Development and application of a field knowledge graph and search engine for pavement engineering

Area of Science:

• Pavement Engineering
• Knowledge Management
• Artificial Intelligence

Background:

• Pavement engineering data in China is fragmented, hindering optimization and intelligent decision-making.
• Effective pavement information is currently underutilized due to data silos.
• There is a need to integrate and manage pavement knowledge assets for advanced engineering.

Purpose of the Study:

• To develop a novel framework for integrating heterogeneous pavement engineering data.
• To create a knowledge graph (KG) for efficient pavement information management.
• To enable intelligent decision-making in pavement design and optimization.

Main Methods:

• Utilized deep learning neural networks to extract knowledge from diverse data sources.
• Developed KG-Pavement, a flexible framework for data ingestion and knowledge graph generation.
• Implemented a pavement information search engine leveraging graph databases for efficient retrieval.

Main Results:

• Successfully integrated siloed pavement data into a unified knowledge base.
• KG-Pavement demonstrated improved efficiency in knowledge retrieval through indexing and constraints.
• Constructed the first pavement information search engine based on a knowledge graph.

Conclusions:

• Knowledge graphs offer an effective solution for integrating pavement engineering information.
• The KG-Pavement framework facilitates intelligent decision-making and knowledge discovery.
• This work represents a significant advancement in applying knowledge graphs to pavement engineering.