Condition-based Maintenance Optimization for Gearbox Using Proportional Hazards Model

This paper proposes a condition-based maintenance policy for gearbox with online monitoring information. The bispectrum entropy is chosen as the extracted degradation feature. Effective maintenance decisions based on proportional hazards model is developed, jointly considering the degradation data and the failure time data. The optimal maintenance policy set two control limits, including both hazard rate control limit and age control limit. The optimal maintenance decisions is calculated by minimize the average per unit time cost rate. The method is illustrated by a simulation case. A comparison with other policies proves the effectiveness of the research method.