Optimal Tool Replacement Decisions Using Wear Data and Proportional Hazard Model  
Author Biao Lu


Co-Author(s) Dehao Zhou


Abstract Effective tool replacement decisions are crucial for the reliable and smooth production. Existing tool replacement models mostly using the hazard function which ignores the impact of tool wear on its failure risk. In fact, tool wear can have a strong impact on its failure risk. Therefore, it is necessary to integrate tool wear into its failure rate model to accurately assess its failure probability. In this regard, this paper develops an integrated hazard function for tools based on the proportional hazard model with tool wear integrated as a covariate. The parameters of the integrated hazard function are estimated using both tool wear data and failure time data. The integrated hazard function is further used to construct a cost model based on the renewal reward criterial for finding optimal replacement time. The proposed model is applied to milling tools in the mill wear experiments. Results show that ignoring the impact of tool wear on its failure risk will lead to underestimation of tool failure rate and consequently lead to too late replacement. The integrated hazard function can accurately evaluate tool failure rate and help make practical replacement decisions.


Keywords tool replacement, proportional hazard model, tool wear, wear data, gamma process
    Article #:  RQD26-68

Proceedings of 26th ISSAT International Conference on Reliability & Quality in Design
Virtual Event

August 5-7, 2021