Prediction of drill-bit breakage from degradation signals using Mahalanobis-Taguchi system analysis

Drilling is a widely used machining process. As the hole drilling operation progresses, a drill-bit gradually degrades until it breaks at the end of its life. Replacing a drill-bit after it is breakage can be costly in certain special applications. At the same time an early tool replacement decision may lead to lower tool life utilisation. This calls for methods that enable accurate prediction of tool failures. In such situations degradation signals using appropriate features are utilised to arrive at the tool replacement decision. In this paper we study two such degradation signals viz., thrust force and torque using 20 derived features (10 features per degradation signal). The methodology used involves a successful application of Mahalanobis-Taguchi System (MTS) analysis that enables an accurate prediction of drill-bit breakage ensuring high tool life utilisation.