結合顧客學優質服務品質的實際，提出咖啡店選擇的指標體系，針對解決具有語言評價資訊的多屬性群決策問題，提出了一種基於二元語義資訊處理的群決策方法。該方法採用近年來最新發展的二元語義概念對語言評價資訊進行處理和運算，並依據傳統投影分析方法的基本思想，通過計算備選方案對正理想方案和負理想方案的投影值，進而計算備選方案對正理想方案相對貼近度，最終確定最優方案。由於在該模式中採用了熵權，從而避免了確定低層次多因素權重的主觀性。通過將該模式顧客在選擇咖啡店中的應用，得出其咖啡店的優質商店的優劣排序，為顧客選擇理想的咖啡店。

A new method is proposed to solve multiple attribute group decision making problems with 2-tuple linguistic assessment information. The paper developed service quality of evaluation index system for coffee industry selection is put forward based on guestology, and a multi-level model of modified ideal solution, based on entropy weight. According to the traditional ideas of project method, the optimal alternative(s) is determined by calculating the project values of every alternative and positive ideal solution and negative ideal solution. It is based on the concept that the optimal alternative should have the largest degree of relative closeness from positive ideal solution. Finally, it also intend to utilize quantitive method for analysis to improve the order preference of coffee industry selection was attained, and the excellent and ideal coffee industry was ascertained for the guest operation.