The quality and efficiency of road maintenance greatly impact on transportation performance. There is an urgent need to automate the process of decision-making for road maintenance, especially for reactive maintenance. However, automating this process is challenging because (1) the expert knowledge required in this process is implicit and difficult to formalize and (2) the defect reports from frontline inspectors are in unstructured human language. This paper proposes a method based on natural language processing (NLP) and knowledge graph reasoning. This method combines the semantic understanding capability of NLP with the structural reasoning capability of the knowledge graph, enabling nuanced and logically rigorous road maintenance decisions. Real-world highway defect reports are used for validation. The experimental results demonstrate a significant improvement of the proposed model over baseline models, with up to 13.5% higher accuracy on the prediction of key tasks. This approach enables a traceable decision-making process, avoiding a black-box model.

The use of graph-based asset information is growing in the Architecture, Engineering, Construction, and Operation (AECO) domain. However, there has been no comprehensive investigation into its internal structure, which hinders the efficiency of information querying and knowledge discovery. This study aims to reveal the internal structure of graph-based IFC (Industry Foundation Classes) information, referred to as IFC-Graph, and to assess the impact of graph structure on the efficiency of information querying and knowledge discovery. First, through a close examination of the IFC standard, five levels of detail were identified and proposed for IFC-Graph. Second, methods for generating graphs at each level were developed. Lastly, the impact of graph structure on information querying and knowledge discovery was assessed. The results show that: (1) relations constitute the main framework of graph-based IFC information and are key to efficiently extracting relational information from IFC-Graph; and (2) the proposed five-level framework makes graph-based asset information more practical to use, by increasing flexibility in graph generation and improving the efficiency of graph queries and knowledge discovery.