Modelling is an essential tool for decision makers to anticipate the future and evaluate the impact of possible interventions under different scenarios. The modelling techniques and forecasting tools most commonly used today, such as 4-stage travel demand models, were developed in a context of clear distinction between mass public transport and private vehicles, and are not well adapted to the scenarios brought about by the myriad of new mobility solutions and the increasing complexity of travel patterns enabled by information and communication technologies. Additionally, conventional approaches to transport and traffic modelling were driven to a large extent by the scarce travel demand data that could be collected through traditional methodologies, which often limits their ability to make the most of the more detailed and granular data available today.
Transport models in the age of big data
New big data sources, such as mobile phone records and smart card data, offer a rich set of opportunities to measure the variability of demand, observe elasticities to changing conditions and learn from planned and unplanned events. The availability of high-detail, high-resolution data for large population samples also opens new paths for the estimation of discrete choice models and the implementation of more disaggregated and dynamic modelling paradigms, such as multi-agent activity-based models, better suited to represent features such as demand-responsive transport services, trip chains and multimodal trips. Nommon brings together its mobility data analytics skills with innovative transport modelling approaches to help our clients understand and forecast the impact of smart mobility solutions and new mobility patterns.