A hybrid approach for low-carbon transportation system analysis: integrating CRITIC-DEMATEL and deep learning features.

Abstract

As supply chains, logistics, and transportation activities continue to play a significant role in China's economic and social developments, concerns around energy consumption and carbon emissions are becoming increasingly prevalent. In light of sustainable development goals and the trend toward sustainable or green transportation, there is a need to minimize the environmental impact of these activities. To address this need, the government of China has made efforts to promote low-carbon transportation systems. This study aims to assess the development of low-carbon transportation systems in a case study in China using a hybrid approach based on the Criteria Importance Through Intercriteria Correlation (CRITIC), Decision-Making Trial and Evaluation Laboratory (DEMATEL) and deep learning features. The proposed method provides an accurate quantitative assessment of low-carbon transportation development levels, identifies the key influencing factors, and sorts out the inner connection among the factors. The CRITIC weight matrix is used to obtain the weight ratio, reducing the subjective color of the DEMATEL method. The weighting results are then corrected using an artificial neural network to make the weighting more accurate and objective. To validate our hybrid method, a numerical example in China is applied, and sensitivity analysis is conducted to show the impact of our main parameters and analyze the efficiency of our hybrid method. Overall, the proposed approach offers a novel method for assessing low-carbon transportation development and identifying key factors in China. The results of this study can be used to inform policy and decision-making to promote sustainable transportation systems in China and beyond.