Failure probabilistic model of CNC lathe based on support vector machine

In this paper, support vector machine (SVM) have been applied to the failure probabilistic model of CNC lathe, and two methodologies (SVM and least square method) have been compared for the analysis of the failure data collected from eighty CNC lathes. The proposed failure probabilistic model based on SVM was more accurate and reliable than that of least square. Hence, the more accurate mean time between failures (MTBF) and failure rate (λ(t)) have been obtained, which aid to deeply understand the reliability status of CNC lathe.