Mobility Simulation
Mobility is a key enabler for efficient functioning of modern cities and urban areas. It impacts almost every aspect of our daily life; thus, it is crucial to understand how to properly adopt mobility infrastructure to rapidly a changing dynamic environment. Moreover, predicting trends and demand in mobility helps to increase efficiency of upcoming infrastructural changes and policies.
The project is lead by Aleksandr Saprykin
In order to do so, several methods are applied, one of them being the mathematical modelling of transportation systems. In reality, mobility behaviour models can be too complicated for analytical or numerical assessment. Mobility simulation helps to reveal complex underlying relations and behavioural patterns, providing substantial information for more efficient policies.
Increased computational capabilities make it possible to use disaggregated approaches, such as agent-based modelling, at large scales. A simulation scenario can consist of millions of agents, each of them with a detailed individual daily plan. Being able to assess a larger number of hypotheses under present and future scenarios can provide greater insight into mobility processes for policy makers.
To perform a simulation with a high level of detail and within short time frames, we explore two ways to utilize modern hardware to accelerate computations:
- Massively parallel mobility simulation models with GPU support
- Distributed computations on high-performance clusters
You can read more about the Methods used to address our research questions.