@unpublished{17,
  keywords = {Simulation;, Transportation, Urban Mobility},
  author = {Chinmaya Samal and Abhishek Dubey and Lillian Ratliff},
  title = {Mobilytics-Gym: A Simulation Framework forAnalyzing Urban Mobility Decision Strategies},
  abstract = {The rise in deep learning models in recent years has led to various innovative solutions for intelligent transportation technologies. While some prediction models focus on predicting the state of network efficiently and accurately, such as estimating traffic congestion, transit delay and so on, other models use those predicted states to find a set of sequential decisions that commuters need to make to travel from their origin to destination. The performance of these models is often evaluated using prediction accuracy. There is a growing need to understand the overall impact of such models on the societal scale. In this paper, we leverage MATSim, an agent-based simulation framework, to incorporate various decision-making models and provide a standardized environment to evaluate the efficacy of these models in terms of its system impact. For example, we describe the integration of a model that captures the altruistic behavior of an agent in addition to the disutility of a user proportional to the travel time and cost. This model can then be used to evaluate the sensitivity of an agent to the system disutility and the monetary incentives given by the transportation authority of the city. We show the effectiveness of the approach and provide the analysis using a case study from the Metropolitan Nashville area.},
  year = {2019},
}