以資料包絡分析法評估郵局之技術效率

Measuring Technical Efficiency of Postal Offices: DEA Approach

林村基(Erwin T. J. Lin) ； 劉盈禎(Ying-Jhen Liou)

《臺灣企業績效學刊》 4卷1期 (2010/12) Pp. 97-123

https://doi.org/10.6697/TBPJ.201012_4(1).0005

摘要

台灣的郵政服務已自民國92年1月1日改制為國營公司，其改制後之經營績效為一頗值得探討之課題，本文以資料包絡分析法估算台中地區61家郵政支局的技術效率，選取兩個產出變數為郵務部門之「信件與包裹投遞總數」及儲匯部門之「臨櫃辦理交易之客戶數」，投入變數則為支局「員工人數」及「營業場所樓地板面積」兩項。實證結果顯示，在CRS假設下有7家支局為相對有效率，在VRS假設下則有15家支局為相對有效率。整體而言，平均純技術效率為0.852，隱含各支局必須在投入要素不增加之情況下，等比例擴增兩項產出17.4%，以達技術有效率。

關鍵字

資料包絡分析法 ； 郵局 ； 技術效率

並列摘要

Taiwan's postal service has undergone a successful transformation as a state-owned public corporation since January 1, 2003. It is interesting to investigate the performance of postal service after becoming a public corporation. In this paper, we estimate technical efficiency for 61 post offices in Taichung area by a nonparametric approach known as data envelopment analysis (DEA) method. We choose the total number of mail handled (including collection and delivery), and the total number of customers served as outputs, meanwhile, the number of labor and capital-related variables are selected as inputs. Our empirical results indicate that of the post offices studied, 7 and 15 offices are evaluated as efficient based on the assumptions of CRS and VRS, respectively. On average, the pure technical efficiency of the postal offices is 0.852, implying that the postal offices should expand their outputs by 17.4 percent, keeping inputs unchanged, so as to attain technically efficient.

並列關鍵字

Data Envelopment Analysis ； Post Office ； Technical Efficiency

參考文獻

吳月萍(2005)。中華郵政於公司化前後績效評估－資料包絡分析法之應用(碩士論文)。國立中正大學國際經濟研究所。

黃敏玲(2004)。階段資料包絡分析法之應用～探討台南郵局所轄支局經營績效二(碩士論文)。國立中正大學國際經濟研究所。

黃旭男、林進財、張元哲(1997)。郵政總局分支機構經營效率及效率變動之評估。交大管理學報。17，125-144。

Ali, A.I.,Seiford, L.M.(1993).The mathematical programming approach to efficiency analysis.The Measurement of Productive Efficiency: Techniques and Applications.(The Measurement of Productive Efficiency: Techniques and Applications).

Andersen, P.,Petersen, N.C.(1993).A procedure for ranking efficient units in data envelopment analysis.Management Science.39(10),1261-1264.