行政院國家科學委員會專題研究計畫 成果報告
技術創新之衡量與評估--以台灣高科技產業之廠商為例
計畫類別： 個別型計畫 計畫編號： NSC93-2213-E-110-041- 執行期間： 93 年 08 月 01 日至 94 年 07 月 31 日 執行單位： 國立中山大學企業管理學系(所) 計畫主持人： 盧淵源 報告類型： 精簡報告 處理方式： 本計畫可公開查詢中 華 民 國 94 年 8 月 1 日
行政院國家科學委員會補助專題研究計畫
█ 成 果 報 告
□期中進度報告
The Model of Measurement and Evaluation for Technology Innovation
in Taiwan High-Tech Firms
台灣高科技廠商技術創新衡測與評估之研究
計畫類別：■ 個別型計畫 □ 整合型計畫
計畫編號：NSC 93－2213－E－110－041
執行期間： 93 年 08 月 01 日至 94 年 07 月 31 日
計畫主持人：盧源源 國立中山大學企管系教授
共同主持人：
計畫參與人員：王俊賢 國立中山大學企管系 博士班研究生
林志用 國立中山大學企管系 碩士班研究生
成果報告類型(依經費核定清單規定繳交)：■精簡報告 □完整報告
本成果報告包括以下應繳交之附件：
□赴國外出差或研習心得報告一份
□赴大陸地區出差或研習心得報告一份
□出席國際學術會議心得報告及發表之論文各一份
□國際合作研究計畫國外研究報告書一份
處理方式：除產學合作研究計畫、提升產業技術及人才培育研究計畫、
列管計畫及下列情形者外，得立即公開查詢
□涉及專利或其他智慧財產權，□一年□二年後可公開查詢
執行單位：國立中山大學企管系
中 華 民 國 94 年 07 月 31 日
中文摘要 此研究是由模糊積分與模糊測度所組成來 衡量廠商創新能力與創新積效評估。其相關 評估準則與量測之建立乃是以技術創新領域 中之 Chain-link 模式與經濟合作發展組織 （OECD）所發展出來的奧斯陸（Oslo）手冊 為依據。此外，此研究並將實際應用在台灣 高科技產業之廠商創新活動評估上。研究結 果可以作為管理者在技術創新能力與創新成 本節省與效率提升之參考。 Abstract
This study attempts to consist of the fuzzy integral and fuzzy measure in order to measure the degree of technology innovation and hi-tech firm innovation performance. The relevant criteria and metrics of this research are adopted chain-link model of technological innovation domain and Oslo manual developed by OECD. Furthermore, this evaluation model will be used to real case for hi-tech firms in Taiwan. The results show that this study provided the manager of the technology innovation competence reference for their innovation relatively cost down and efficient improvement.
Key words: innovation performance, fuzzy
measure, fuzzy integral, Oslo manual
Introduction
Hi-tech firm face the pressure that a world market for products, and the increasing tendency of global competition, which need to have technological innovation if they hope to improve firms’ competitive competence as well as competitive advantages. Technology innovation measurement is a difficult task there is no widely accepted method for has been developed so far. Many methods exist, and they all have their advantages and disadvantages according to how they are used. This lack of uniformity in the framework for technological innovation compliance testing creates a serious problem for companies which seek registration to these standards.
According to Oslo manual (OECD, 1992), that provide powerful methods of depth analysis technology innovation especially that engaging in technological innovation firms. Hence, this study applies the characteristics list in Oslo manual (OECD, 1992) to measure the technology innovation. n this research, we hope that the proposed measurement model can measure the technological innovation deeply. There are many empirical studies have pointed out that corporative innovation performance is extremely difficult because of the existence and interaction of many innovation processes such as Meyer-Kraher, (1984); Chakrabarti and Souder (1984); Hagedoorn and Cloodt, (2003). To describe the interaction between input of innovation and output of innovation that the traditional statistical method is not suit to handle effect of interaction. Thus, this study employs fuzzy measure and choquet integral to deal with interactive phenomenon in firm innovation activities. In order to describe the interaction among information sources, fuzzy measure can be used. (Wang, and Klir, 1992; Murofushi, et al., 1994; Wang, et al., 1999) Particularly, firm innovation processes are interactive and non-linear system that fuzzy measure, and fuzzy integral can be handle this situation. The purposes of this research are: 1) To analyzes the extent of, and basis for, achieving technology innovation within the Taiwan high-tech firms, and 2) An evaluation system should be established for technology innovation to detected and assess technology innovation in order to improve their efficiency in technological innovation and increase firms’ competitive advantage.
Indicators of innovative
In this study the evaluation technology innovation performance of hi-tech firms are adopted the Oslo manual (OECD, 1992, 1997) with frame of chain-link model and
modified them with proposed indicators and some relative literature to suit firm innovation performance evaluation. The Oslo manual takes the original frame of concepts, definitions and methodology and updates than to incorporate survey experience and improved understanding to the innovation process and also to take in a winder range of industries. It also provided technology innovation measurement factor and criteria. The well-known innovation model of chain-link model provides to assess the importance of innovation indicators for different stages of the innovation process which cover and link both the innovation process and overall firm performance. The scope of the Oslo manual is to provide a baseline for further individual refinement and description of technology innovation (OECD, 1992, 1997) and it is developed to a baseline of technology innovation model with associated innovation characteristic and indictors reveal in the Figure 1.
Outcome and conclusion
Firm innovation performance has three stage of innovation should be assess. The following table presents various firm innovation performances values (c)∫hdgand λ
values in different stage of innovation processes. From Table 4 firm A, it is found that there are high degrees dependent and mutual influences at firm innovation process, that are λ=-0.986, among indicators in the innovation input ( '
1
x ). Similarly, the other
innovation stage, such as innovation throughput ( '
2
x ) and its λ=-0.882, innovation
output (x ) and its 3' λ=-0.994 showed that that their λ values close to -1 which represents a complete dependent and mutual influence among those innovation indicators of them. Moreover, the Choquet integral (c)∫hdg in Eq.
(13) is employed to determine the aggregated value of each innovation stage based on innovation indicators and the aggregate value of firm overall innovation performance (OIP) value based on stage of innovation listed in Table 6.
Table 4 Summary and aggregated value of different
stage of innovation processes
Inn. S t. Ind. ) (xi h gi λ g (c)∫hdg (λ value) 1 x 0.565 0.616 gλ(x3) 0.681 2 x 0.621 0.739 ( , ) 4 3 x x gλ 0.883 3 x 0.837 0.681 gλ(x3,x4,x2) 0.978 ' 1 x 4 x 0.672 0.616 gλ(x3,x4,x2,x1) 1.000 0.777 (-0.986) 5 x 0.621 0.558 gλ(x6) 0.558 6 x 0.569 0.616 gλ(x6,x5) 0.871 ' 2 x 7 x 0.532 0.558 gλ(x6,x5,x7) 1.000 0.593 (-0.882) 8 x 0.375 0.558 gλ(x11) 0.747 9 x 0.587 0.805 gλ(x11,x9) 0.954 10 x 0.551 0.739 gλ(x11,x9,x10) 0.993 ' 3 x 11 x 0.892 0.747 gλ(x11,x9,x10,x8) 1.000 0.811 (-0.994)
Table 1 presents the measured ratings h(xi)
and their grades of importance g as i evaluated by experts, the λ value of g , λ obtained by Eq. (10) and the aggregated value
∫hdg c)
( for the three stage innovation. In
table 1 (c)∫hdg represents the overall
innovation performance value of the evaluators to the three innovation stage of innovation input ( ' 1 x ) equal to 0.777, innovation throughput ( ' 2 x ) equal to 0.593, and innovation output ( ' 3 x ) equal to 0.811. Use
the same procedure above that we can found various hi-tech firms different stage innovation performance values listed in Table 5.
Table 5 Various firms aggregated value of different stage of innovation processes
∫hdg c)
( (λ value)
Inn. S
t.
Firm A Firm B Firm C Firm D
' 1 x 0.777 (-0.986) 0.851 (-0.986) 0.673 (-0.967) 0.754 (-0.996) ' 2 x 0.593 (-0.882) 0.640 (-0.882) 0.642 (-0.893) 0.820 (-0.983) ' 3 x 0.811 (-0.994) 0.813 (-0.954) 0.847 (-0.966) 0.804 (-0.991)
Table 6 Overall innovation performance of firm A Firm A Inn. h(xi) gi gλ (c)∫hdg (λvalue ) ' 1 x 0.777 0.806 ( ) ' 3 x gλ 0.674 ' 2 x 0.593 0.746 ( , ') 1 ' 3 x x gλ 0.946 (OIP) ' 3 x 0.811 0.674 ( , , ') 2 ' 1 ' 3 x x x gλ 1.000 0.790 (-0.982)
In the assessing the overall innovation performance, the following table presents four hi-tech firms overall innovation performance value. Table 7 presents the (c)∫hdg value
obtained from different stage of innovation processes (see Table 4). Choquet integral yields final rank of innovation performance shown in Table7.
Table 7 Rank of innovation performance obtained by Choquet integral
Firm λ value (c)∫hdg Rating Firm A -0.982 0.790 4 Firm B -0.958 0.831 1 Firm C -0.967 0.799 3 Firm D -0.993 0.814 2
In this study, proposed methodology for innovation performance and ranking can be used for hi-tech firm overall innovation performance evaluation. Choquet integral can be used in deal with experts’ subjective judgments and provide credible, accurate ranking of alternative. During the decision making procedure a small number of input data cover the whole range of actual innovation grades.
References
1. Chakrabarti, A. K. and Souder, W. E., “Critical Factors in Technological Innovation and Their Policy Implications.” Technovation, Vol. 2, 1984, pp.255-275.
2. Hagedoorn, J. and Cloodt, M., “Measuring Innovative Performance: Is there an Advantage in using Multiple Indicators?” Research Policy, Vol. 32, 2003, pp.1365-1379.
3. Meyer-Krahmer, F., 1984. Recent Results in Measuring Innovation Output, Research Policy, Vol. 13, pp.175-182.
4. Murofushi, M. Sugeno, M. Machida, Non-monotonic Fuzzy Sets and Systems, 1994, Vol. 64, pp.73-86.
5. OECD, The Measurement of Scientific and Technological Activities,
1992, 1997.
6. Wang, Z. and Klir, G. J., Fuzzy Measure Theory, New York: Plenum Press, 1992
7. Wang, Z., Leung, K. S., and Wang J., A Genetic Algorithm for Determining Nonadditive set Functions in Information Fusion, Fuzzy Set and System, Vol. 102. pp. 463-469.
Innovation Performance Innovation Input (x1') Innovation Throughput (x2' ) Innovation Output (x3') Measure
Figure 1 Hierarchy model for firms technology innovation measuring
Innovate Stage Indicators Metrics
R&D Intensity (x1) R&D expenditure (x2) Percentage of researchers(x3)
Non-R&D innovation expenditure (x4)
Number of patents (x8)
Sales of innovative products (x9) Yearly growth rate of sales (x10)
Export intensity of innovation products (x11) New product announcements (x5)
Innovation density (x6) Intensity of collaboration (x7)
