The Modeling Framework for Behavior, Energy, Autonomy, and Mobility

  • What is BEAM?

    BEAM extends the Multi-Agent Transportation Simulation Framework (MATSim)
    to enable powerful and scalable analysis of urban transportation systems.


    Embedding discrete choice models in agent-based simulations.

    People make decisions every day that impact their engagement with the transportation system. Through stated preference and revealed preference analysis, we have sophisticated models of how people make these decisions.


    BEAM enables us to directly embed these discrete choice models in a virtual environment where we can then see how human preferences will impact the performance of the system as new technologies and policies are implemented.


    Coupling mobility, vehicle energy consumption, EV charging behavior, and charging control strategies with power sector simulation models.

    BEAM enables detailed analysis of the energy impacts of changing mobility trends as well as the potential impacts of EV adoption and the benefits of managing charging in order to support grid reliability and access emerging markets for grid flexibility services.


    The future of transportation and electric systems will be dominated by the emergence of autonomy and distributed control.

    From smart devices to fully autonomous vehicles, transportation and the energy economy will be transformed by our ability to enhance vehicles and appliances with the ability to intelligently sense and respond to the world around them.

    The interconnected impacts of these rapidly evolving technologies simply cannot be understood in isolation. BEAM will therefore serve as a test bed for new ideas in managing the charging profiles of EVs or assessing the opportunities and challenges associated with fully autonomous vehicles active on 20th century road networks.


    Mobility is endogenous to BEAM​

    BEAM enables analysis of new technologies in a manner that fully respects the fluid nature of mobility in urban systems. A new transit stop reshapes the mobility behavior of people who live and work nearby; dynamic pricing on Uber changes the loading on public transit in real-time; new EV chargers increase both EV adoption and day-of decisions on which vehicle to drive.


    BEAM provides an integrated analytical environment to sort out these tradeoffs between the multitude of competing mobility options and services.

  • BEAM Blog

    The visualization above shows trips by agents as simulated in BEAM moving throughout the San...
    BEAM needs to be scalable. Our grand vision for BEAM is to simulate entire states or nations, not...
    In mobility simulation modeling, routing is by far the most computationally expensive aspect of...
  • Multi-Modal Urban Systems

    Agents in a BEAM choose between driving, public transit, walking, biking, or using shared or networked mobility services like Uber/Lyft/etc. They evaluate this choice by using a simulated trip planning service and then evaluating and sampling from a personal utility function.

  • Electric Vehicles

    BEAM enables highly resolved simulations of EV charging behavior and interactions with charging infrastructure.


    See BEAM in action simulating the detailed interactions between EV drivers in the San Francisco Bay Area and the charging infrastructure.


    For more details on BEAM features, download this overview.

  • The BEAM Team

    BEAM is a joint effort between Lawrence Berkeley National Laboratory and the Institute for Transportation Studies at UC Berkeley.

    Colin Sheppard


    Colin Sheppard is a co-manager of model development and is leading the LBNL effort to use BEAM to analyze vehicle grid integration opportunities, challenges, and value.

    Dr. Rashid A. Waraich


    Dr. Waraich worked at ETH Zurich for several years as a MATSim developer of core capabilities and energy related extension. He is a co-manager of BEAM model development and is leading the LBNL effort to integrate BEAM with UrbanSim.

    Michael Zilske


    Dr. Zilske recently graduated from TU Berlin, where he played a central role in maintaining and extending the MATSim framework and conducted research on analyzing mobility patterns from passive data sources.

    Sid Feygin


    Sid Feygin is a Ph.D student at UC Berkeley doing researching at the interface of transportation science and artificial intelligence. Sid played a key role in designing the foundational architecture of BEAM.

    Dr. Anand Gopal

    Emeritus Advisor

    Dr. Anand R. Gopal is a Research Scientist in the International Energy Studies Group of the Energy Technologies Area (ETA) at the Lawrence Berkeley National Laboratory (LBNL).He is the Deputy Leader for ETA’s Sustainable Transportation Energy Program.

    Dr. Alexei Pozdnukhov

    Emeritus Advisor

    Alexei holds a Ph.D. in computer science from EPFL, Switzerland, following his research in machine learning methods and computer vision that he carried out at IDIAP Research Institute in Martigny, Switzerland. He then worked on remote sensing and spatial data mining at the University of Lausanne (UNIL). Most recently, he co-directs the Smart Cities Research Center at UC Berkeley.

    Andrew Campbell


    Andrew is a Ph.D. candidate in the UC Berkeley Transportation Engineering program. His research employs machine learning and agent-based microsimulation to predict travel location decisions.

    Sangjae Bae


    Sangjae Bae is a Ph.D student at UC Berkeley. He researches EV grid integration, optimal operations, and scalable spatial analytics.

  • Contact Us

    We want to hire and/or collaborate with passionate modelers and

    transportation / electric systems engineers. Please reach out!

All Posts