A two-stage EDM method based on KU-CBR with the incomplete linguistic intuitionistic fuzzy preference relations
Introduction
In recent years, the frequent emergencies in the world have seriously affected people’s health and social stability, such as: COVID-19, locust disaster in East Africa, forest fires in Australia and so on. As the key issue in the emergency management field, emergency decision making (EDM) can reduce the loss of emergency (Gao, Xu, Liang, & Liao, 2019). Thus, the research on EDM have become one of the most important hotspots and frontier issues in the field of emergency management (Ding et al., 2021, Jiang et al., 2021, Wu et al., 2021). The EMD process can be divided into six stages: problem definition, goal setting, alternative generation, alternative selection, organizational implementation and feedback modification (Zhou, Wu, Xu, & Fujita, 2018). Among these stages, the alternative generation and alternative selection mainly determine the efficiency of EDM, which are discussed in this paper.
In the alternative generation stage, it is an effective methodology to refer to the past experience, which can be realized by case based reasoning (CBR) technology (Chen et al., 2021, Zhang et al., 2021). CBR systems compare a new problem with the historical cases and retrieve similar historical cases (Paulson & Juell, 2004). Therefore, a CBR system should contain two basic elements: case representation to describe the cases, and similarity measurement for retrieving similar cases. However, previous CBR studies generally focus on case retrieval, and lack case representation (Bannour et al., 2020, Zhang et al., 2020, Zheng et al., 2018). In fact, the efficiency of case retrieval is determined directly by the case representation method used. As a result, it is more logical to propose a case retrieval method with the consideration of case representation. When considering case representation, its content and form need to be studied. Due to the dynamic nature of emergencies, an emergency can be in different scenarios from start to finish, and a historical emergency case can be composed of multiple scenarios (Huang, Nie, & Luo, 2020). In order to facilitate the subsequent case retrieval, it is necessary to express the emergency scenario uniformly, and the knowledge-unit model just meets this demand. The knowledge-unit model was proposed in literature (Wang, 2011), which could meet the needs of representation and management of cross-disciplinary knowledge in emergency management, and had been well validated in Chen, Dong, Wang, Xiao, and Gong (2011) and Han, Li, and Su (2019). Therefore, we first construct an emergency case representation model by using the knowledge-unit model. On the basis of emergency scenario representation, a scenario similarity measurement is proposed to retrieve similar historical scenarios. Based on which, a knowledge-unit based case-based reasoning (KU-CBR) method is proposed to generate the alternatives set.
In the alternative selection stage, some experts are required to evaluate the alternatives. In EDM problem, the preference relation is one of the most popular tools to express experts’ evaluation information about the alternatives, as it only require the experts to compare a pair of alternatives at one time. Many kinds of preference relations have been applied to EDM problems (Sun et al., 2021, Xu et al., 2019). For example, Gao et al. (2019) proposed an EDM method using the probabilistic linguistic preference relations. Song and Li (2019) proposed a large group EDM by multigranular probabilistic fuzzy linguistic preference relations to represent subgroup’s preferences information. Among these preference relations, the linguistic preference relation is the most popular, which are simple and straightforward and can mitigate the cost of inaccuracies to some extent (Krishankumar, Ravichandran, Ahmed, Kar, & Tyagi, 2019). However, the elements in these linguistic preference relations can only describe the membership degree of one alternative preferred to the other. To make up this limitation, by combining linguistic variables with intuitionistic fuzzy sets, Chen, Liu, and Pei (2015) introduced linguistic intuitionistic fuzzy numbers (LIFNs) that can express membership degree and non membership degree by using linguistic variables. Meng, Tang, and Fujita (2019) proposed the linguistic intuitionistic fuzzy preference relations (LIFPRs), which are more suitable for complex EDM problems. Jin et al. (2019) investigated the models to deal with the individual consistency and group consensus for LIFPRs. For the decision-making model with preference relations, the definition of preference relations’ consistency and alternatives’ priority weights are two important issues, which are determined on the basis of complete preference relations (Cao et al., 2021, Wang, 2013a, Wu et al., 2019). However, in some cases, some experts may not be able to express their preference for two alternatives, which leads to the incomplete preference relation (Li et al., 2021, Zhang et al., 2014). Thus, we first construct a model to determine the missing values in the incomplete preference relationship. Then, we build the models to generate the LIFPRs with acceptable consistency. Finally, the priority weights of alternatives is derived to rank alternatives.
Based on the above analysis, a two-stage EDM method which includes the alternative generation stage and the alternative selection stage is proposed. The remaining part of this paper proceeds as follows. In Section 2, we first give the knowledge-unit model, and introduce some definitions about the linguistic term set and LIFPRs. In Section 3, we first show the basic procedure of the EDM method, then concretely analyze the steps. In Section 4, a case study of typhoon is provided to illustrate the use of the EDM method, and the alternative selection method with LIFPRs is compared with another method to demonstrate its effectiveness. We conclude this study in Section 5.
Section snippets
Preliminaries
In this section, the knowledge-unit model is presented. Then, some basic concepts and definitions related to the linguistic term set and LIFPRs are introduced, and the consistency of LIFPRs is analyzed, which is a basis of the decision making method with LIFPRs.
The two-stage EDM method
In this section, a two-stage EDM method is proposed. The basic procedure of the EDM method is shown in Fig. 1, and the detailed analysis of the two stages is as follows.
Case application
In this section, the EDM method is used to solve the case of typhoon emergency. The case is as follows:
On August 8, 2020, a typhoon landed in the coastal area of a city. When landing, the wind near the center has 16 levels, the wind speed is 52 m/s, and the minimum pressure of the center is 930 HPA. After landing, at the speed of 15 kilometers per hour, it gradually turned westward. According to statistics, more than 5200 houses have collapsed.
Using the two-stage EDM method to solve the above
Conclusions
Based on the KU-CBR as well as LIFPRs, in this paper, we propose a method for solving EDM problem at two levels, i.e., the alternative generation stage and the alternative selection stage.
In the alternative generation stage, a KU-CBR method is developed to generate alternatives. First of all, the emergency case representation model is constructed with knowledge-unit, which can represent the case information completely and structurally. On this basis, a scenario similarity measurement method for
CRediT authorship contribution statement
Liyuan Zhang: Writing – review & editing, Funding acquisition. Chunlei Liang: Conceptualization, Methodology, Writing – original draft. Tao Li: Methodology, Supervision. Wentong Yang: Acquisition of data, Software.
Acknowledgments
This work is supported by the National Social Science Foundation of China [grant number 19CGL045].
References (36)
- et al.
Case-based reasoning for crisis response: Case representation and case retrieval
Procedia Computer Science
(2020) - et al.
A bidirectional feedback mechanism for balancing group consensus and individual harmony in group decision making
Information Fusion
(2021) - et al.
A heterogeneous GRA-cbr-based multi-attribute emergency decision-making model considering weight optimization with dual information correlation
Expert Systems with Applications
(2021) - et al.
An extended TODIM approach for group emergency decision making based on bidirectional projection with hesitant triangular fuzzy sets
Computers & Industrial Engineering
(2021) - et al.
Hybrid similarity measure for case retrieval in CBR and its application to emergency response towards gas explosion
Expert Systems with Applications
(2014) - et al.
Expected consistency-based emergency decision making with incomplete probabilistic linguistic preference relations
Knowledge-Based Systems
(2019) - et al.
An assertive reasoning method for emergency response management based on knowledge elements C4. 5 decision tree
Expert Systems with Applications
(2019) - et al.
Linguistic intuitionistic fuzzy preference relations and their application to multi-criteria decision making
Information Fusion
(2019) Derivation of intuitionistic fuzzy weights based on intuitionistic fuzzy preference relations
Applied Mathematical Modelling
(2013)Derivation of intuitionistic fuzzy weights based on intuitionistic fuzzy preference relations
Applied Mathematical Modelling
(2013)
A novel emergency decision-making model for collision accidents in the yangtze river
Ocean Engineering
Deviation measures of linguistic preference relations in group decision making
Omega
Large group emergency decision-making method with linguistic risk appetites based on criteria mining
Knowledge-Based Systems
The concept of a linguistic variable and its application to approximate reasoning—I
Information Sciences
A CBR-based decision-making model for supporting the intelligent energy-efficient design of the exterior envelope of public and commercial buildings
Energy and Buildings
Conditions of two methods for estimating missing preference information
Information Sciences
Emergency decision making for natural disasters: An overview
Int. J. Disaster Risk Reduct.
Knowledge unit model for emergency management of unconventional incident
J. Intell.
Cited by (12)
Advanced linguistic intuitionistic fuzzy Maclaurin symmetric means for MAGDM
2024, Applied IntelligenceCognitively Inspired Group Decision-Making with Linguistic q-Rung Orthopair Fuzzy Preference Relations
2023, Cognitive Computation