In the present era of globalization, the transportation system needs to be more flexible to meet customer demand. The port system has significant contributions in fluent operations of the freight transportation system. With the increase in transport demand, the port faces multiple complexities such as lower utilization of resources, capacity issues, etc. The forecasting of container traffic at the port would help the port planning team, and managers analyze the port system's infrastructural investment and optimization. In this study, the annual data of the past 20 years (1999-2019) of container traffic in TEUs at three Indian major ports, namely, Kolkata, Tuticorin and Cochin have been considered. The grey forecasting model and Auto-Regressive Integrated Moving Average (ARIMA) model is developed to analyze the container traffic data. Further, the Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) is used to test the model's accuracy. The results of the study show that both models are fit to forecast the container traffic data. Precisely, the grey forecasting model fits better than ARIMA for two ports Tuticorin and Cochin with MAPE of 10.52% & 4.80%, respectively. The findings of this study would guide the practitioners and planning managers in decision-making related to port optimization.In the present era of globalization, the transportation system needs to be more flexible to meet customer demand. The port system has significant contributions in fluent operations of the freight transportation system. With the increase in transport demand, the port faces multiple complexities such as lower utilization of resources, capacity issues, etc. The forecasting of container traffic at the port would help the port planning team, and managers analyze the port system's infrastructural investment and optimization. In this study, the annual data of the past 20 years (1999-2019) of container traffic in TEUs at three Indian major ports, namely, Kolkata, Tuticorin and Cochin have been considered. The grey forecasting model and Auto-Regressive Integrated Moving Average (ARIMA) model is developed to analyze the container traffic data. Further, the Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) is used to test the model's accuracy. The results of the study show that both models are fit to forecast the container traffic data. Precisely, the grey forecasting model fits better than ARIMA for two ports Tuticorin and Cochin with MAPE of 10.52% & 4.80%, respectively. The findings of this study would guide the practitioners and planning managers in decision-making related to port optimization.