Publication | International Conference of the International Building Performance Simulation Association 2017

Simulating the Behavior of Building Occupants using Multi-agent Narratives

A Preliminary Study in a Generic Hospital Ward

Abstract

Simulating the Behavior of Building Occupants using Multi-agent Narratives: A Preliminary Study in a Generic Hospital Ward

Davide Schaumann, Simon Breslav, Rhys Goldstein, Azam Khan, Yehuda E. Kalay

International Conference of the International Building Performance Simulation Association 2017

In architectural design it is of cardinal importance to anticipate how people will use a building prior to its construction and occupation. Conventional multi-agent simulation methods represent occupant movement and activities to assess the day-to-day performance of households and office buildings. In these environments, behavior is usually driven by individual schedules or comfort-related actions. In other kinds of settings, such as hospitals, airports, or factories, behavior is driven by codified sets of collaborative procedures which dynamically adapt to the spatial and social context. To address these building types, we propose a narrative-based approach whereby a variety of behavior patterns involving multiple occupants can be simulated and visualized. A scheduling method coordinates the narratives using Operations Research techniques. The method is demonstrated through a preliminary study, which involved collecting data in an existing hospital environment, modeling narratives computationally, and simulating them in an abstracted layout of a generic hospital ward.

Download publication

Related Resources

Publication

2023

Generative design for COVID-19 and future pathogens using stochastic multi-agent simulation

Proposing a generative design workflow that integrates a stochastic…

Publication

2019

Occupancy Visualization towards Activity Recognition

We present a sensor visualization system that integrates data streams…

Publication

2022

COIL: Constrained Optimization in Workshop on Learned Latent Space

Constrained optimization problems can be difficult because their…

Publication

2019

Demo: Semantic Human Activity Annotation Tool Using Skeletonized Surveillance Videos

Human activity data sets are fundamental for intelligent activity…

Get in touch

Something pique your interest? Get in touch if you’d like to learn more about Autodesk Research, our projects, people, and potential collaboration opportunities.

Contact us