Prediction of Equipment Spare Parts Consumption Based on ElasticNet and Markov Chain Models

By analyzing the influencing factors of equipment spare parts consumption, dividing the historical consumption data into training set and test set, using the training set to establish ElasticNet regression model, making preliminary prediction of equipment spare parts consumption, and then using Markov chain, correcting the prediction results, the calculations show that the average relative error of this method on the prediction of both the training set and the test set has been reduced substantially, and the average relative error of the prediction of the training set and the test set has been reduced. It improves the prediction accuracy of equipment spare parts consumption, which is of great guiding significance for improving the quality of equipment protection.