Electricity Consumption Forecasting Using the Prophet Model in Industry: A Case Study

Forecasting electricity consumption is a key mechanism for a wide range of industries to develop strategies or take precautions. This case study primarily aims to predict the electricity consumption of the machines on the production line through sensors and analyzers containing both thermocouples and devices storing the use of electricity. To do so, one of the most powerful methods, which is known for its ability to learn the main characteristics of the data for time-series models, the Prophet method has been utilized. Moreover, the capability of the Prophet method relying on both the use and not the use of temperature data in forecasting electricity consumption has been discussed. The achievement of the method has been supported through the tables and figures in both univariate and multivariate cases. Comparing the RMSE, MAE, and SMAPE scores, the results have shown that the not use of temperature data has been better than those of the use of temperature in the prediction.