Agent-based services for the validation and calibration of multi-agent models

Agent-based modelling in the form of multi-agent models has been increasingly applied to the simulation of spatial phenomena in silico. Validation and calibration are recurrent problems. The complexity of these models with large numbers of parameters can make validation procedures intractable. In this paper, the novel concept of using agent-based technologies to create services that assist in the validation and calibration of multi-agent models is developed. Such agent-based services offer an efficient solution where large numbers of model runs need to be carried out. In this paper, the agent-based services are collaborative sets of agents that perform calibration and sensitivity analysis as a key task in model validation. In a case study, the prototype agent-based validation services are implemented for a multi-agent wayfinding model as a means of proof-of-concept. The case study demonstrates how agent-based services can be deployed for testing the robustness of emergent patterns through sensitivity analyses and used for model calibration. (C) 2008 Elsevier Ltd. All rights reserved.