蝴蝶蘭被喻為台灣近年來最具市場實力與潛力的花卉產品，其產值與銷售額均居所有花卉之冠，在有利可圖之下民間業者及國外花卉公司紛紛投入，造成該產業競爭激烈。當經營者面對全球蝴蝶蘭產業的競爭，如何在有限的資源下，發揮最有效率的營運成果，以因應國內及全球性之競爭環境，已經是該產業經營的重要課題。 本研究運用資料包絡分析法(DEA)來探討蝴蝶蘭產業生產單位之經營績效。以人事費用、能源費用及生產成本作為投入變數，營業收入及稅前利益作為產出變數，透過DEAP 2.1軟體運算得到之各項效率值加以分析，以協助管理者了解自身營運成效的好壞，並提供營運效率不佳的單位改善的方向與幅度，作為績效改進之依據。 研究結果發現：從整體來看20個DMU各項效率值平均皆在0.9以上，與有效率(1.000)差距不大，顯示四個生產場平均營運良好。而從參考群體分析與差額變數分析，可提供無效率DMU學習對象以及改善投入產出量的建議值。根據麥氏指數分析，以MPI生產力變動指數而言，整體的平均值為1.020，表示整體生產場的生產力是呈現進步的情形，並已接近固定規模報酬或漸向長期最適生產規模趨近。 Phalaenopsis has been Taiwan’s most well-known and well accepted flower product in the global market. The gross output value and sales turnover from Phalaenopsis exceeded that generated by other flowers in Taiwan. Many private companies as well as overseas enterprises have entered this industry causing fierce competitions. Facing competitions from global suppliers, the most important issue in today’s Phalaenopsis industry is to maximize productivity and achieve the most efficient performance even with limited resources. In this research, Data Envelopment Analysis (DEA) method was adopted for analyzing the productivity of Phalaenopsis production plants. Personnel expenses, energy expenses, and production cost were adopted as input variables, and sales revenue and earnings before tax were adopted as output variables. Different efficiency values generated by DEAP 2.1 software were analyzed to assist the managers in understanding the performance of their plants. The result of analysis may offer directions for improvement and may also serve as a reference for the plant seeking to enhance performance. From the research, it was found that the average technical efficiency value for 20 DMUs was above 0.9; the result was quite close to the efficiency value of 1.000. It indicated that the four production plants were performing well. The results from reference group analysis and slack variable analysis can be utilized as reference in the process of enhancing performance for inefficient DMUs and improving productivity outputs. The Malmquist productivity index was 1.020 in average; it indicated that the overall productivity was going in an upward direction and it was close to the state of Constant Returns to Scale (CRS) or it was moving toward optimum scale in the long run.
