Agent-based models under uncertainty

Background: Monte Carlo (MC) is often used when trying to assess the consequences of uncertainty in agent-based models (ABMs). However, this approach not appropriate epistemic rather than aleatory, that is, it represents a lack knowledge variation. The free-for-all battleship simulation modelled here inspired by children’s game, where each an agent. Methods: The contrast MC implementation against interval for uncertainty. In case, our form uncertain radar. method, occludes status agents (ships) and precludes analyst from making decisions about them real-time. Results: In highly environment, after many time steps, there can be ships remaining whose unknown. contrast, any invariably tends conclude with small number steps. Thus, misses quantitative conclusion. some results are generated implementation, e.g. identities surviving ships, which revealed nearly mutual though fewer total compared MC. Conclusions: We have demonstrated possible implement intervals ABM, but broad, may useful generating overall bounds system do provide insight on expected outcomes trends.