The need for effective public transport systems, especially metro rail networks, has greatly expanded due to the fast growth of urbanisation. Reducing congestion, increasing service quality, and optimising resource allocation all depend on accurate passenger flow forecasts. A machine learning-based method for forecasting passenger flow in metro systems is presented in this study.The suggested solution makes use of past passenger data, including time-based characteristics like date, time, peak hours, and seasonal fluctuations. To deal with missing values, normalise data, and identify pertinent features, data preparation techniques are used. To examine temporal patterns and predict future passenger demand, machine learning models like Decision Trees, Long Short-Term Memory (LSTM) networks, and Linear Regression are used. Metrics like Mean Absolute Error (MAE), Mean Squared Error (MSE), and Root Mean Squared Error (RMSE) are used to assess the models' performance. The results of the experiments show that the suggested method offers precise and trustworthy passenger flow forecasts, facilitating improved metro operations planning and administration.By facilitating data-driven decision-making, this study demonstrates how machine learning approaches might enhance the sustainability and efficiency of urban transit systems