中国安全科学学报 ›› 2017, Vol. 27 ›› Issue (2): 1-6.doi: 10.16265/j.cnki.issn1003-3033.2017.02.001

• 安全科学技术基础学科 •    下一篇

面向行为安全的泛场景数据理论与应用研究

佟瑞鹏 副教授, 陈策, 刘思路, 卢恒, 马建华   

  1. 中国矿业大学(北京) 资源与安全工程学院,北京 100083
  • 收稿日期:2016-11-01 修回日期:2017-01-11 出版日期:2017-02-28 发布日期:2020-11-22
  • 作者简介:佟瑞鹏 (1977—),男,黑龙江穆棱人,博士,副教授,主要从事行为安全与管理、风险建模与评估、公共安全与健康等方面的研究。E-mail:tongrp@cumtb.edu.cn。
    陈 策 (1991—),男,河北保定人,硕士研究生,主要研究方向为行为安全与管理。E-mail:mychence@126.com。
  • 基金资助:
    国家自然科学基金资助(51674268)。

Research on theory and application of pan-scene data for behavioral safety

TONG Ruipeng, CHEN Ce, LIU Silu, LU Heng, MA Jianhua   

  1. School of Resources & Safety Engineering, China University of Mining & Technology (Beijing), Beijing 100083, China
  • Received:2016-11-01 Revised:2017-01-11 Online:2017-02-28 Published:2020-11-22

摘要: 为进一步挖掘统计数据中的安全规律,在场景研究基础上,分析泛场景数据的理论内涵,提出从时间、位置区域、行为个体、不安全动作、行为性质、行为痕迹、风险等级等7个维度表征不安全行为泛场景数据。通过对现实场景和抽象场景2个层面的采集数据进行单维度统计分析及多维度关联规则挖掘,实现了泛场景数据的应用。从全国225起瓦斯爆炸事故中采集871条不安全行为泛场景数据,并对其进行统计分析和关联规则挖掘。结果表明,泛场景数据的理论和方法可用来有效挖掘和释放统计数据价值,从多维度深入分析不安全行为的内在特征规律。

关键词: 泛场景数据, 不安全行为, 数据挖掘, 关联规则, 瓦斯爆炸事故

Abstract: In order to further excavate the safety rules in the statistical data,the theoretical connotations of pan-scene data were analyzed on the basis of scene research.The pan-scene data on unsafe behavior were described by 7 dimensions namely time, position, behavioral individual, unsafe action, behavioral nature, behavior trace and risk level.The pan-scene data can be collected from real scenes and abstract scenes.Then,the application of the pan-scene data was implemented through the single dimensional statistical analysis and multi-dimensional association rules mining.871 pan-scene data on unsafe behaviors in 225 gas explosion accidents were collected.Then,a statistical analysis and association rules mining were carried out for the data.The results show that the theory of the pan-scene data can be used for mining and releasing data value effectively and analyzing the intrinsic characteristics of unsafe behavior from multi dimensions deeply.

Key words: pan-scene data, unsafe behavior, data mining, association rule, gas explosion accident

中图分类号: