国家关键技术选择:三维综合指数方法研究

The selection of national key technology: Research on three-dimensional comprehensive index method

国家间关键技术竞争正在加剧,如何有效、准确地选择出中长时期内应重点发展的关键技术是一个极具价值的命题。本文深入研究关键技术的概念、选择原则、选择指标、数据处理和分析方法等,提出国家关键技术选择的三维综合指数方法,相关结论如下:(1)国家关键技术是指在未来较长时期内,能对发展全局和国家安全产生重要影响的技术;(2)选择原则包括重要性、战略性、前瞻性、实用性和可行性等五个方面;(3)关键技术选择的综合指数包括研究开发水平、综合重要程度和综合可能性等三个维度;(4)考虑专家判断的一致性,以变异系数和Kendall系数修正德尔菲调查结果,使其更具科学性与稳健性;(5)引进Kruskal-Wallis秩和检验和Probit定序回归方法,结合专家的认同度构建实现可能性指数,并引入三角模糊函数法,可使专家权重更稳健。

Science and technology has become the decisive factor to promote the economic and social development, and the competition between countries will gradually focus on the key technologies field. How to effectively and accurately select the key technologies that should focus on in the long term be developed in the future medium and long term is a significant research issue. Based on literature review, this paper deeply research the definition, selection principle, indicator system, data processing, and selection method of key technologies and put forwards the method of three dimensional composite index for national key technology selection. The main conclusions are as follows. Firstly, the definition of national key technology in this paper is intended to meet the needs of economic and social development in the long period and play a key role in improving industrial competitiveness and ensuring national security so as to achieve sustainable competitive advantage the technology that can bave an important impact on the overall development and national security in a long period of time. This definition fully recognizes the timeliness and long-term nature of key technologies. Secondly, five selection principles of national level key technology are importance, strategic, foresight, practicability and feasibility. Thirdly, this paper integrates the three dimensions of research and development level, comprehensive importance and comprehensive possibility to form a comprehensive indicators of key technology selection Fourth, this paper uss the coefficient of variation and Kendall coefficient to modify the Delphi survey results to make more scientific and robust. Fifth, considering the possibility index, Kruskal-Wallis rank sum test and Probit ordered regression method are introduced to construct the experts’ identification degree. And triangular fuzzy function method is introduced to make the weight given by experts’ judgment more stable in the three-dimensional comprehensive index method.