availability and replacement policy is built and a software supporting availability and replacement policy is developed. The detailed work is listed as follows:First of all, an imperfect maintenance model is built, and Quasi-Renewal theories are applied to the description of imperfect maintenance process.Moreover, instantaneous availability of system with imperfect maintenance is analysed, especially that in the exponential distribution assumption.Later, the imperfect maintenance process of one-component system is descripted with Quasi-Renewal methods, and a model of stationary availability is presented. Besides, the maintenance costs of per unit time are solved, grounded upon which a replacement policy model is built to get the maximal long-run expected reward per unit time. These models provide information on equipment status, maintenance costs and replacement policy for equipment users.Finally, on the software platform of EISIS(Equipment Integrated Support Information System), the stationary availability and strategy models which have been discussed are designed to achieve, so the system supporting availability and replacement policy is developed. Thereafter, some numerical examples are presented to demonstrate the model proposed in the thesis, and information on stationary availability, costs and policy is offered as reference for management departments.