基于R&D网络现实特点,考虑研发企业之间的合作程度,采用BBV模型构造R&D网络.从运行过程视角,对研发企业的技术风险进行识别.在此基础上,提出不同节点企业之间的技术风险触发机制.引入风控能力和风险大小等参数,基于SIR模型构建了R&D网络的技术风险传播动力学模型,并对该理论模型进行数值仿真.仿真结果发现:技术风险总能迅速地传播,影响范围最大时整个网络中有大约半数节点企业感染技术风险.三类技术风险中,实现性技术风险传播速度最快,影响范围最大且最难移除.在考虑边加权的无标度网络中,技术风险传播的相对缓慢,影响范围相对较小,不同风险干扰情况下网络的鲁棒性差异不大.企业抗风险能力存在临界值,如果40%的企业遭受三种技术风险,60%的企业遭受一种技术风险时能够成功移除风险,此时技术风险对R&D网络的冲击最小.本研究成果对提高R&D网络的抗风险能力具有一定现实意义.%Along with the world economy’s incessant informative and networked process, R&D firms often cannot simply rely on their own strength to cope with market change and the cross-integration trend of technology, which inevitably allow different companies to establish a cooperative relationship. However, the dynamics and uncertainty in technology market makes the whole process full of risks, in which technical risk becomes an important one. Once it occurs, it is easy to propagate and result in network paralysis. Currently, researches on technical risk in R&D network mostly focus on the static level, such as concepts defining, type identification, quantitative analysis, without taking into account the dynamic propagation process. Few dynamic studies are mostly limited to qualitative description, or only the establishment of a universal risk propagation model, which does not distinguish propagation mechanism between technical risk and other kinds of risk. Based on such background, in order to improve the anti-risk ability of the companies and reduce the probability of technical risk propagation, it is necessary to explore dynamic technical risk propagation process in R&D networks. In the first part, we review the relevant literature and find that R&D network shows the characteristics of scale free. Furthermore, considering the interaction strength between enterprises, we use BBV model to construct the R&D network based on realistic characteristics of R&D network. In the second part, we identify technical risks in R&D companies from the perspective of operational procedures, which are subdivided into strategic technical risk, organizational technical risk and realization technical risk. In the third part, we propose a trigger mechanism of technical risks among different note enterprises, and classify node enterprises into three kinds of states based on the SIR epidemic model. Then, the SIR model is improved by combining the characteristics of technology risk propagation in R&D network. The evolution rules of technology risk propagation and node state transition formula are put forward by introducing parameters such as risk-control ability and magnitude of risks. Finally, we build technical risk propagation dynamics mode of R&D network and run numerical simulation. Simulation results show that technical risk can propagate fast. About half of the node enterprises are infected when its impact reaches the maximum. In three kinds of technical risks, realization technical risk is the one which spreads the fastest and has maximum impact. Furthermore, it is the most difficult one to remove. In the weighted scale-free network, the technical risk propagates relatively slow, and its impact is smaller compared with the one in the scale-free network. The robustness of the network under different types of risk interference is similar. The anti-risk ability of the enterprises displays a critical value. If 40% of enterprises suffer from three kinds of technical risks and 60% of firms suffer from one kind of technical risk, the impact caused by technical risk is minimum on R&D network. In view of the simulation results, we propose some recommendations for R&D enterprises to prevent and control technical risk. For example, technical risk event response plans should be more targeted, which means taking various solutions for different types of technical risk. The construction of the prevention mechanism needs to develop from the actual situation of enterprises, and strengthen the prevention of realization technical risk. Enterprises should focus on the whole network and try to achieve the strongest anti-risk ability of the whole network instead of only considering the case of one single enterprise and increasing the cost of anti-risk blindly when establishing strategies. To sum up, this study enriches the theoretical knowledge of technical risk propagation, and improves the SIR model, which is more in line with the characteristics of technical risk propagation in R&D network. At the same time, it has practical significance for R&D enterprises to fully identify all kinds of technical risk, formulate corresponding prevention and control strategies, and improve the anti-risk ability of the whole network.
展开▼
