The knowledge graph of Volvo Cars Logistics

The automotive industry relies heavily on efficient supply chain and logistics operations to ensure timely delivery of components, reduce costs, and meet customer demands. However, the industry faces challenges in managing its complex logistics networks, especially the transportation of materials and goods from suppliers. This study aimed to leverage the potential of graph databases, specifically the Neo4j graph database, to create a comprehensive representation of the transportation network in Volvo cars. By using this database and the querying capabilities of Cypher, this research eased the visualization of transportation connections, also recognizing the best paths possible to reach a plant from the supplier. The data were cleaned, and if needed, synthetic data was added for missing CO2 emission values. Then, the data was stored, and new paths were created in Neo4j based on the previous paths. Linkurious was used to better visualize paths connecting suppliers and plants. This visualization tool facilitated understanding the locations of packages, suppliers, and plants. Subsequently, using Cypher query language, the effectiveness of the solution was evaluated in terms of cost reduction, time savings, and CO2 emissions reduction in paths found between three locations of all the packages traversing them. The evaluation of our paths revealed that, under various scenarios, over 50% of paths of our packages achieved better cost savings compared to alternative routes, while nearly 16% saw improvements in time compared to their initial routes. Additionally, when imposing a CO2 limitation, more than 20% of cases achieved cost optimization. These findings demonstrate the system's efficacy in optimizing pathfinding even when there is a constraint on one of the factors, contributing to enhanced efficiency and sustainability in the automotive industry.