摘要
高技术产业的技术创新效率是一个国家、区域和产业科技研发实力与经济竞争力的重要体现。随着高技术产业的蓬勃发展,技术创新效率在提升高技术产业的国际竞争力方面占据越来越重要的位置。分解技术创新效率的方法也需要从以往单系统的“黑箱”结构向多系统“灰箱”结构转变。本文提出了权重加和模型和乘积模型的三阶段网络结构模型,该模型的多变量的改进算法替代单变量算法一定程度解决了各子阶段优先级不同导致的整体效率最优解不唯一的问题。另外,学者往往单独考虑权重加和法或乘积法,而忽略了测算模型的系统误差,模型中同时运用权重加和法与乘积法两种方法探索了模型测算结果的误差效应。利用这两种模型评估我国高技术产业多系统技术创新效率;同时考虑具有共享投入与自由中间产出的R&D投入资源分配方式,对两种模型的实证结果进行对比分析。分析结果表明:权重加和模型有放大效率的风险,会导致效率值失真;相对于权重加和模型来说,乘积模型更适用于多系统的效率求解。此外,资源合理分配相对于不考虑生产系统内部R&D投入资源合理分配来说,在一定程度上能够增大整体效率。因此,在提高生产系统整体效率方面,合理分配R&D投入资源比一味加大R&D投入更有效。
The R&D efficiency of high-tech industry is an important indicator of a country, region and industry’s technological innovation capability and economic competitiveness. With the rapid development of high-tech industry, the efficiency of technological innovation is playing a more important position in promoting the international competitiveness of high-tech industries. The evaluation criteria and requirements of technological innovation of high-tech industry or enterprises are significantly higher than general industrial enterprises. The technological innovation needs high calibre of personnel, long research and development cycle, and high complexity of innovation system structure. These complex factors increase the difficulty of R&D efficiency measurement in high-tech industries or enterprises. The method of decomposing the efficiency of technological innovation needs to change from the "black box" structure of the previous single system to the multi-system "grey box" structure.At present, measuring the efficiency of technological innovation in high-tech industry mainly run up against the following problems. First of all, the traditional measuring method of the technological innovation efficiency basically treats the production system as a "black box", which means considering the external input resources and output results, nevertheless ignoring the resource utilization in the process of technological innovation in the production system. It is impossible to accurately understand the utilization rate of input resources and the stage effectiveness of output results in the innovation process. On the one hand, although leading to increasing R&D input resources, but still unable to effectively improve the efficiency of production system technology innovation;on the other hand, it will lead to the final being enlarged output results of the technology innovation effectiveness. For example, when measuring the efficiency of technological innovation, the system is regarded as a "black box", that is, a single system structure. Traditional measurement methodscan not judge the effective utilization of R&D investment resources, so that when improving the efficiency of technological innovation, it can not accurately find the breakthrough point to increase R&D investment. Secondly, the traditional measurement methods can not distinguish whether all the scientific and technological achievements are effectively transformed into economic achievements, which leads to the amplification of the final output effectiveness, and the "virtual high" phenomenon of the efficiency measurement results. In recent years, more and more scholars have paid attention to the "grey box" structure. Scholars have been studying the effective ways to open the traditional "black box", analyzing its complex internal structure, exploring the two-stage or even multi-stage network structure to restore the real innovation production system as much as possible, and establishing a reasonable efficiency evaluation model. The "grey box" structure can directly or indirectly reflect the stage of innovation system, and clarify the utilization situations of input resources, the transformation of intermediate output and the final output, which is helpful to improve the accuracy of efficiency evaluation. Accurately measuring the efficiency of high-tech industry or enterprise technological innovation is the prerequisite to improve the competitiveness of enterprises. Optimization of innovation system structure and rational allocation of R&D input resources are effective methods to accurately measure the efficiency of industry or enterprise technological innovation.Liang et al.(2008) proposed a strict serial two-stage model based on DEA network structure, and used the model to measure and evaluate efficiency. MaJianfeng and He Feng(2014) extended the model of Liang et al.(2008), established a Two-stage DEA model considering both shared input and free output, and put forward a weighted sum method to measure the efficiency of the Two-stage DEA model. Xie Jianhui et al.(2016) also expanded the model of Liang et al.(2008), assuming that the second stage has external input, or the first stage has free output, and proposed the centralized decision-making(product method) model and non cooperation(game theory) model to evaluate the efficiency of the generalized two-stage network structure DEA. Both the weighted sum method of Ma Jianfeng et al.(2014) and the product method of Xie Jianhui et al.(2016) adopt the method of one variable to measure the optimal solution of the overall efficiency by using the step size change. Those applications have the same problem, that is, once the first stage is selected as a variable or the second stage is selected as a variable, the range of efficiency change in the other stage is fixed. It can not be realized that when the overall efficiency reaches the optimal level, the first stage and the second stage also obtain their respective optimal solution steps at the same time. Therefore, this paper proposes an improved algorithm to solve this problem.In this paper, under the three-stage network structure model, the weight-sum model and product model are proposed to evaluate technology innovation efficiency of multiple systems in Chinese high technology industries, which explores and puts forward a multi-variable algorithm replacing the single variable algorithm to solve this problem, in order to solve the problem that the optimal solution of the whole efficiency is not unique due to the different priorities of each stage. And scholars often consider the weight adding method or product method alone, ignoring the systematic error of the calculation model. Meanwhile, it uses two methods to explore the error effect of the model calculation results. As well as, considering about the modes of R&D resource allocation with shared inputs and free intermediate outputs, and carrying out comparative analysis between the two models. The result shows that the weight-sum model has an efficiency-amplified risk which leads to the distortion of efficiency value, and the product model is more suitable for solving the efficiency of multiple systems. While compared with overlooking R&D resource allocation, the rational R&D resources allocation of the multiple systems can help to increase the overall efficiency to some extent. As a result, the reasonable distribution of R&D resources is more effective than some blind increase of R&D investment for promoting the industrial system entirety efficiency.
作者
李作志
苏敬勤
刘小燕
Li Zuozhi;Su Jingqin;Liu Xiaoyan(Faculty of Management and Econom ics,Dalian University of Technology,Dalian 116024,Liaoning,China;Management School,Tianjin Polytechnic University,Tianjin 300387,China)
出处
《科研管理》
CSSCI
CSCD
北大核心
2019年第12期31-41,共11页
Science Research Management
基金
国家自然科学基金重点课题:“新技术环境下的组织创新研究”(项目批准号:71632004)
天津市哲学社会科学规划研究项目:“碳源环境下天津滨海新区低碳承载力分析与补偿机制研究”(TJGL13-018)
关键词
高技术产业研发效率
网络结构DEA
共享投入
自由中间产出
R&D efficiency of high-tech industry
network DEA structure
shared input
free intermediate output
