Prediction of Sugar Production using Multivariate Linear Regression

Sugar is a widely consumed commodity all over the world. India produced 33 million metric tons of sugar in 2018-2019, making it the world's largest sugar producer. Sugar production is affected by a variety of factors, including the region under sugar cane, sugar cane production, cane yield per hectare, the number of sugar factories in operation, and the total amount of sugar cane crushed. The current paper, titled "Prediction of Sugar Production Using Multivariate Linear Regression," focuses on sugar production forecasting. Forecasting allows you to keep track of your market by foreseeing challenges and opportunities. Forecasting accuracy is just as important for a company as it is for a person. To render the model using multivariate linear regression for predicting total sugar production, time-series data of sugar production from 1931 to 2018 was used. Cooperative Sugar, Vol. 51, No. 6, February 2020, is the source of the knowledge. The model was trained with 80% of the data and tested with 20% of the data. The association between different parameters is determined using a heatmap. For predicting total sugar output, a multivariate linear regression model is the best fit.