Proceedings of the 2nd International Conference on Public Management, Digital Economy and Internet Technology, ICPDI 2023, September 1–3, 2023, Chongqing, China

Research Article

Evaluation and Prediction of Enterprise Safety Management Efficiency Based on DEA-BP Neural Networks

Download82 downloads
  • @INPROCEEDINGS{10.4108/eai.1-9-2023.2338749,
        author={Lixing  Zhu and Jie  Xiao and Tiancheng  Wang and Jinhai  Tang and Chengyifan  Li and Junjie  Yan},
        title={Evaluation and Prediction of Enterprise Safety Management Efficiency Based on DEA-BP Neural Networks},
        proceedings={Proceedings of the 2nd International Conference on Public Management, Digital Economy and Internet Technology, ICPDI 2023, September 1--3, 2023, Chongqing, China},
        publisher={EAI},
        proceedings_a={ICPDI},
        year={2023},
        month={11},
        keywords={enterprise security management efficiency evaluation;data envelopment analysis (dea);back propagation neural network (bpnn)},
        doi={10.4108/eai.1-9-2023.2338749}
    }
    
  • Lixing Zhu
    Jie Xiao
    Tiancheng Wang
    Jinhai Tang
    Chengyifan Li
    Junjie Yan
    Year: 2023
    Evaluation and Prediction of Enterprise Safety Management Efficiency Based on DEA-BP Neural Networks
    ICPDI
    EAI
    DOI: 10.4108/eai.1-9-2023.2338749
Lixing Zhu1, Jie Xiao1, Tiancheng Wang2,*, Jinhai Tang2, Chengyifan Li2, Junjie Yan2
  • 1: Guangdong University of Science and Technology
  • 2: Zhuhai College of Science and Technology
*Contact email: 1393334922@qq.com

Abstract

The evaluation and prediction of enterprise security management efficiency is an important and challenging topic. In this study, we use a hybrid model of Data Envelopment Analysis (DEA) and Back Propagation Neural Network (BPNN) to evaluate and predict the efficiency of enterprise safety management. This model considers multiple input variables, including security inputs, number of employees, and enterprise size, and is trained and tested on a virtual enterprise dataset. The results show that all the input variables affect the efficiency of an enterprise's security management to varying degrees, with the effects of security inputs and number of employees being particularly significant. Through principal component analysis, we further visualized the relationship between input variables and efficiency values in a two-dimensional space. The results of this study have important theoretical and practical implications for understanding and improving enterprise safety management, as well as providing a new perspective to explore the complexity of this field.