Journal of Artificial Societies and Social Simulation · 2016 · 247 citations · 0 references
Natural SciencesEnvironmental ManagementDynamical SystemsEnvironmental ModelingUncertainty QuantificationAgent-based ModelingAgent-based SimulationComplex Systems ModelingEcological ModelingEnvironmental Systems ModelingSensitivity AnalysisComplexityParameter Sensitivity AnalysisRenewable Resource DynamicsResource Dynamics ModelingSensitivity Analysis In Dynamic ModelsSensitivity Analysis In Simulation ModelsSensitivity Analysis Of Model OutputsSensitivity Analysis Of Model Parameters
Existing sensitivity analysis methods are inadequate for agent‑based models, whose multi‑level, nonlinear interactions and emergent behavior limit the information obtainable from classical input‑output linkages. This study evaluates the performance of three established sensitivity analysis methods for agent‑based models. The authors applied extended one‑at‑a‑time, variance‑based model‑fitting, and Sobol decomposition techniques to a simple case study of free‑roaming, reproducing agents harvesting a diffusing renewable resource. All three methods provide useful insights but none offers a complete view, so the authors recommend extended OAT as the initial approach for uncovering mechanisms and patterns in agent‑based models.
Existing methodologies of sensitivity analysis may be insufficient for a proper analysis of Agent-based Models (ABMs). Most ABMs consist of multiple levels, contain various nonlinear interactions, and display emergent behaviour. This limits the information content that follows from the classical sensitivity analysis methodologies that link model output to model input. In this paper we evaluate the performance of three well-known methodologies for sensitivity analysis. The three methodologies are extended OAT (one-at-a-time), and proportional assigning of output variance by means of model fitting and by means of Sobol’ decomposition. The methodologies are applied to a case study of limited complexity consisting of free-roaming and procreating agents that make harvest decisions with regard to a diffusing renewable resource. We find that each methodology has its own merits and exposes useful information, yet none of them provide a complete picture of model behaviour. We recommend extended OAT as the starting point for sensitivity analysis of an ABM, for its use in uncovering the mechanisms and patterns that the ABM produces.