A. Nouri Qarahasanlou - Faculty of Technical & Engineering, Imam Khomeini International University, Qazvin, Iran
M. Ataei - Faculty of Mining, Petroleum and Geophysics Engineering, Shahrood University of Technology, Shahrood, Iran
R. Shakoor Shahabi - Faculty of Technical & Engineering, Imam Khomeini International University, Qazvin, Iran

Whether directly in the form of expenses or indirectly, the objective of maintenance in the mining industry is self-evident in time losses and loss of production. In this paper, the reliability-based maintenance is examined with a different insight than before. The system goes back to the Good As New (GAN) state or too Bad As Old (BAO) maintenance state; why so, the maintenance of the system shifts to the midrange state. On the other hand, the implementation of repairs is strongly influenced by the environmental factors that are known as the “risk factors”. Therefore, an analysis requires a model that integrates two basic elements: (۱) incompleteness of the maintenance effect and (۲) risk factors. Thus, an extensive proportional hazard ratio model (EPHM) is used as a combination of the Proportional Hazard Model (PHM) and the Hybrid Imperfect Preventive Maintenance model (HIPM) in order to analyze these elements. In this regards, four different preventive maintenance strategies are proposed. All four strategies are time-based including constant interval or periodic (the first and second strategies) and cyclic interval (the third and fourth strategies). The proposed method is applied for a Komatsu HD۷۸۵-۵ dump-truck in the Songun copper mine as a case study. The PM intervals with a mean value of risk factors for the four activities to reach the ۸۰% reliability for the first and second strategies are about ۵ and ۴۸ hours. These intervals for the third strategy are calculated as ۴۸.۳۶, ۱۱.۵۸, ۱۰.۲۵, and ۹.۰۳۵, and for the fourth strategy are ۵.۰۶, ۴.۰۷۸, ۳.۴۵۹, and ۱.۹۲.

Reliability, Maintenance, Expected Proportional Hazard Ratio Model (Ex-PHM), Proportional Hazard Model (PHM), Hybrid Imperfect PM Model (HIPM)