Sustainable spatial and semantic-web enhanced pathfinding in dynamic domains (SPEED): a case study of grain transportation in Ukraine

Abstract

Optimal routing of goods is crucial when addressing supply chain challenges. Within this context, grain transportation stands out as a significant sub-issue. For this research,grain transportation is defined as the process of determining the best routes between grain elevators and railway stations. As in most supply chain and path finding problems, this process is also challenging due to the complexity and dynamics of the domain. In this work, we propose a novel application that uses the KNowledge Acquisition and Representation Methodology (KNARM) and the KnowWhereGraph to enhance the path-finding process during transportation of goods, in this specific case grain. KNARM is a methodology that allows for creating and maintaining modular ontologies that can represent complex domains and support reasoning. KnowWhereGraph, is a densely connected, cross-domain knowledge graph and geo-enrichment service stack that provides rich and up-to-date geospatial information. By integrating these two components, our application aims to leverage the semantic and spatial knowledge to find more accurate and efficient paths for grain transportation. To find the optimal path, we use the A* algorithm, which is a heuristic search algorithm that can find the optimal path between two locations, taking into account the criteria of cost, time, and risk. The A* algorithm can also adapt to dynamic and uncertain situations, such as changes in weather, traffic, or security conditions by updating the paths. We also discuss the challenges and limitations that we faced during the ontology development and data integration process, and how we resolved or mitigated them. Our work demonstrates the potential of using ontologies and knowledge graphs to enhance path-finding problems in complex and dynamic domains. We show that our application can find accurate, efficient, and robust paths for grain transportation, based on the feedback from domain experts and GIS experts. We also provide new insights into the modeling of the factors that may affect grain transportation in Ukraine, such as weather, traffic, and security conditions.