【摘要】：在社會(huì)活動(dòng)中,往往需要對(duì)同類的決策單元進(jìn)行評(píng)價(jià)或排序,從中選擇最優(yōu)方案。通常情況下,由于決策會(huì)涉及到多個(gè)屬性,這些屬性都是能夠影響到我們判斷的不可忽略的重要因素時(shí),就會(huì)造成了決策的復(fù)雜性。對(duì)于多投入多產(chǎn)出的決策單元的排序問(wèn)題,交叉效率通過(guò)線性規(guī)劃,綜合考慮測(cè)算得到的自評(píng)效率值和他評(píng)效率做出評(píng)價(jià)。但是如何處理模型中的多重最優(yōu)解造成的問(wèn)題,一直都是該領(lǐng)域的研究熱點(diǎn)。另一方面,猶豫模糊集的出現(xiàn),能夠允許屬性值以多個(gè)互不干擾的隸屬度,同時(shí)獨(dú)立存在,為綜合考慮多重最優(yōu)解的決策信息提供了一種新的視角。在二者的基礎(chǔ)上,提出了基于猶豫模糊集的交叉效率群決策方法。研究的關(guān)鍵問(wèn)題是在猶豫模糊集的基礎(chǔ)上,如何構(gòu)建群決策交叉效率矩陣,以及根據(jù)該矩陣,如何對(duì)決策單元進(jìn)行排序。主要工作進(jìn)展如下:1、研究了猶豫模糊集構(gòu)成的交叉效率矩陣下,決策單元的排序問(wèn)題。方法Ⅰ:將問(wèn)題轉(zhuǎn)化為群決策問(wèn)題,將改進(jìn)的進(jìn)取型、進(jìn)取型、中立性、仁慈型和改進(jìn)的仁慈型這五個(gè)經(jīng)典模型代表從悲觀到樂(lè)觀過(guò)渡的五名專家,得到5組判斷決策單元有效性的隸屬度,將猶豫模糊集引入到交叉效率矩陣中;在交叉效率矩陣中,引入多目標(biāo)屬性權(quán)重優(yōu)化方法,計(jì)算效率值的屬性權(quán)重;根據(jù)決策單元與正負(fù)理想決策單元的貼近度,進(jìn)而對(duì)決策單元排序。2、為了直觀地反應(yīng)決策單元的綜合效率值,對(duì)方法Ⅰ進(jìn)行了改進(jìn),提出了一種基于決策者風(fēng)險(xiǎn)態(tài)度的交叉效率群決策方法(方法Ⅱ)。該方法充分利用數(shù)據(jù)信息,從決策單元整體和局部?jī)蓚�(gè)角度考慮,將區(qū)間猶豫模糊集引入交叉效率,為避免極端值影響決策,引入風(fēng)險(xiǎn)態(tài)度因子,通過(guò)潛在信息函數(shù)求出各效率值的權(quán)重,得到綜合效率值作為評(píng)價(jià)結(jié)果。3、實(shí)證分析。通過(guò)前面提到的方法Ⅱ,對(duì)中部六省的區(qū)域創(chuàng)新效率進(jìn)行研究,分析“十一五”期間和“十二五”期間各省的創(chuàng)新效率狀況以及變動(dòng)情況。本文針對(duì)交叉效率的多重最優(yōu)解,引入猶豫模糊集,研究了基于猶豫模糊集的交叉效率群決策方法,并對(duì)中部六省的區(qū)域創(chuàng)新效率進(jìn)行了實(shí)證分析,為交叉效率理論研究提供了新的方向。
[Abstract]:In social activities, it is often necessary to evaluate or rank the same decision making units, and select the best scheme from them. Usually, because of the multiple attributes involved in decision making, these attributes are important factors that can influence our judgment, which will lead to the complexity of decision making. For the ranking problem of decision making units with multiple inputs and outputs, the cross efficiency is evaluated by linear programming, and the self-evaluation efficiency and the other evaluation efficiency are considered synthetically. However, how to deal with the problem caused by multiple optimal solutions in the model has always been a hot topic in this field. On the other hand, the emergence of hesitant fuzzy sets allows attribute values to exist independently with multiple non-interference degrees of membership, which provides a new perspective for decision information considering multiple optimal solutions. Based on the two methods, a method of cross-efficiency group decision making based on hesitant fuzzy sets is proposed. The key problem of the study is how to construct the cross-efficiency matrix of group decision making based on the hesitant fuzzy set and how to sort the decision units according to the matrix. The main works are as follows: 1. The scheduling problem of decision units under the cross efficiency matrix of hesitant fuzzy sets is studied. Method I: transform the problem into a group decision making problem. The five classical models of improved enterprising, neutral, merciful and improved benevolent type represent five experts in the transition from pessimism to optimism. Five groups of membership degrees are obtained to determine the validity of Decision-making units, and the hesitant fuzzy sets are introduced into the cross-efficiency matrix. In the cross-efficiency matrix, the multi-objective attribute weight optimization method is introduced to calculate the attribute weight of the efficiency value. According to the degree of closeness between the decision making unit and the positive or negative ideal decision making unit, the decision making unit is sorted. 2. In order to directly reflect the comprehensive efficiency value of the decision making unit, the method I is improved. A method of cross-efficiency group decision making based on decision makers' risk attitude is proposed (method 鈪，

本文編號(hào)：2393208