基于科学的产业创新主体专利合作伙伴识别研究

近年来，发展基于科学的产业已成为全球科技和产业竞争的焦点。基于科学的产业具有开放式创新特点，产业创新主体致力于寻找合作伙伴以充分利用外部资源。专利信息涵盖丰富的技术信息、法律信息及经济信息，是创新主体寻找合作伙伴并开展创新合作的重要资源，但鲜有学者运用专利信息研究基于科学的产业创新问题。构建基于科学的产业专利合作伙伴识别研究框架，以石墨烯这一典型的基于科学的产业为例，采用专利技术组合和社会网络分析方法，从行动者网络理论的4个行动者转译环节出发，全面展现我国石墨烯产业中最具潜力的国内和跨国专利合作伙伴识别过程，以挖掘专利合作机会并提高企业竞争优势。最后，从政府、高校、企业3个层面为我国石墨烯产业创新发展提供决策参考。

Identification of Patent Collaboration Partners in A Science-based Industry

The development of science-based industry has become the focus of global science and technology as well as industrial competition. Science-based industry has the characteristics of open innovation. The innovation subject in the industry is committed to finding partners to make full use of external resources, so as to obtain patent collaboration opportunities and competitive advantages. For how to choose partners, domestic and foreign scholars mainly analyze the evaluation factors and research motivation of partners through qualitative research methods such as expert knowledge. Thus, the research results are greatly affected by subjective factors and lack certain objectivity. Patent information covers rich technical, legal and economic information. Patent information is an important information resource for innovation subjects to find partners and carry out innovation collaboration. However, there is a lack of in-depth research on science-based industrial innovation and development by using patent information, and the research on exploring patent partners in science-based industry from a predictive perspective is still blank. #br#　　Aiming at exploring patent collaboration opportunities and competitive advantages in science-based industry, this study introduces actor-network theory to the research field of science-based industry, employs patent technological portfolio and social network analysis, and proposes a new method of patent partner identification. Taking graphene, a typical science-based industry, as an example, this study proves the effectiveness of the method. After looking through related literature and consulting technical experts' advice, 67 835 graphene technology patents from 2011 to 2019were obtained downloaded from Derwent Innovation Index. This method combines the technology level with the patentee level from a predictive perspective, and deeply discusses the logical translation process of identification for potential domestic and foreign patent partners. The actor-network theory divides the translation of patent partners' identification into four processes: problematization, interessement, mobilization and enrollment. It also selects appropriate patent indicators to measure the process of interessement, mobilization and enrollment. The process of problematization aims to extract the research problems in line with the common interests of all actors according to the current situation of industrial innovation and development. This study focuses on how to find potential patent partners in science-based industries. In the process of interessement, starting from the common interests among actors, it looks for potential actors to join the network. Based on patent analysis the study identifies the technology hotspots and fronts in the science-based industry from the technology level which shows the common innovation opportunities of patentees in the industrial competition;it also verifies the patentees with competitive advantages in the science-based industry from the patentee level which indicates the potential important actors in the interactive network. The process of mobilization is designed to provide a strategy for connecting with other actors. This paper constructs the technology association network of the patentee in key technology fields, combines the technology level with the patentee level, and digs out technical focus among important patentees. Based on the common technical focus among patentees, this paper further constructs the strategy of identifying potential patent partners in the science-based industry. In the process of enrollment, this paper adopts the strategies in the mobilization process and forms an interactive network between actors. Based on the index of technical similarity, this paper constructs patent partner identification networks, identifies China's potential domestic and foreign patent partners, and explores patent collaboration opportunities and competitive advantages. #br#　　The results show that the most potential domestic patent partners in the field of graphene include Zhejiang University, Harbin Institute of Technology, Chengdu New Keli Chemical Technology Co., Ltd., Tsinghua University and Central South University; the most potential foreign patent partners in the field of graphene include Samsung Electronics Co., Ltd, LG Chem Ltd., Nanotek Instruments Inc, LEE Y T, Sungkyunkwan University, and Korea Advanced Institute of Science and Technology; the key technology areas with the most collaboration potential at home and abroad are E05-U05C(Nanofilm), L03-H05(Vehicles), L03-A02B(Nonmetal conductors-carbon and graphite), A08-S02(Solvents; swelling agents) and A10-E05B(Chemical modification by carbonization). #br#　　 The contributions of this paper are threefold. First this paper uses patent information to provide a useful supplement for the quantitative analysis of innovation and development in the science-based industries. Second it expands the existing theoretical system of science-based industry research by proposing a new methodology of patent partner identification. Lastly from the predictive perspective, this paper combines these two levels through patent technological portfolio and social network analysis, and identifies the potential domestic and foreign patent partners of the innovation subjects in China's graphene industry.#br#