Shape mobility for a more sustainable future

NEVS Engineering department is creating a strong foundation of skills for new technologies and innovations. We have a committed passion to create and deliver iconic and desirable global products and services. By challenging conventions, we design premium electric vehicles and mobility experiences that are simple, engaging and distinctive, but that also shape a brighter, cleaner future for all.

We are now looking for 2 students for Master Thesis work concerning:

Traffic Simulation Setup Calibration (Multivariate Optimization)

Description

Traffic simulation is a vital component that provides insight into planning of mobility services and fleet management. As with any simulation setup, the input data heavily affects the outcome or result of the simulation, thus making this step a key focus for any case study or scenario. Setting up a microscopic traffic simulation environment involves allocation of various input parameters such as origin, destination, purpose, demographic and behaviour to one or many agents in the simulation. This thesis will focus on the optimization of several input parameters to setup a simulation scenario in a microscopic traffic simulation tool known as SUMO (Simulation of Urban MObility). The goal of the optimization is to minimize the difference between the outputs of the simulation and the available data.

In the figure below it shows a graph that represent the problem on a high level, assuming that we are only interested in calibrating with respect to traffic flows. The problem here is to generate trips between and within zones that result in flows between nodes that are as close as possible to the available data. Some flows are more important to get right, so weighting should be applied. Note also that we may only specify origin and destination zone for each trip and the simulation tool will provide the optimal route for each trip. What is optimal will in turn depend on the observed flows, so this is a dynamic system.

Figure 1 Network representation

Another example of what we need to consider is that some nodes are more likely to be origins or destinations and it will vary between different times of day, e.g. schools, workplaces, shops.

The input to simulations is various routes based on geographic districts of different size. The goal is to find the optimal distribution of these routes while considering the profile of the districts (with regards to households, workplaces, schools and demographics)

We are looking for 2 students from Engineering Mathematics and Computational Science, Computer Science - algorithms, languages and logic, Systems Control and Mechatronics or similar background.

Please apply as soon as possible! The selection work and interviews will be ongoing continuously and the opportunity will be closed as soon as right candidates have been found for it.

If you have any questions, please contact:

Camilla Lindeblad - Chapter Leader SW/Control T&S, Phone number: +46 73 966 64 89.

Katarina Karlsson - HR Business Partner & Recruitment, Phone number: +46 73 966 61 06.