Safety early warning research for highway construction based on case-based reasoning and variable fuzzy sets

Area of Science:

• Civil Engineering
• Safety Science
• Artificial Intelligence

Background:

• Highway construction is a high-risk industry with a history of safety issues, particularly in China.
• A significant factor contributing to accidents is the lack of effective early warning systems.
• Existing safety measures need enhancement to address the specific risks in highway construction.

Purpose of the Study:

• To develop an intelligent early warning system for highway construction safety.
• To integrate case-based reasoning (CBR) with advanced analytical techniques for improved safety.
• To address key challenges in safety warning systems, including index selection and accident analysis.

Main Methods:

• Case-based reasoning (CBR) framework for safety early warnings.
• Association rule mining for analyzing accident causes.
• Support vector machine classifiers for predicting warning degrees.
• Variable fuzzy qualitative and quantitative change criterion modes for forecasting.

Main Results:

• The proposed CBR system effectively integrates multiple intelligent technologies for safety early warnings.
• Solutions are provided for critical issues like index selection, accident cause association, and warning degree forecasting.
• The methods demonstrate effectiveness and synergy in practical safety warning applications.

Conclusions:

• An intelligent and effective highway construction safety early warning system has been established.
• The integration of CBR and data mining techniques offers a robust solution for enhancing construction safety.
• This approach provides valuable tools and technologies for proactive safety management in highway projects.