Supervised machine learning algorithms for measuring and promoting sustainable transportation and green logistics

Abstract

The sustainable development of freight transport has received much attention in recent years. The new regulations for sustainable transport activities established by the European Commission and the United Nations have created the need for road freight transport companies to develop methodologies to measure the social and environmental impact of their activities. This work aims to develop a model based on supervised machine learning methods with intelligent classification algorithms and key performance indicators for each dimension of sustainability as input data. This model allows establishing the level of sustainability (high, medium or low). Several classification algorithms were trained, finding that the support vector machines algorithm is the most accurate, with 98% accuracy for the data set used. The model is tested by establishing the level of sustainability of a European company in the road freight sector, thus allowing the establishment of green strategies for its sustainable development.