SWUTC Research Project Description
Title of Project: Using GPS and ITS Data to Calibrate the Micro Simulation Model VISSIM
Project Number: 167963
Principal Investigator:
Fengxiang Qiao
(713) 313-1915
P.I. Affiliation: Texas Southern University
Project Monitor:
Hameed Mrchant
Senior Planner
(713) 739-4000
Project Status: Active
Date Started: 9/1/05
Estimation Completion Date: 8/31/06
Estimated Cost - Current Fiscal: $34,700
Estimated Cost - Total Planned: $34,700
Project Summary:
Project Abstract:
VISSIM, as one of the most popular microscopic, behavior-based multi-purpose traffic simulation programs, has become an invaluable and cost-reducing tool for transportation planning and traffic engineering. VISSIM simulates lane changing behaviors and driving behaviors, including two versions of the Wiedemann model: urban driver and freeway driver. The car-following mode of the freeway driver model involves 10 tunable parameters: CC0 through CC9. The calibration of these and other parameters are very essential to the simulation results and thus have attracted more and more attentions from researchers. In this projects, the real time ITS data and on-road GPS data will be used as one of the resources for the calibration and evaluation of the simulation model VISSIM. The test beds will be selected in Houston with various scenarios designed, while the calibration procedures and corresponding computer programs will be constructed. Genetic Algorithm and other possible optimal searching algorithms will be used as the criteria of calibration. With the calibrated parameters, simulated and measured travel variables (such as speed, travel time, queue length, etc.) under different travel patterns will be compared and evaluated. Corresponding suggestions will be recommended to researchers and engineers.
Project Objectives:
The goal of this research is to develop a methodology and the corresponding procedure for the calibration of VISSIM parameters. To achieve this objective, the research will:
Task Descriptions:
Task 1: Conduct literature search and develop the analytical framework
The literature review will focus on the practices of calibrating parameters of VISSIM and other simulation software. Literature will be found through scientific papers, reports and websites. Based on the literature search, the analytical framework for VISSIM parameter calibration will be synthesized.
Task 2: Identify the proper calibration methodology and procedure
Before the identification of calibration methodology, traffic patterns to be simulated (e.g. free flow, fair, slightly congested, highly congested) should be firstly identified. Possible optimal searching algorithms such as the Genetic Algorithm (GA) may be used as the criteria of calibration. Natures of ITS data and GPS data should be fully considered when constructing the methodology and procedure.
Task 3: Select a test bed and retrieve the real time ITS data
The research team will make a selection of several test bed in Houston, Texas, where the various travel patterns are normally while the on-line real time ITS data can be easily obtained as well. After selected the test bed, the research team will retrieve the real time ITS data and store them in a well-organized database for the consequent calibration and evaluation analysis. In the mean time, VISSIM coding for the selected typical travel pattern should be conducted.
Task 4: Collect real time GPS data for calibration purposes
GPS will be equipped to the testing vehicles while the global positioning data is collected while traveling along the selected travel patterns. The testing vehicles should following the typical driving behaviors of the traced traffic flows. In order to obtain the actual driving behaviors, sufficient rounds for the same travel patterns should be ensured.
Task 5: Calibrate the parameters using the collected ITS data and GPS data
The retrieved ITS data and collected GPS data will be used in the calibration of parameters following the procedures developed in Task 2. Modifications of the procedures might be possible depending on the identified problems. This task will use the coded typical VISSIM systems as were conducted in Task 3. The result of this task is a set of calibrated parameters for different travel patterns.
Task 6: Compare and evaluate the simulated and measured travel variables and make recommendations
The calibrated parameters will be used in VISSIM simulation for illustrated examples. Travel variables from VISSIM simulation and from the real time measurement will be compared and the parameter calibration effects will be evaluated. If the results are not satisfied, modifications to the calibration methodology and procedures should be made so that better evaluation outcomes will be achieved accordingly.
Task 7: Document research results and findings
This task will document all research results and findings from this project. A comprehensive final report summarizing the calibration processes will be delivered.
Index Terms:
Traffic Simulation; Traffic Models; Calibration; Statistical Analysis; Intelligent Transportation Systems; Global Positioning System; Automatic Data Collection Systems