透過您的圖書館登入
IP:3.237.31.131
  • 學位論文

服務復原策略效益評估模式建立之研究-以馬可夫鏈為分析工具

To Establish An Evaluation Model for Service Recovery Strategy by Markov Chain Decision Processes

指導教授 : 顧志遠
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


由於服務業時代的來臨,企業已逐漸重視服務系統管理的重要性,企業管理者,均致力於提昇顧客的滿意度為職志,以確保顧客價值,提昇顧客忠誠度,使企業永續經營。 服務復原在管理學派裡有服務品質論及服務善後行銷管理論,但均回歸於探討顧客終身價值,可見顧客滿意的價值在於顧客的忠誠度(人氣,不斷地回購);但企業無論如何地運用管理技巧及運籌維幄,最後終難逃其將本求利的宿命。過度投資服務復原策略,雖然造就顧客上門,但並非效益最大,甚至在做賠本生意;投資不足,顧客將轉向同業消費,造成蕭條倒閉。故以最佳化,最適化的理念經營管理服務復原策略才是正本清源之道。 顧客在接受服務後,滿意的狀態描述,事實上是一個動態變化的過程,透過隨機動態規劃狀態空間及馬可夫鏈狀態模擬,可以有效地預測服務復原策略運用與顧客滿意度的狀態變化情形。馬可夫決策過程之應用折現成本法,正符合商業將本求利的本質,是為服務復原策略效益評估模式極佳的工具。其步驟如下:(一)規劃隨機變數的機率狀態空間。(二)以機率矩陣作相關線性規劃。(三)馬可夫決策過程最佳折現成本運算。本研究特以符合其特性的餐飲業為例,研究發現:動態觀察的結果可以驗證靜態觀察的準確性;而服務的滿意狀態正是動態變化的系統層次。許多研究僅將顧客滿意狀況作關聯性探討,在企業管理上的功效仍是有限;本研究為彌補前述缺憾,並驗證服務復原效益影響構面的正確性,特以隨機動態規劃為啟始,顧客(忠誠度、回購率)的滿意狀態為馬可夫鏈狀態空間,同時考量成本及獲利的整體效益,建立起服務復原策略效益評估模式。

並列摘要


Due to the service epoch is comming, business is becomming to attach importance to service system management. The manager devotes to promote customer-satisfied, than to posses customer value and, to buy again, and for permanent in trade. The contribution of service recovery is customer value, the trade’s fatalism is to make money, that an investment goes to far for service recovery, than that make to waste. Optimality is the best policy, In fact, the alternation of customer-satisfied is an alternation state. Dynamic Programming can be used to describe the alternation of customer-satisfied state. Markov chain state (sample) spaces for probabilistic dynamic programming can explain and predict the alternation of customer-satisfied. Markovian decision process models of Policy-improvement algorithms for finding optimal polices: (1) Set random variables and state (sample) space by probability distributions. (2) Linear programming formulation. (3) Finite-period Markovian decision process and the method of successive approximations. It’s good for explanation and prediction of trade to make money. The case is the service system of restaurant. To examine the static state model’s accuracy of service recovery’s benefit. Note that a finite number of time periods are now being considered. This problem is analogous to deterministic dynamic programming, except that the Markov system evolves according to some probabilistic law of motion rather than evoking in a deterministic fashion. The deterministic dynamic programming solution is suggestive of the solution to this “probabilistic” dynamic programming problem. To establish an evaluation model for service recovery strategy by Markov chain decision processes.

參考文獻


45.沈煥順,民國八十九年,以問題解決為導向的企業流程改造概念模式之探討,私立中原大學,企業管理研究所,碩士論文。
68.劉世足,民國九十年六月,顧客對服務品質認知與服務復原策略關係之研究,私立中原大學,企業管理學系,碩士論文。
62.蔡政宏,民國八十九年十二月,休閒遊樂業服務品質測量工具應之實證研究,正修學報,十三期,p.241~255。
63.林國瑞等,民國八十九年十二月,台北市立體育學院推廣教育非學分班(游泳訓練班)之教學及服務品質調查研究,北體學報,八卷, p.129~145。
64.劉宗其、李奇勳、黃吉村、渥頓等,民國九十年一月,服務失誤類型、補償措施與再惠顧率之探索性研究─以CIT法應用於餐飲業為例,管理評論,零卷一期, p.6~97。

被引用紀錄


洪彬皓(2002)。應用模糊馬可夫決策發展自由化電業市場多狀態競價策略〔碩士論文,中原大學〕。華藝線上圖書館。https://doi.org/10.6840%2fcycu200200190

延伸閱讀