Skip to content

faq 107514910

Billy Charlton edited this page Sep 5, 2018 · 2 revisions

calculating system optimum using MATSIM

by Ido Klein on 2017-01-25 09:42:26


Does anybody know of modules in MATSIM that allow the calculation of the network's system optimum?


Comments: 1


Re: calculating system optimum using MATSIM

by Ihab Kaddoura on 2017-02-14 09:18:18

Dear Ido,


the system optimum is addressed in several modules / approaches.

The basic idea is to add a correction term to the agents’ generalized cost function to provide the incentive towards a system optimal behavior. Applying the basic economic principal of Pigouvian taxation, the correction term can be interpreted as a monetary toll which needs to be set according to the marginal external costs, e.g. delays, noise damages, air pollution costs, (…) that are imposed on other travelers, residents or the society as a whole.

In MATSim, adding a correction term requires two steps:

1) Adjust the agents’ scoring.
2) Adjust the agents’ routing, i.e. the travel disutility computation which is used for the least cost path calculation.


There are different use cases / applications…

… in context of evacuation scenarios: Please have a look at the evacuation contrib (org.matsim.contrib.evacuation) as well as the PhD thesis of Gregor Lämmel: https://depositonce.tu-berlin.de/handle/11303/3290

… in context of optimal pricing schemes.

Congestion pricing:

Approach (a): Queue-based congestion pricing: Compute the delay imposed on following travelers and charge the external delay cost from the causing agent (see package playground.vsp.congestion)

Approach (b): List pricing: Compute the congestion level for each road segment and time bin and set/update the toll accordingly (see package playground.ikaddoura.decongestion)

some references (for approach a):

* I. Kaddoura. Marginal congestion cost pricing in a multi-agent simulation. Investigation of the Greater Berlin Area. Journal of Transport Economics and Policy, 49(4):560–578, 2015. https://svn.vsp.tu-berlin.de/repos/public-svn/publications/vspwp/2014/14-19/

* I. Kaddoura and B. Kickhöfer. Optimal road pricing: Towards an agent-based marginal social cost approach. VSP Working Paper 14-01, TU Berlin, Transport Systems Planning and Transport Telematics, 2014. https://svn.vsp.tu-berlin.de/repos/public-svn/publications/vspwp/2014/14-01/


Environmental effects (air pollution + noise): Map the environmental damages back to road segments and transport users and internalize the costs (see org.matsim.contrib.noise and org.matsim.contrib.emissions + playground.vsp.airPollution).

some references:

B. Kickhöfer, K. Nagel; Towards High-Resolution First-Best Air Pollution Tolls; in Networks and Spatial Economics, Springer https://svn.vsp.tu-berlin.de/repos/public-svn/publications/vspwp/2013/13-18/

I. Kaddoura, L. Kröger, and K. Nagel. User-specific and dynamic internalization of road traffic noise exposures. Networks and Spatial Economics, 2016. DOI: 10.1007/s11067-016-9321-2. https://svn.vsp.tu-berlin.de/repos/public-svn/publications/vspwp/2015/15-12/

I. Kaddoura and K. Nagel. Simultaneous optimization of traffic congestion and noise exposures. VSP Working Paper 16-06, TU Berlin, Transport Systems Planning and Transport Telematics, 2016. https://svn.vsp.tu-berlin.de/repos/public-svn/publications/vspwp/2016/16-06/

A. Agarwal, B. Kickhöfer, K. Nagel; The Internalization of Congestion and Air Pollution Externalities: Evaluating Behavioral Impacts; submitted to IATBR 2015 https://svn.vsp.tu-berlin.de/repos/public-svn/publications/vspwp/2015/15-11/


Hope this helps. Best regards,
Ihab

Clone this wiki locally