Improving E-Bike Trip Efficiency through perfect duration and Battery Consumption Prediction Using Machine Learning Techniques and Power BI Analytics

Abstract

Global warming and rise in temperature in cities due to vehicle pollution are the main causes to increase the usage of pollution free Electrical Bikes. These vehicles operate on batteries; therefore it is needed to estimate the 
batteries power consumption and enhance the trip efficiency. This paper illustrates about data-driven approach for estimating the time required for trips as well as battery usage in e-bikes share systems. Feature engineering has been assigned to expand the dataset through environmental, temporal, and operational factors. Several machine learning models such as Linear Regression, Decision Tree, and Random Forest regression have been built and evaluated via performance measures including Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and R-Squared (R²) score. Out of all the tested models, the random forest regression executed best with RMSE of 290.02 and R² score of 0.75. In addition, Power Business Intelligence (Power BI) has been used for interactive data visualization. The proposed approach intensifies operational efficiency and supports sustainable urban mobility planning.