The increasing demand for logistics and the urgent need for environmental sustainability requires innovative solutions to optimize transportation routes and reduce greenhouse gas emissions. This study explores the role of artificial intelligence in improving logistics efficiency and environmental performance by applying various regression models to predict travel times and emissions using real-world logistics datasets. These datasets include factors such as vehicle types, traffic conditions, weather, emissions, distance, fuel consumption, and package attributes. The study employs machine learning models like Linear Regression, Ridge and Lasso Regression, Support Vector Machines, Decision Trees, Random Forests, Gradient Boosting, XGBoost, Gaussian Processes, and Multi-layer Perceptron Regressors, as well as advanced deep learning techniques including LSTM, RNN, CNN, and ARIMA for time series forecasting. The evaluation framework uses metrics such as Mean Squared Error, Mean Absolute Error, R-squared, and Mean Absolute Percentage Error, with hyperparameter tuning to enhance model performance. It also incorporates dynamic route recalculation, emissions impact analysis focusing on CO2 and other greenhouse gases, cost-benefit optimization, and scenario planning. The results identify the most effective models for route optimization and emission reduction, demonstrating AI’s potential to improve logistics and sustainability and reduce the ecological footprint of transportation. This research aligns with several United Nations Sustainable Development Goals (SDGs), including Industry, Innovation, and Infrastructure (Goal 9), Sustainable Cities and Communities (Goal 11), Responsible Consumption and Production (Goal 12), Climate Action (Goal 13), and Partnerships for the Goals (Goal 17).