Agent-based Modeling Of Sequential Innovation Diffusion On Network

It is the necessary requirement of technology progress to promote innovation diffusion, especially for the developing countries catching up the developed countries. To find a good answer to solve this problem, the paper focuses on the innovation diffusion phenomenon on network.The paper establishes an agent-based model to simulate the behaviour of the enterprise agents on a space-technology network to invest on different R&D activities to achieve sequential innovation. The agents will be put in different structure network and different patent system to study the effect on agents' behaviours and innovation diffusion, to prove the small-world feature of sequential innovation diffusion on network.In Chapter 2, the paper discusses the theoretical basis of the study, the sequential innovation theory and the patent theory, describes the features of the small-world network, and brings forth the structure designing of the model of sequential innovation diffusion on network.In Chapter 3, the paper explains the designing of the model of the sequential innovation diffusion on network in detail by three parts as market and patent system, the agent behaviour, and network structure, and also builds the market and institutional arrangement, defines the agent behaviour function in entrepreneurship, production and R&D activities, determines the network structure of agents' interaction.In chapter 4, the paper shows the implementation approach of the agent-based simulation system of the sequential innovation diffusion from technological view, and explains the function of each module in this system, such as multi-model simulation, parameterized model setting, dynamic display of the simulation and basic statistics function.In Chapter 5, the simulation is performed to analyze the agent behaviour function and innovation diffusion by using the sequential innovation diffusion network ABM system. The study also compares the difference among different patent arrangement and different network structure and draw conclusions as below:The enterprise agents obtain patent by innovation and become the source of diffusion, and the other agents can acquire the technology in diffusion process by imitation or paying for technology adoption. The variation of patent system arrangement will affect the number of diffusion source and the R&D behaviour of the agent. The space-technology 2D network we built in the paper proves to be a small-world network and the process of sequential innovation diffusion on network have the characteristics of small-world network.