Assessing the Socio-Economic Potential of Electric Vehicle Charging Infrastructure: A Machine Learning based Approach for Marrakech-Safi Region, Morocco

The evaluation of socio-economic potential plays a pivotal role in advancing sustainable transportation systems, particularly in the field of the infrastructure for electric vehicle charging. In this study, we present a comprehensive methodology that integrates the Analytic Hierarchy Process (AHP) and machine learning techniques to evaluate the socio-economic potential of the Marrakech-Safi region. By employing AHP, we determine the target variable, and subsequently apply RF (Random Forest) and SVM (Support Vector Machine) models incorporating 11 key factors such as demographics, road network, public facilities, typology, and power grid. The findings of our study reveal that the RF model, with an accuracy of 96.37%, outperforms the SVM model, which achieved an accuracy of 94.81%, in accurately predicting the socio-economic potential of our region. Building upon these results, we employ the RF model to project the potential of the Casablanca-Settat region, uncovering promising opportunities for the construction of an infrastructure for electric vehicle charging, notably in the city of Casablanca. The information provided by this study hold significant implications for decision-makers and policymakers involved in the planning and promotion of sustainable transportation infrastructure. By leveraging the combination of AHP and machine learning techniques, our methodology provides a solid framework for evaluating a region's socioeconomic potential, contributing to the formulation of informed strategies for sustainable transportation systems.