分享興趣，傳播快樂，
增長見聞，留下美好。
親愛的您，這裡是LearingYard學苑！
今天小編為您帶來精讀博士論文《廣義乘性偏好資訊下考慮專家共識的多屬性群決策方法》5.5算例分析（2）。
歡迎您的通路！
Share interest, spread happiness,
increase knowledge, and leave beautiful.
Dear, this is the LearingYard Academy!
Today, the editor brings the " 5.5 example analysis (2) of the intensively read PhD thesis 'A Multi-Attribute Group Decision-Making approach considering expert consensus under generalised nultiplicative preference information'".
Welcome to visit!
一、内容摘要（Content Summary）
本期推文将從思維導圖、精讀内容、知識補充三個方面介紹精讀博士論文《廣義乘性偏好資訊下考慮專家共識的多屬性群決策方法》的5.5算例分析（2）。
This issue of the article will introduce the 5.5 example analysis (2) from three aspects: mind mapping, in-depth reading content, and supplementary knowledge, focusing on the doctoral dissertation "Multi-attribute Group Decision-making Method Considering Expert Consensus under Generalized Multiplicative Preference Information".
二、思維導圖（Mind Mapping）
三、精讀内容（Detailed Reading Content）
（一）對比分析（Comparative Analysis）
為了更好的探讨本章提出的基于 IMPR 的考慮一緻性與共識的MAGDM方法的有效性，将本章得到的排序結果與其他六種方法所得結果進行對比，具體決策結果和參數對比分析結果如下圖所示。其中，這六種對比的方法中的專家權重資訊、共識水準門檻值以及一些參數均沿用各個方法本身現有的取值，但一緻性與共識水準計算方法則統一使用本章建構的測度。
To better explore the effectiveness of the MAGDM method proposed in this chapter, which is based on the consideration of consistency and consensus using IMPR, the ranking results obtained in this chapter are compared with the results obtained by six other methods. The specific decision results and parameter comparison analysis are shown in the figure below. Among these six comparative methods, the expert weight information, consensus level threshold, and some parameters are all based on the existing values of each method. However, the consistency and consensus level calculation methods are uniformly based on the measures constructed in this chapter.
從上表可以看出，本章提出的決策方法得到的最優備選方案為A4，而未考慮專家之間的共識水準的決策方法所得最優備選方案大多為A3，并且現有 IMPR環境下考慮共識水準的方法所得結果也為A3。然後作者對結果進行了詳細分析，如下圖所示。
From the table above, it can be seen that the optimal alternative solution obtained by the decision method proposed in this chapter is A4, while the optimal alternative solutions obtained by the decision methods that do not consider the consensus level among experts are mostly A3. Additionally, the results obtained by the methods considering the consensus level under the existing IMPR environment are also A3. The authors then conducted a detailed analysis of the results, as shown in the figure below.
根據以上分析可以得到本章構造的多屬性群決策（MAGDM）流程在處理直覺乘性偏好資訊（IMPR）下的決策事件的優勢如下。
Based on the analysis above, the advantages of the Multi-Attribute Group Decision Making (MAGDM) process constructed in this chapter for dealing with decision events under Intuitionistic Multiplicative Preference Information (IMPR) are as follows.
（1）本章提出方法中定義了在專家參與評估程度不同的情境下IMPR的一緻性的衡量準則，并分别建構了專家參與重新評估與未參與重新評估時的IMPR一緻性優化方法，提高了MAGDM結果的準确性與合理性；（2）本章提出方法中定義了在專家參與評估程度不同的情境下IMPR之間的共識水準的計算方法，并建構了專家參與重新評估與未參與重新評估時的IMPR的共識達成方法，提高了MAGDM結果的可靠性；（3）本章提出的專家未參與重新評估的一緻性與共識水準改進疊代算法的可行性與合理性均得到了證明；（4）本章建構了IMPR下的誘導集結算子，以不同專家所給出的IMPR的一緻性與共識水準為誘導分量，自動生成決策者的權重資訊，有效完善資訊集結中決策者權重的求解環節。
(1) The proposed method in this chapter defines a measure criterion for the consistency of IMPR in situations where the degree of expert participation in the evaluation is different, and constructs consistency optimization methods for IMPR when experts participate in reassessment and when they do not, thereby improving the accuracy and rationality of the MAGDM results; (2) The proposed method in this chapter defines a method for calculating the consensus level among IMPR in situations where the degree of expert participation in the evaluation is different and constructs methods for achieving consensus among IMPR when experts participate in reassessment and when they do not, thereby improving the reliability of the MAGDM results; (3) The feasibility and rationality of the iterative algorithm for improving the consistency and consensus level of experts not participating in reassessment proposed in this chapter have been proven; (4) This chapter constructs the induction set settlement operator under IMPR, which uses the consistency and consensus level of IMPR provided by different experts as the inducing components to automatically generate the weight information of decision makers, effectively improving the solution process of decision maker weights in information aggregation.
（二）管理啟示（Managerial Insights）
在管理啟示部分，作者首先介紹了本章的具體研究内容：本章運用直覺乘性偏好資訊表征專家對于備選方案的兩兩對比評估結果，分别構造了兩種直覺乘性偏好資訊下的一緻性與共識水準改進方法，又依據不同專家給出的偏好資訊之間的一緻性與共識水準差異，提出誘導直覺乘性集結算子得到群體評估資訊，形成了直覺乘性偏好資訊下的基于一緻性與共識水準的群決策流程。
In the managerial insights section, the authors first introduced the specific research content of this chapter: This chapter uses intuitionistic multiplicative preference information to represent experts' pairwise comparison evaluation results of alternative solutions, and constructs two methods for improving consistency and consensus levels under two different intuitionistic multiplicative preference information. Based on the difference in consistency and consensus levels among different experts' preference information, the authors propose an induced intuitionistic multiplicative aggregation operator to obtain group evaluation information, forming a group decision-making process based on consistency and consensus levels under intuitionistic multiplicative preference information.
然後，作者介紹了本章構造的方法的實際應用情況，如下圖所示。
Then, the authors introduced the practical application of the method constructed in this chapter, as shown in the figure below.
根據本章的研究内容，可以得出本章提出直覺乘性偏好資訊下的MAGDM方法能夠解決實際決策工作中的如下關鍵問題。（1）專家有意願或有必要參與重新評估工作時的非一緻與非共識偏好資訊的識别與改進問題；（2）專家無意願或不需要參與重新評估工作時的偏好資訊的可接受一緻性與共識條件達成問題；（3）專家權重的客觀确定問題；（4）群體偏好資訊一緻性的延續問題，為特定條件下的複雜決策問題提供一定的指導作用。
Based on the research content of this chapter, it can be concluded that the MAGDM method proposed under intuitionistic multiplicative preference information in this chapter can solve the following key problems in actual decision-making work: (1) Identification and improvement of non-consistent and non-consensus preference information when experts are willing or necessary to participate in reassessment work; (2) Reaching acceptable consistency and consensus conditions of preference information when experts are unwilling or do not need to participate in reassessment work; (3) Objective determination of expert weights; (4) Continuation of consistency of group preference information, providing guidance for complex decision-making problems under specific conditions.
四、知識補充——決策偏好（Supplementary Knowledge - Decision Preferences）
偏好是指效用理論的一個概念。是指決策人對收益和風險的态度。決策人對一方案或後果的偏好強烈程度，稱為偏好程度。偏好可分為以下三種：（1）厭惡風險型，即對等的收益和損失，隻有當損失的機率小于1/2時，決策人才可能投資。（2）追求風險型，它是與厭惡型相反的類型。（3）相對風險中立型，即引入風險對決策人的偏好無明顯作用。
Preference is a concept in utility theory. It refers to the attitude of a decision-maker towards benefits and risks. The degree of preference of a decision-maker for a scheme or consequence is called the degree of preference. Preferences can be divided into the following three types: (1) Risk-averse type, that is, for equal benefits and losses, a decision-maker may only invest when the probability of loss is less than 1/2; (2) Risk-seeking type, which is the opposite of risk-averse type; (3) Relatively risk-neutral type, that is, the introduction of risk has no obvious effect on the preference of decision-makers.
偏好實際是潛藏在人們内心的一種情感和傾向，它是非直覺的，引起偏好的感性因素多于理性因素。偏好有明顯的個體差異，也呈現出群體特征。
Preferences are actually a kind of emotion and tendency hidden in people's hearts. They are non-intuitive and are more influenced by emotional factors than rational factors. Preferences exhibit significant individual differences and also show group characteristics.
今天的分享就到這裡了，
如果您對文章有獨特的想法，
歡迎給我們留言。
讓我們相約明天，
祝您今天過得開心快樂!
That's all for today's sharing.
If you have a unique idea about the article,
please leave us a message,
and let us meet tomorrow.
I wish you a nice day!
參考資料：ChatGPT、百度百科
參考文獻：
王睿. 廣義乘性偏好資訊下考慮專家共識的多屬性群決策方法 [D]. 四川: 西南交通大學, 2022.
本文由LearningYard學苑整理并發出，如有侵權請在背景留言！
文案| Ann
排版| Ann
稽核| Wei