Information-Gain-Based Multi-Criteria Decision-Making Approach for Optimizing Freight Electric Vehicle Charging Station Siting using Global Positioning System Data

Abstract

This study focuses on optimizing the siting of freight electric vehicle charging stations (FEVCS), a pivotal initiative for enhancing the efficiency and sustainability of freight transportation systems. We develop a novel optimal location selection model based on the Developing Geographic PageRank framework, which leverages global positioning system data from medium- and heavy-duty electric trucks. This model integrates trajectory density with other geographic information to enhance decision-making. Machine learning is also utilized in a multi-criteria decision-making process to improve the objectivity and accuracy of the criteria weighting process. The results show that the distance to existing charging stations is the most critical factor. This study provides a research framework for FEVCS planning while laying the foundation for urban planners to create a resilient green transportation system.