Full realization of all potentials in predictive and proactive maintenance highly depends on the accuracy of long-term predictions of the remaining useful life of manufacturing equipment. Parametric linear prediction techniques, such as Autoregressive Moving Average modeling (ARMA), are routinely used to trend and predict future behavior of any time series, but are frequently not appropriate for long-term prediction because of the highly complicated and non-stationary nature of manufacturing processes. In this paper, we propose a novel method that is capable of achieving high long-term prediction accuracy by comparing signatures from two degradation processes using measures of similarity that form a Match Matrix. Through this concept, we can effectively include large amounts of historical information into the prediction of the current degradation process. Similarities with historical records are used to generate possible future distributions of features, which is then used to predict probabilities of failure over time by evaluating overlaps between predicted feature distributions and feature distributions related to unacceptable equipment behavior. Experimental results show that the proposed method results in a significant improvement of long-term prediction accuracy compared with ARMA modeling-based prediction.
Forecast number of new cases of Corona Virus Disease (COVID-19) in Ethiopia, using the case-based autoregressive integrated moving average model
