A Machine Learning Approach to Electricity Usage and Billing Trends in Chittagong: Population Dynamics, Industrial Expansion, and Sustainable Energy Planning

Chittagong, the economic center and primary port city of Bangladesh, is experiencing swift urbanization and industrialization, leading to a substantial increase in electricity demand. This study examines the correlation between population growth, industrial development, and infrastructure expansion in relation to electricity consumption and billing trends across four operational zones-East, West, North, and South—governed by the Bangladesh Power Development Board (BPDB). Data on electricity consumption and billing from 2007 to 2022, alongside population statistics from 1991 to 2022, were examined utilizing machine learning models such as Random Forest, Support Vector Machine (SVM), Logistic Regression, KNearest Neighbors, and Gradient Boosting to predict trends in electricity consumption and billing. The Random Forest model exhibited performance, attaining up to 94 % accuracy in predicting electricity consumption trends and billing trends, especially in the East circle. Findings indicate significant regional disparities: the East and West circles demonstrate the most pronounced growth in electricity consumption due to accelerated industrialization and population growth, whereas the North circle exhibits emerging demand potential. Conversely, the South circle exhibits a decrease in consumption, indicating diminished industrial activity. These insights provide essential assistance for policymakers and energy planners to formulate sustainable energy plans suited to the evolving urban and industrial environment of Chittagong.