(106f) Agent-Based Modeling and Simulation to Complex Adaptive Eco-Industrial Systems

Eco-industry is an important way to improve the material efficiency, to resolve the environment pollution, and to achieve the sustainable development. An eco-industrial system is characterized by diversity, autonomy, opening, emergence and evolution. Its evolution is resulted from the dependent decision-making, the interaction, and the symbiosis of all enterprises in the system. Aiming to the complexity characters, the complex adaptive system (CAS) theory is applied to investigate the development mechanism and the evolution process of eco-industrial systems by the agent-based modeling and simulation. The research results can be used to guide the eco-industry practice in China.

The basic principle of CAS theory is that the adaptivity brings up the complexity. In the evolution of a complex system, agents of the system can change their behaviors by learning; thereby, they can finally harmonize and adapt with their environment each other. All complex adaptive systems share four properties (aggregation, nonlinearity, flowing, and diversity) and three mechanisms (categorization by tagging, anticipation through internal models, and decomposition in building blocks).

An agent-based modeling and simulation method, which is suitable for eco-industrial systems, is proposed in this paper. The modeling design process is pyramidal, from element, flow, module, agent, aggregation to system, which is from microscope to macroscope. "Element", "flow", "module", and "agent" are the micro design, and "aggregation, system" are the macro design. The emphasis of the model is the various agents in the system. "Element", "flow", "module", and "agent" are designed to describe the characters and behaviors of the agents. "Aggregation" and "system" are formed by interactions of all agents under some interactive mechanisms. The macro phenomena of the system therefore come out because of the interactions.

CAS theory and agent-based modeling and simulation method are applied into a practical example: Kaiyang national demonstration eco-industrial base of phosphorus-coal-based chemical industry in China. The agent-based models involve many factors which may affect systems' behaviors, such as technology, policies, environment, society and market. The model can clearly express complex hierarchy and relationship in the system. A graphic, dynamic and interactive simulation platform is developed. The study results show that better economic and environmental benefits of the system can be achieved under the direction of eco-industrial idea. Meanwhile problems in the system evolution process are also found, and the reasons and solutions are analyzed and presented. This could be very helpful for eco-industrial practice.

Key words: Eco-industry, Complex adaptive system(CAS), Agent, Model, Simulation