Research on optimization method of coal mine safety investment driven by digital twin-enabled INFO-SVR

• Published in: Process safety and environmental protection Vol. 200; p. 107345
• Main Authors: Wang, Yunrui, Li, Wenxuan, Wang, Le, Li, Rui, Wang, Haoning
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.08.2025
• Subjects: Accident losses; Coal mine; Digital twin; Optimization; Safety investment; Safety investment; Digital twin; Coal mine; Accident losses; Optimization
• ISSN: 0957-5820
• DOI: 10.1016/j.psep.2025.107345

To make effective use of safety investment resources, and reduce accident risk and economic loss, an optimization method of safety investment based on digital twin vector weighted average algorithm optimization support vector regression (INFO-SVR) is proposed. Consequently, a five-dimensional digital twin optimization model for coal mine safety investment is constructed, elucidating the optimization process within this model. The study provides a detailed explanation of constructing safety investment indicators under digital twins, calculating indicator weights, and fitting coal mine safety investment with accident losses. To ensure fitting accuracy, three algorithms (SSA, GWO, INFO) are employed to optimize the parameters of Support Vector Regression (SVR). Using the Y coal mining enterprise as a case, the established model is applied to optimize the safety investment, and the results show that the economic loss of accidents caused by the optimized safety investment is reduced by 14.9 %, effectively improving the safety management level of coal mining enterprises, reducing the accident risk and accident loss. This study provides a scientific basis for safety investment decision-making in coal mining enterprises and provides a new way for the safety management in the industrial field.