大型研究機構績效評估模型之建構

Abstract

知識經濟與全球化競爭的時代，科技對一個國家經濟發展居關鍵地位。研究機構的績效是一個國家經濟發展與競爭力很重要的一環，因此被賦予促進經濟成長的重要角色及協助建立新興產業與促進產業升級與創新的任務。因為科技資源很有限，為了提升研究機構的績效，各國政府對一些由政府提供研究經費的大型研究機構績效評估更重視，紛紛頒布相關的規定辦法，要求這些研究機構從體檢開始提升組織績效與競爭力。我國國科會與經濟部也於91年分別公佈了「中華民國科技組織績效評鑑作業手冊」與「經濟部科技專案研究發展計畫作業手冊」。

相對於已開發國家，我國是出口導向國家，企業規模相對較小，研發資源更有限，因此工業技術研究院（簡稱工研院）等研究機構就被政府付賦此重要任務與角色。實務上，大型研究機構本身已經有建立相關績效評估機制，本研究透過文獻探討，發現學術研究無論國內外針對研究機構組織績效評估的文獻相對很少；研究機構績效評估主要也集中在人事、財務、技術指標、客戶滿意度等。對大型研究機構組織內效率與生產力研究相對尚缺。本研究經分析各國主要大型研究機構特性發現工研院相對產出效益與對產業貢獻顯著，因此以工研院為實證個案研究對象。有關如何應用客觀公正與科學方法之量化模式來評估不同部門間效率問題，經文獻探討分析，本研究以DEA來構建大型研究機構之組織結構的效率評量模型，並以工研院各研發單位多投入及多產出效果的評量模式，實證證明DEA的交叉分析模式結合多目標模式產生之結果可信度高且較以往績效評量客觀公正，同時可使院與所、中心的經營團隊更能了解各單位本身真正的營運效率。此模式比較質的分析，有利於組織間溝通與建立共識，節省時間，同時其能發覺隱性效率問題提前改善。本研究同時以多變量的因子分析與群落分析來收歛歸納各研究單位之屬性及角色定位為技術研發創新型、技術前瞻型、工業技術服務型、技術引進及合作開發型、以及資訊服務型等五個群落；應用正準相關分析，分析出工研院各單位研發投入及研發產出間的正準相關係數及樣本得點，並利用迴歸分析求得樣本得點之迴歸式。另外本研究運用層級分析法(AHP)透過問卷調查該院中高階研發與營運主管，再針對各單位主管進行問卷，再輔助第二階段針對工研院最高層的經營團隊為主的訪談，建構出工研院績效評量指標之權重，其中最主要共識結果是絕大多數主管認為「專利」最能表現研究機構之價值。最後根據工研院的相關績效評估辦法與活動，綜合整理出乙套研究機構績效評估IPOSI模型。基本上大型研究機構組織任務與目標相同，特別是國內經濟部所屬財團法人研究機構，本研究評量模式可適用。

Science and technology development is known to be a significant factor in a country’s economical strength. The government-funded research organizations can play an important role in promoting economical growth. Taiwan’s Industrial Technology Research Institute (ITRI) is a notable example.

Taiwan’s government has placed increasing importance on not-for-profit research organizations, based on the ITRI model, to help moderate-sized industries that cannot afford R&D on their own. As such organizations proliferate, a systematic method to measure their performance is highly desirable in order to make best use of the limited resources. The same method can also help research organizations monitor their effectiveness and make management decisions. However, there is no generally agreement on the method to be used, although some tools are available.

This study focuses on ITRI, beginning with a comparison with a number of large research organizations in several nations. The data envelopment analysis (DEA) method is then used to evaluate organizational structure and efficiencies of various ITRI departments. The contributions to industries are separately verified. This analysis can improve communications between research organizations and their funding agencies. This approach also makes the management decisions and communications within the organization easier.

A factor analysis is performed to find characteristics and roles of various research divisions. The subsequent cluster analysis identifies five main types: Innovative R&D, Visionary technology, Technical Services, Technology Acquisition and Collaborative R&D, and Information Service. Further analysis based on Canonical Correlation yields relative coefficients and relations among various input/output elements, pointing the way for improving the organizational performance. Depending on the weighting factors, there is certain flexibility in allocating resource for the most desired result. For example, more capital investment may be called for in some cases, but not in others.

The basic missions and goals of large research organizations are similar in essence. The method and findings for ITRI here are expected to be applicable in other organizations.