This paper analyzed the spreading effect of industrial sectors with complex network model under perspective of econophysics. Input-output analysis, as an important research tool, focuses more on static analysis. However, the fundamental aim of industry analysis is to figure out how interaction between different industries makes impacts on economic development, which turns out to be a dynamic process. Thus, industrial complex network based on input-output tables from WIOD is proposed to be a bridge connecting accurate static quantitative analysis and comparable dynamic one. With application of revised structural holes theory, flow betweenness and random walk centrality were respectively chosen to evaluate industrial sectors’ long-term and short-term spreading effect process in this paper. It shows that industries with higher flow betweenness or random walk centrality would bring about more intensive industrial spreading effect to the industrial chains they stands in, because value stream transmission of industrial sectors depends on how many products or services it can get from the other ones, and they are regarded as brokers with bigger information superiority and more intermediate interests.

In the face of the anti-globalization trend and the shrinking of the global value chain, ensuring the safety of the global layout of the industrial chain and the sustainability of each country’s internal intermediate product production cycle has become an important new development strategy for all countries. The sustainability of the internal and external cycles of production systems is closely related to global value chains. Based on the world input-output model, we define the trade pattern of intermediate goods in various countries from the perspective of trade intermediary attributes, and propose two indicators by which to measure the dependence of China on the global value chain in the process of “dual circulation” development: the degree of vertical specialization (VSD) and the import share of domestic total consumption (IMS); China’s super-large market leads to low values of both VSD and IMS. China’s high-tech industry has the highest degree of external dependence in the process of participating in dual circulation, and there has been a fluctuation cycle since 2009. The external dependence of different industries shows heterogeneity.

The input-output table is comprehensive and detailed in describing the national economic system with complex economic relationships, which embodies information of supply and demand among industrial sectors. This paper aims to scale the degree of competition/collaboration on the global value chain from the perspective of econophysics. Global Industrial Strongest Relevant Network models were established by extracting the strongest and most immediate industrial relevance in the global economic system with inter-country input-output tables and then transformed into Global Industrial Resource Competition Network/Global Industrial Production Collaboration Network models embodying the competitive/collaborative relationships based on bibliographic coupling/co-citation approach. Three indicators well suited for these two kinds of weighted and non-directed networks with self-loops were introduced, including unit weight for competitive/collaborative power, disparity in the weight for competitive/collaborative amplitude and weighted clustering coefficient for competitive/collaborative intensity. Finally, these models and indicators were further applied to empirically analyze the function of sectors in the latest World Input-Output Database, to reveal inter-sector competitive/collaborative status during the economic globalization.

The input-output table is very comprehensive and detailed in describing the national economic systems with abundant economic relationships, which contain supply and demand information among various industrial sectors. The complex network, a theory, and method for measuring the structure of a complex system can depict the structural characteristics of the internal structure of the researched object by measuring the structural indicators of the social and economic systems, revealing the complex relationships between the inner hierarchies and the external economic functions. In this paper, functions of industrial sectors on the global value chain are to be distinguished with bipartite graph theory, and inter-sector competitive relationships are to be extracted through resource allocation process. Furthermore, quantitative analysis indices will be proposed under the perspective of a complex network, which will be used to bring about simulations on the variation tendencies of economies’ status in different situations of commercial intercourses. Finally, a new econophysics analytical framework of international trade is to be established.

In the context of economic globalization, innovation has become a major drive for the sustainable development of enterprises, which emphasizes the importance of studying the influencing factors of enterprise innovation output. The purpose of this study is to clarify the influence mechanism of different indicators on enterprise innovation output, and then provide relevant suggestions for improving enterprise innovation ability. This paper takes 562 enterprises in Chaoyang Sub-park and Fengtai Sub-park of Zhongguancun in Beijing within the time span between 2015 and 2016 as the research objects, and constructs a comprehensive indicator system of influencing factors for enterprise innovation output from the perspective of enterprise co-opetition relationship, factor input and environmental factors. A quantitative model of innovation output and influencing factors was built and then solved, via spike-and-slab sparse function and stepwise regression, aiming at analyzing the influence of different indicators on enterprise innovation output. In addition, this paper also classifies enterprises according to their innovation level and explores the influence of indicators on different types of enterprises. The innovation of this study lies in the modeling of competition and cooperation between enterprises and the establishment of a relatively comprehensive indicator system of influencing factors for enterprise innovation output. The results show that the degree of technological collaboration between enterprises, the level of financing and the degree of capital and labor input in innovation activities have significant positive effects on enterprise innovation output. On the contrary, product and service competition, as well as capital competition, tends to have a negative impact, which urges enterprises to pay heed to the intensity of competition faced by products and services as well as the diversity of financing sources and investment targets to reduce the negative impacts. In addition, enterprises with different levels of innovation should take customized measures in terms of factor input and co-opetition relationships, in that some indicators such as network structure indicators negatively influence the innovation output of enterprises with lower levels of innovation, but has positive impacts on those with higher levels of innovation.