Optimal Preventive Maintenance Scheduling for Serial Mechanical System with Deteriorated Component Using Genetic Algorithm ─ A Case Study of Paper Facility

• Main Authors: Yen-chang Hsieh, 謝炎璋
• Other Authors: Jyh-Horng Chou
• Format: Others
• Language: zh-TW
• Published: 2004
• Online Access: http://ndltd.ncl.edu.tw/handle/62544807455033998156

碩士 === 國立高雄第一科技大學 === 機械與自動化工程所 === 95 === This paper presents optimal algorithm procedure with improvement genetic algorithm by using hybrid Taguchi genetic algorithm for serial mechanical system with deteriorated component. We propose minimize searching to the cost-effective object function in the periodic maintenance mode which has been constrained by uncertainly maintenance intervals, Then, the optimal preventive maintenance schedule was obtained, and the result is robust and effective. Paper machine facility system in paper mill is used as an example to demonstrate the proposed algorithm. First, collect the historical failure data of the system to construct statical reliability function. Next, in order to descript mechanical system deteriorated behavior during operating period, the statical reliability function is curve fitted to the dynamic reliability function. In addition, we according to the effect that maintenance activities to reliability and hazard to set up system reliability-cost mode and system benefit mathematical model. Finally, under the constraints of reliability after maintenance and uncertainly maintenance intervals, the objective is the minimum cost of time of operating, and plan preventive maintenance activities with preventive maintenance or preventive repair or preventive replacement, and obtain optimum activities-combination schedule.