Most real world systems are complex in nature and agent based modeling allows researchers to create models with the required complexity and test out what-if scenarios to understand the behavior of the system. These models are helpful in understanding emergent phenomena and help to devise policy interventions. We aim to develop powerful population data generation, simulation, and visualization systems to enable researchers to understand how very large populations (hundreds of millions of agents) interact and behave.
We are currently working on a variety of problems in this area, from trying to find good ways to clean census data, to validating synthetic demographics. Much of this work feeds into (and is implemented in) the Bharatsim project, a collaboration between Ashoka University and ThoughtWorks, funded by the Bill & Melinda Gates Foundation and the Mphasis F1 Foundation.