A generic agent-based framework for modeling business ecosystems: a case study of electric vehicle home charging

Abstract

Modeling and simulation have been popularly used for system investigation and evaluation. With proper evaluation, distribution system operators can decide on a reasonable course of action for encouraging energy flexibility and make predictions on the recommended timing and magnitude of system updates under different scenarios. However, there is no efficient tool for system operators to quickly set up and perform simulations of alternative scenarios for system updates before planning their course of action, without much experience with programming or system modeling. This paper proposes an agent-based modeling framework for developing agent-based simulation models of business ecosystems that can be applied to multiple evaluation scenarios by simple configuration of agents and roles. There are two steps in this proposed framework: Step 1 – Interface and role interactions design and Step 2 – Agent architecture and connections design. In addition, the framework depends on a pre-step that covers mapping and architecture development of the business ecosystem to be modeled. The framework is demonstrated with a case study of an energy business ecosystem consisting of an electricity distribution grid with 137 connected domestic consumers. The case study shows that the proposed agent-based modeling framework supports the development of agent-based models for simulating energy business ecosystems. To verify the behavior of the developed agent-based simulation models, a verification procedure of the agent models is briefly discussed, which includes unit, integration, and system testing approaches similar to the ones used in software testing.