Data-driven ecological driving behaviour evaluation and green supply chain improvement

• Main Author: Zhou, Zhongcan
• Other Authors: Chen Songlin
• Format: Thesis-Master by Coursework
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Engineering; Green supply chain improvement; Eco-driving; Short trips; Clustering; Energy consumption
• Online Access: https://hdl.handle.net/10356/174109
• Institution: Nanyang Technological University

In the 14th Five-Year Plan, China has emphasized the importance of promoting green development, and the policy once again emphasized the importance of promoting energy conservation, thus improving the green supply chain. This dissertation delves into eco-driving evaluation and its crucial role in improving the green supply chain. Initially, over 3 million GPS trajectory data pieces were collected, enabling trip construction and feature extraction. Subsequently, short trips were classified and matched based on OSM road data, then the remaining 198529 pieces of short trips are clustered via Kmeans algorithms, resulting in 20 driving status categories across five road types. After that, an energy consumption model and eco-driving evaluation model were constructed in order to calculate the number for all short trips. Next chapter clarify the results and explain the relationships among road types, driving status and eco-driving scores. This dissertation also shows the relationship between eco-driving and green supply chain improvement, offering actionable strategies from the aspects of supplier testing, logistics optimization, and technological innovation. This dissertation shows its practical significance in amplifying sustainability within the green supply chain.