Abstract
Most Industrial production systems are subject to deterioration with usage and age. Such system deterioration may lead to higher production costs and lower product quality and increase possibility of breakdown. Maintenance management plays a critical role in preventing deterioration and failures of the system. Executing maintenance management in the production systems will promote the system availability. There are two aspects on increasing system availability: operators are trained to decrease the human error rate, and the other way, the equipment will be designed to tolerate the slight human error and recover it in time. There are two objectives of this study, 1) A stochastic model of the maintenance management would be developed. The maintenance management model was applied to construct the relationships among maintenance cycle, maintenance personnel allocation, human recovery factor, and system’s tolerance time. In addition, a simulated experiment was conducted to find out the supplementary parameters such as individual latent human error, individual critical human error, recovery rate, and system’s tolerance time. 2) This model would be applied to an aircraft maintenance management decision support system to help maintenance managers to accomplish tasks. In application, maintenance management records of the a local Airline were used and a aircraft maintenance management decision support system （AMMDSS） was structured for the maintenance management department of the airline company to provide maintenance management information including total maintenance cost, standard maintenance task process, the optimal combination of the maintenance personnel, and the reliability of the maintenance management. Being limited to the conditions of maintenance management model and aircraft maintenance record, the parameters of maintenance model can’t be used to AMMDSS completely. However, the concepts of recovering human error, locating maintenance personnel, and setting maintenance cycle will be designed into AMMDSS. User subjectively validate the system be helpful to maintenance management.