On the multi-agent stochastic simulation of occupants in buildings

This paper introduces a new general platform for the simulation of occupants' presence and behaviours. Called No-MASS (Nottingham Multi-Agent Stochastic Simulation), this generates a synthetic population of agents, predicts their presence and, in the case of residences also their activities and inferred locations, as well as their use of windows, lights and blinds. Using the Functional Mockup Interface, No-MASS is coupled with EnergyPlus: EnergyPlus parses environmental parameters to No-MASS which in turn parses back the energetic consequences of agents' behaviours. After describing the architecture of No-MASS and the form of the integrated models, we demonstrate its utility through two use cases: a house and an office. We close by outlining how No-MASS has been extended to more comprehensively simulate the behaviours of agents occupying multiple buildings, including behaviours for which data is scarce, social interactions between agents, and a generalization of No-MASS to simulate electrical devices and their interactions.