A case study on the application of predictive analytics toward forecasting swing door failure

Predictive maintenance is a maintenance approach that involves monitoring machines in order to predict their failures. This case study focuses on predicting failure of swing doors employed in a facility and scheduling maintenance based on the predicted failure date. Swing doors in a facility are checked regularly. Problems such as a door stays open only for a short period of time, a door opens only after a hard push act as signals that some larger problem may be growing. With this information, maintenance can be scheduled before the door goes out of service. Hold on Time (HoT), which is the specified time during which a swing door is held open, is collected for all swing doors in a facility. By closely following the trend exhibited by hold on time readings observed overtime, abnormal operations could be predicted beforehand, and maintenance can be carried out before any doors fail. Piecewise regression is the technique adopted for prediction and is implemented using R.