MAICE

Search SwRI

               SwRI Home

Screen display of an urban protest scenario, provided by  SwRI-developed MAICE Station, a software analysis tool that uses agent-based modeling to evaluate emergent behaviors during simulated group interactions.

MAICE Station screen display during a hostage event scenario. Data can be examined to secure a detailed analysis of actions and behaviors.

Current events with national and international security implications have markedly increased the need for modeling systems in the area of human aggregate and crowd behavior.

MAICE Station

In response, Southwest Research Institute has developed the Modeling of Aggregates of Individuals and Crowd Evaluation (MAICE) Station™, a software analysis tool using agent-based simulation and modeling to evaluate emergent behaviors based on the actions and interactions of simulated autonomous individuals in one or more groups.

MAICE Station scenarios and associated models simulate the simultaneous actions of numerous independent agents to recreate, analyze and predict the outcomes of complex human behaviors and interactions. In MAICE Station, individuals make independent decisions, follow and lead, show aggression and resistance, and communicate and interact with the environment, contributing to the overall dynamics of the crowd. This analysis approach generates emergent aggregate behaviors from the interactions of lower level individuals.

In 2010, Southwest Research Institute released MAICE Station Version 3.0, the robust crowd modeling software platform that offers unique individual and crowd behavior analysis capabilities. MAICE Station allows law enforcement, military and event management personnel to analyze, model and research the effects of the behaviors of hundreds or even thousands of individuals in a potentially volatile crowd.

Scenario Visualization Using MAICE

MAICE Station includes modeling utilities for constructing aggregates and scenarios using extensive sets of attributes and behavior settings. Rather than model a crowd as a single unified entity, MAICE Station simulates crowds by aggregating hundreds or thousands of unique, independent individuals. MAICE Station provides a variety of scenario visualization tools and can be used as an analysis platform as well as a briefing and training tool for event management or post event evaluation.

Designed to run on a standard desktop PC, MAICE Station supports rapid prototyping of crowd simulation and modeling scenarios. Analysts can build a scenario using any map image background and populate it with completely customized individuals. During runtime, simulated individuals can:

• Communicate between themselves

• Act on the environment

• Act on other individuals

• Make decisions affecting the progress of the scenario

An analyst may pause, fast forward, single step, or slow down scenario execution speed. Analysts have complete control over the behaviors and display appearance of individuals in the scenario. A zoom window provides close-up views of areas of interest. Simulated individuals may be monitored visually during runtime or log data may be examined for detailed analysis of actions and behaviors.

MAICE Station Features

Support for the use and analysis of lethal and non-lethal countermeasures is also provided. All stimuli are completely customizable—both the intended effects built into the stimulus as well as specific effects on individuals and aggregates.

A complete set of file management utilities is available for saving and reusing:

• Scenarios

• Templates

• Aggregate individual data sets

• Presets

• Profiles

• Master files

• Objects

• Customized components

The features also allow a user to:

• Analyze multiple timelines

• Evaluate the effects of changes in deployment of resources

• Study variations in behavior

Several experiments have been conducted with the U.S. military to validate behavior models and scenarios created with MAICE Station. MAICE Station has also been integrated with industry standard constructive simulation programs such as RealWorld and OneSAF with impressive results, including demonstrating a MAICE Station generated crowd of 10,000 individuals within a OneSAF scenario.

MAICE Station Applications

Example applications of agent-based modeling include:

• Supply chain optimization and logistics

• Modeling of consumer behavior, including word of mouth, social network effects

• Distributed computing

• Workforce management

• Portfolio management

• Traffic congestion

• Evacuation modeling

In these and other applications, the system of interest is simulated by capturing the behavior of individual agents and their interconnections. Agent-based simulation tools can be used to test how changes in individual behaviors will affect the system's emerging overall behavior. Agent-based models have been used to analyze:

• The spread of epidemics the threat of biowarfare

• The growth and decline of ancient civilizations

• The human immune system 

For more information about MAICE Station capabilities and techniques at SwRI or how you can contract with SwRI, please contact Susan Porter, at susan.porter@swri.org or (210) 522-2812.