Proper prediction of electricity consumption is one of the most important parameters in planning for a system. One of the most common tools for forecasting is regression-based methods. In this research, using the multiple linear regression model, the prediction of electricity consumption in Mashhad fire stations has been discussed. The statistical population of the current research is the stations of the Mashhad Fire Department, of which 45 stations were selected as a statistical sample.

Pearson correlation and regression statistical methods were used to analyze the research data. The results of the research show that there is a strong and direct correlation between the three independent variables of the number of firemen stationed at the station, the area of ​​the station and the number of fire trucks with the dependent variable of the station's annual electricity consumption. Finally, by using the multiple regression model, a model was obtained to calculate the annual electricity consumption based on the mentioned independent variables.

In order to confirm the obtained model, it was shown by using ridge regression analysis that despite the presence of strong collinearity between independent variables, it did not affect the estimation of coefficients of the multiple regression model and the coefficients of the ridge regression model were very close and equal to the normal multiple regression coefficients. 

The results of this research can be used to predict the amount of the annual budget of electricity consumed by fire stations and to optimize and modify the consumption pattern.