Advanced Transport Modelling VT07 (1N1154)

Course administrator: Lars-Göran Mattsson, lgm[at]infra.kth.se, tel: 08-790 85 63
Course assistant: Erik Jenelius, jenelius[at]infra.kth.se, tel: 08-790 96 35

Transport models are used in many contexts to analyse policies and their effects, such as effects on emissions, cost and benefit, redistribution or effects on accidents. Policy examples include new roads, intelligent transport systems, new public transportation, increased or decreased cost for private car or public transportation, and flexible hours for shopping, working or schools. After the course, the student should be able to program a transport demand model and understand its limitations, and use a model to analyse a current policy measure, such as congestion charging.

Aim

After completing the course, the student should be able to 

• use the theory for discrete choices and stochastic utility maximization to formulate models related to transport demand, in particular travel frequencies, destination choice and mode choice 

• program a transport forecasting model for analysing a real and current policy measure, such as congestion charges 

• account for models and theory for route choice 

• explain network equilibrium models and equivalent optimization formulations. Account for pros and cons with the concept of network equilibrium 

• use software for network equilibrium on a real world application to analyse traffic flows. 

• use decision support systems to analyse realistic planning problems 

• search relevant literature for a current application within transport demand modelling

Syllabus

• Theory for discrete choices, stochastic utility maximization, econometric estimation, entropy methods and gravity model. 

• Theory and algorithms for network equilibrium 

• Cost benefit valuation and effect evaluations: accidents, emissions and value of time. 

• Literature seminar

First, the theory is presented within lectures, which are the followed up by computer assignments, normally four.

Finally, in a literature study assignment, the student will search information to find a solution to a given problem. The suggested solution will be presented as a written report and at a seminar.

Prerequisites

Knowledge of mathematical models, as defined by mathematical courses mandatory within the programs for Civil Engineering and Urban Management (Traffic Engineering), Engineering Physics (Optimisation and Systems Theory), Vehicle Engineering or Mechanical Engineering (Systems Engineering).

Follow-up

Mathematical economics, Probability theory, Optimization modelling. Applications within the programs for Urban Planning and Traffic Engineering.

Requirements

Written exam (3p) and assignments (2p).

Required reading

Hensher, D.A., and Button, K.J., 2000, Handbook of Transport Modelling, Pergamon Press. In particular chapters 1, 3, 5, 9, 10, 13, 17, and 19.

Train, K., 2003, Discrete choice methods with simulation, Cambridge University Press.
Included chapters: 2, 3, 5, 6, 8.

Extensively:
3.4
5.3-rest of chapter 5
6.5
6.8
8.3.2-8.4
8.6-rest of chapter 8

Note that chapter 8.5 is included "intensively".

Also, all of chapter 10 is included extensively.

Extensively means that it is included, but if you know the rest you will pass the exam. For higher grades, knowledge from chapter 10 may be good.

Additional material

Lecture 3 (Mar 23):
OH Logit 1
OH Logit 2
Logit theory

Discrete choice excersises:
Excersises 1
Excersises 2
Excersise 1, 3 and 9 on the first paper were done during the Tutorial 1 on Apr 10.

Lecture 9 (Apr 17, on meso and microsimulation):
Presentation part 1
Presentation part 2

Lecture 12 (Apr 26, on valuation and CBA):
OH Part 2

Instructions for Lab 1 (corrected version)

Old exams (from the course 1H1401 OPTTLA)

Schedule

Location for lectures and tutorials if not otherwise stated is the seminar room at TLA, Teknikringen 78 B 1st floor.