Combination of microscopic and macroscopic traffic simulation for by jbw10297


									                                      Project description

         Combination of microscopic and macroscopic traffic simulation
                   for modeling Kungsgatan in Norrköping

In recent years the interest for micro simulation techniques to model traffic in congested road
networks has gained momentum. The interest is often due to development of real-time traffic
management and control strategies that react to the highly dynamic and variable nature of traffic
conditions and driver behavior. The micro models are generally used for networks consisting of a
couple of intersections. For planning analysis at city level, static or dynamic assignment tools are
more suitable than microscopic models. Changes in a small part of a network often influence a larger
area and a microscopic simulation analysis of a constrained area has often to be combined with a
macroscopic assignment analysis on city level.

Aim and purpose
The aim of this project is to build, calibrate, validate and experiment with a small traffic network
using the microscopic traffic simulation software AIMSUN in combination with a traffic assignment
model. The area along Kungsgatan shall be studied. The purpose of the project is to give insights in
how traffic simulation models in combination with static assignment models can be used for
analyzing real world traffic control problems.

Project supervisor: Johan Janson Olstam
Examiner: Clas Rydergren

The project shall be carried out according to the project organization description.

Project group
The project should be conducted by a group of 3-4 students. This project requires that the students
have taken the course TNK095 Traffic planning and simulation.

Project description
The project covers the following steps

    -   Plan and carry out a traffic simulation study for the area described below.
    -   Collect necessary input data to a traffic simulation model and analyze and evaluate the
        uncertainty in the input data.
    -   Create, calibrate and validate a traffic simulation model in Aimsun.
    -   Find a suitable way to use output data from the micro model as input to the macro model
        and vice versa.
    -   Decide on which outputs that are relevant to analyze for different types of traffic simulation
        applications and be able to conduct suitable output analyzes. The student should also be able
        to reflect on how the output data is influenced by road design, driver behavior, travel
        demand, etc.
    -   Run the micro and macro model iteratively to get an “equilibrium” solution.

The project group needs to collect the necessary data for building the model. The model shall be
calibrated and validated. The group should also propose an alternative design of the network and
analyze this design and compare it to the original design using the combined traffic simulation and
traffic assignment model.

Suitable design changes may for example be:

    -   Change of regulation to traffic signal, roundabout, yield, or stop.
    -   Change of signal timings and signal coordination.
    -   Change the geometric intersection design: number of lanes, left or right turn pockets, bus
        lanes, bus stop placements, etc.

Kungsgatan is an urban arterial, carrying a big part of the traffic travelling between the southern and
northern parts of Norrköping. It is one of few possibilities to pass in southbound and northbound
direction through Norrköping. There are lots of pedestrians and bicycles both crossing and travelling
along Kungsgatan. The emission levels at Kungsgatan are often higher than allowed. Therefore, the
municipality wants to redesign Kungsgatan to decrease the amount of traffic travelling on
Kungsgatan. They want to evaluate the traffic effects both at a local and global (city) level.

For this project, an “initial” Aimsun network and an “initial” macro (Matlab based) network are
available. The Aimsun network is from 2005 and consists of 5 signal controlled intersections (3 traffic
actuated signals and 2 pedestrian crossings). Your task is to update the Aimsun model with respect to
geometry, traffic demand, signal settings, etc. so that it corresponds to the current design. The main
task is to find a suitable way to combine the microscopic and the macroscopic model.
Project goals
The project shall result in a method for combining an Aimsun model and a Matlab based static
assignment model. At least one alternative design shall be suggested, evaluated and compared to the
basic scenario using the combined model.

Appropriate delimitations are specified by the project group.

Project organization
Project supervision

A project leader and contact person has to be appointed for in each project. Each project leader
informs the course examiner of their existence and present the project group and project number.

Every project is appointed a supervisor. One appointment with the supervisor each week is
scheduled. Other appointments have to be booked directly with the supervisor.

Project requirements and documentation
There are a number of decision points. Each decision point is associated with a deliverable. Each of
these points is given in the list below. Each decision shall be discussed with the supervisor, or handed
in to the examiner. Preliminary decision points:

October 26: Project start

November 4: Project specification and project plan delivery

December 7: Final report submission

December 11: Final report presentations

December 18: Project assessment and written opposition submission

Contact persons
Johan Janson Olstam,

Clas Rydergren,

Project grading
Each project can be graded, Fail, 3, 4 or 5, or the corresponding ECTS grade. Each step below will be
evaluated regarding the complexity, realization, systematic treatment of input data, modeling and
output data.

For the grade 3, at least the following steps have to be carried out

    -   A detailed project specification and project plan, carefully written in English.
    -   A systematic and well documented data collection
    -   An Aimsun model over the above described area.
    -   A method for combining the Aimsun model with a traffic assignment model of Norrköping.
    -   Verification and calibration of the model output.
    -   Evaluate one design change in the modeled traffic system and compare to the basic scenario
        with regard to both local and area wide effects.
    -   A final report carefully written in English.

For the grade 4, the grade 3 steps have to be carried out with very good results and the following
steps have to be carried out:

    -   A systematic and complete data collection for the described area.
    -   Calibration based on data collected on gap acceptance, travel times and queue lengths.
    -   Validation of the output from the Aimsun model.
    -   Evaluate one major (complex) design changes in the modeled traffic system and compare the
        results with each other and to the basic scenario, using suitable statistical methods.

For the grade 5, the grade 3 and 4 steps have to be carried out with extremely good results and the
following steps have to be carried out:

    -   Systematic modeling of all signalized intersection in the area, either with time or traffic
        actuated signal timings.
    -   Relate to the results from the combined model to earlier documented studies with combined
        models in the literature.
    -   Systematic validation of the model, based on input data that is not used for calibration.

To top