01557776

Component

https://doi.org/10.3141/2491-06

Microsimulation programs, before being used for traffic analysis in practice, should be calibrated to reproduce the traffic characteristics and to represent the driver population in reality. A driver behavior questionnaire survey conducted in China revealed that significant differences in driving behavior existed in drivers. Calibration should be done by taking these differences in driving style into account. In this study, 30 drivers in Changsha, China, were invited to participate in in-car tests and complete a questionnaire. Factor analysis revealed the interdependencies between observed acceleration or deceleration characteristics and other variables. Four driving types were classified according to the four factor scores. Further analysis of repeated trips validated the consistency of driver membership within a class. A censoring method was applied to calibrate the desired acceleration functions and desired speed distributions in a simulation model for these four driving types. Distinct differences were found in desired speed and maximum desired acceleration or deceleration of the four driving types. It is expected that the categorization of driving types will be useful in model calibration.

01586813

15-3955

English

pp 53–60

2015

Transportation Research Record: Journal of the Transportation Research Board

9780309369275

Print

Figures (5) ; References (21) ; Tables (4)

Behavior; Calibration; Drivers; Driving; Factor analysis; Microsimulation; Speed distribution; Surveys; Traffic models

China

Highways; Operations and Traffic Management; Planning and Forecasting; I72: Traffic and Transport Planning

TRIS, TRB, ATRI

Dec 30 2014 1:17PM

• Analysis of Network Exit Functions for Various Urban Grid Network Configurations
• Capacity Drop: Relationship Between Speed in Congestion and the Queue Discharge Rate
• Comparative Analysis of Traffic State Estimation: Cumulative Counts-Based and Trajectory-Based Methods
• Developing a Large-Scale Hybrid Simulation Model: Lessons Learned
• Empirical Study of Scooter–Vehicle Mixed Traffic Propagation on Urban Arterials
• Multiscale Traffic Flow Model Based on the Mesoscopic Lighthill–Whitham and Richards Models
• On Traffic Relaxation, Anticipation, and Hysteresis
• Real-Time Traffic Network State Prediction for Proactive Traffic Management: Simulation Experiments and Sensitivity Analysis
• Spatiotemporal Clustering of Urban Networks: Real Case Scenario in London
• Trajectory Data and Flow Characteristics of Mixed Traffic
• Use of Microsimulation for Examination of Macroscopic Fundamental Diagram Hysteresis Patterns for Hierarchical Urban Street Networks
• Vehicle Dynamics Model for Estimating Typical Vehicle Accelerations