Support vector machine regression analysis of LHD failures

Abstract

Load-haul-dump (LHD) machines are the main material handling equipment in underground mines. Reliability analysis of the machine, in the form of forecasting the time between failures (TBFs), is a reliability index for scheduling, preventive maintenance. Support vector regression (SVR) was employed for reliability analysis of an LHD unit. The LHD’s engine was detected as the most critical system by a Pareto analysis. TBFs of 32 failures were obtained. Twenty-five records were used for SVR modelling and the remaining for testing. Mean absolute percentage error value of less than 2% was achieved, which demonstrated the applicability of SVR.

Keywords:

• support vector machine regression (SVR)
• load-haul-dump (LHD)
• reliability analysis