主管:中国科学院
主办:中国优选法统筹法与经济数学研究会
   中国科学院科技战略咨询研究院

中国管理科学 ›› 2026, Vol. 34 ›› Issue (2): 195-206.doi: 10.16381/j.cnki.issn1003-207x.2023.0990cstr: 32146.14.j.cnki.issn1003-207x.2023.0990

• • 上一篇    下一篇

基于多粒度概率语言和双参照点的应急决策方法

王增强1(), 蒲云2   

  1. 1.西华大学管理学院,四川 成都 610039
    2.西南交通大学交通运输与物流学院,四川 成都 610031
  • 收稿日期:2023-06-13 修回日期:2023-11-07 出版日期:2026-02-25 发布日期:2026-02-04
  • 通讯作者: 王增强 E-mail:wzqlinger@126.com
  • 基金资助:
    国家自然科学基金项目(71872153);国家自然科学基金项目(71401142);四川省哲学社会科学基金项目(SCJJ25ND021)

An Emergency Decision-Making Method Based on Multi-granularity Probabilistic Linguistic Terms and Double Reference Points

Zengqiang Wang1(), Yun Pu2   

  1. 1.School of Management,Xihua University,Chengdu 610039,China
    2.School of Transportation and Logistics,Southwest Jiaotong University,Chengdu 610031,China
  • Received:2023-06-13 Revised:2023-11-07 Online:2026-02-25 Published:2026-02-04
  • Contact: Zengqiang Wang E-mail:wzqlinger@126.com

摘要:

为了以有效的方式反映决策信息的不确定性和决策成员的有限理性,提出了基于多粒度概率语言和双参照点的应急决策方法。首先,决策成员使用偏好的概率语言评估标度表征各项关键风险因素权重的评估信息,本文给出一种改进的多粒度概率语言信息处理方法,确定了各项关键风险因素的权重;其次,将能够体现决策成员损失规避特征的累积前景理论运用到更能描述决策环境不确定性的多粒度概率语言决策环境中,并结合确定的双参照点计算各个可行方案预估效果的感知价值;再次,计算各个可行方案预估概率的收益感知权重和损失感知权重,在此基础上,集结各项关键风险因素的权重,确定不同可行方案的最终感知价值,获得突发事件应对的最佳方案;最后,以某地森林火灾的应对为例,验证了所提方法的有效性,并通过与现有方法的对比分析证明了所提方法的优越性。

关键词: 突发事件, 应急决策, 多粒度概率语言, 累积前景理论, 双参照点

Abstract:

In order to promote the comprehensive development of economy, over-exploitation of various resources have increased, resulting in various emergencies such as natural disasters and accidents. It is of great significance to study how to improve the effectiveness of decision-making under the imperfect and complex scenarios. The main problems are uncertainties and risk issues in the procedures, which are difficult to be effectively addressed. For reflecting the uncertainty of decision information and limited rationality of decision-makers in an effective way, a new emergency response decision-making based on multi-granularity probabilistic linguistic and double reference points is presented. Firstly, multi-granularity probabilistic linguistic variables are used to express the decision-makers’ evaluation of each key risk factor, and the weights of key risk factors are determined based on an improved method to address multi-granularity probabilistic linguistic information. Secondly, the cumulative prospect theory, which can embody the characteristic of loss-aversion of decision-makers, is applied to the multi-granularity probabilistic linguistic environment, then, the perceived values of each feasible alternative’s estimated effect are calculated by the certain double reference points. Thirdly, the earn perceived weight and loss perceived weight of each feasible alternative’s estimated probability is determined, on this basis, the ultimate perceived value of each alternative is obtained by combining the weights of key risk factors, and the best alternatives can be determined. Finally, the response to forest fire illustrates the effectiveness of the proposed method, and verifies the superiority by comparative analysis. The proposed integrated model based on multi-granularity probabilistic linguistic and double reference points not only represent the imprecise and vague information more effectively, but also consider the decision makers’ bounded rationality under risk comprehensively, and then provide some clues for the effective processing of information theories in emergency decision-making process.

Key words: emergency events, emergency decision-making, multi-granularity probabilisticlinguistic, cumulative prospect theory, double reference points

中图分类号: