scholarly journals Uncertainty Quantification of the Traffic Assignment Model

2020 ◽  
Author(s):  
Mundher Seger ◽  
Lajos Kisgyörgy
Keyword(s):  
Uncertainty Quantification ◽  
Traffic Flow ◽  
Travel Demand ◽  
Traffic Assignment ◽  
Posterior Distributions ◽  
Demand Model ◽  
Assignment Model ◽  
Wide Range ◽  
Least Squares Adjustment ◽  
Travel Demand Model

Forecasting of traffic flow in the traffic assignment model suffered to a wide range of uncertainties arising from different sources and exacerbating through sequential-stages of the travel demand model. Uncertainty quantification can provide insights into the level of confidence on the traffic assignment model outputs, and also identify the uncertainties of the input Origin-Destination matrix for enhancing the forecasting robustness of the travel demand model. In this paper, a systematic framework is proposed to quantify the uncertainties that lie in the Origin-Destination input matrix. Hence, this study mainly focuses on predicting the posterior distributions of uncertainty Origin-Destination pairs and correcting the biases of Origin-Destination pairs by using the inverse uncertainty quantification formulated through Least Squares Adjustment method. The posterior distributions are further used in the forward uncertainty quantification to quantify the forecast uncertainty of the traffic flow on a transport network. The results show the effectiveness of implementing the inverse uncertainty quantification framework in the traffic assignment model. And demonstrate the necessity of including uncertainty quantification of the input Origin-Destination matrix in future work of travel demand modelling.

scholarly journals Predicting and Visualizing the Uncertainty Propagations in Traffic Assignments Model Using Monte Carlo Simulation Method

2018 ◽  
Vol 2018 ◽  
pp. 1-11
Author(s):  
Mundher Seger ◽  
Lajos Kisgyörgy
Keyword(s):  
Monte Carlo Simulation ◽  
Monte Carlo ◽  
Traffic Flow ◽  
Travel Demand ◽  
Simulation Method ◽  
Transport Network ◽  
Monte Carlo Simulation Method ◽  
Traffic Flows ◽  
Demand Model ◽  
Travel Demand Model

Uncertainty can be found at all stages of travel demand model, where the error is passing from one stage to another and propagating over the whole model. Therefore, studying the uncertainty in the last stage is more important because it represents the result of uncertainty in the travel demand model. The objective of this paper is to assist transport modellers in perceiving uncertainty in traffic assignment in the transport network, by building a new methodology to predict the traffic flow and compare predicted values to the real values or values calculated in analytical methods. This methodology was built using Monte Carlo simulation method to quantify uncertainty in traffic flows on a transport network. The values of OD matrix were considered as stochastic variables following a specific probability distribution. And, the results of the simulation process represent the predicted traffic flows in each link on the transport network. Consequently, these predicted results are classified into four cases according to variability and bias. Finally, the results are drawn into figures to visualize the uncertainty in traffic assignments. This methodology was applied to a case study using different scenarios. These scenarios are varying according to inputs parameters used in MC simulation. The simulation results for the scenarios gave different bias for each link separately according to the physical feature of the transport network and original OD matrix, but in general, there is a direct relationship between the input parameter of standard deviation with the bias and variability of the predicted traffic flow for all scenarios.

Accuracy and Performance of Improved Speed-Flow Curves

1998 ◽  
Vol 1646 (1) ◽  
pp. 9-17 ◽  
Author(s):  
Richard G. Dowling ◽  
Rupinder Singh ◽  
Willis Wei-Kuo Cheng
Keyword(s):  
Travel Demand ◽  
Traffic Assignment ◽  
Demand Model ◽  
Power Functions ◽  
Flow Curves ◽  
Speed Flow ◽  
Demand Models ◽  
Higher Power ◽  
The Impact ◽  
Travel Demand Model

Skabardonis and Dowling recommended updated Bureau of Public Road speed-flow curves for freeways and signalized arterials to improve the accuracy of speed estimates used in transportation demand models. These updated curves generally involved the use of higher power functions that show relatively little sensitivity to volume changes until demand exceeds capacity, when the predicted speed drops abruptly to a very low value. Skabardonis and Dowling demonstrated that the curves provide improved estimates of vehicle speeds under both uncongested and queueing conditions; however, they did not investigate the impact of these curves on the performance of travel demand models. Practitioners have been concerned about the impacts of such abrupt speed-flow curves on the performance of their travel demand models. Spiess has stated that higher power functions are more difficult computationally for computers to evaluate and that more abrupt speed-flow curves adversely affect the rate of convergence to equilibrium solutions in the traffic assignment process. In this paper the impact of the Skabardonis and Dowling updated speed-flow curves on the performance of selected travel demand models is investigated. The updated speed-flow curves were found to significantly increase travel demand model run times. However, it is demonstrated that an alternative speed-flow equation developed by Akçelik has similar or better accuracy and provides much superior convergence properties during the traffic assignment process. The Akçelik curve significantly reduced travel demand model run times.

Parsimonious Modeling and Uncertainty Quantification for Transportation Systems Planning Applied to California High-Speed Rail

2012 ◽  
Vol 2266 (1) ◽  
pp. 49-59 ◽  
Author(s):  
Quentin Noreiga ◽  
Mark McDonald
Keyword(s):  
Uncertainty Quantification ◽  
High Speed ◽  
Travel Demand ◽  
Mode Choice ◽  
Demand Model ◽  
Parameter Uncertainties ◽  
Main Mode ◽  
High Speed Rail ◽  
The Impact ◽  
Travel Demand Model

This paper presents a parsimonious travel demand model (PTDM) derived from a proprietary parent travel demand model developed by Cambridge Systematics (CS) for the California high-speed rail system. The purpose of the PTDM is to reduce computational expense for model simulations, optimization and sensitivity analyses, and other repetitive analyses. The PTDM is used to quantify the significance of parameter uncertainties with the use of mean value first-order second moment methods for uncertainty quantification and sensitivity analysis. The PTDM changes the model resolution of the parent travel demand model from a traffic analysis zone to a county-level analysis. The three-step model contains trip frequency, destination choice, and main mode choice models and is calibrated to match the results of the CS model. The main mode choice model predicts primary mode choice results for car, commercial air, conventional rail, and high-speed rail. The PTDM uses data and models similar to parent models to show how uncertainty in travel demand model predictions can be quantified. This paper does not attempt to assess the reliability of parent model forecasts, and the results should not be used to evaluate uncertainty in the California High-Speed Rail Authority's rider ship and revenue forecasts. However, the uncertainty quantification methodology presented here, when applied to the CS model, can be used to quantify the impact of parameter uncertainty on the forecast results.

scholarly journals An Integrated Pipeline Architecture for Modeling Urban Land Use, Travel Demand, and Traffic Assignment

2018 ◽  
Author(s):  
Paul Waddell ◽  
Geoff Boeing ◽  
Max Gardner ◽  
Emily Porter
Keyword(s):  
Land Use ◽  
Travel Demand ◽  
Traffic Assignment ◽  
Urban Land ◽  
User Equilibrium ◽  
Urban Land Use ◽  
Analysis Tool ◽  
Demand Model ◽  
Pipeline Architecture ◽  
Assignment Model

Integrating land use, travel demand, and traffic models represents a gold standard for regional planning, but is rarely achieved in a meaningful way, especially at the scale of disaggregate data. In this report, we present a new pipeline architecture for integrated modeling of urban land use, travel demand, and traffic assignment. Our land use model, UrbanSim, is an open-source microsimulation platform used by metropolitan planning organizations worldwide for modeling the growth and development of cities over long (~30 year) time horizons. UrbanSim is particularly powerful as a scenario analysis tool, enabling planners to compare and contrast the impacts of different policy decisions on long term land use forecasts in a statistically rigorous way. Our travel demand model, ActivitySim, is an agent-based modeling platform that produces synthetic origin--destination travel demand data. Finally, we use a static user equilibrium traffic assignment model based on the Frank-Wolfe algorithm to assign vehicles to specific network paths to make trips between origins and destinations. This traffic assignment model runs in a high-performance computing environment. The resulting congested travel time data can then be fed back into UrbanSim and ActivitySim for the next model run. This technical report introduces this research area, describes this project's achievements so far in developing this integrated pipeline, and presents an upcoming research agenda.

Model Calibration and Forecasts of Air Travel Demand with Categorized Household Socioeconomic Attributes

2020 ◽  
Vol 2674 (6) ◽  
pp. 363-371
Author(s):  
Jungin Kim ◽  
Ikki Kim ◽  
Jaeyeob Shim ◽  
Hansol Yoo ◽  
Sangjun Park
Keyword(s):  
Travel Behavior ◽  
Household Income ◽  
Travel Demand ◽  
Behavioral Model ◽  
Income Group ◽  
Air Travel ◽  
Jeju Island ◽  
Demand Model ◽  
Air Travel Demand ◽  
Travel Demand Model

The objectives of this study were to (1) construct an air demand model based on household data and (2) forecast future air demand to explain the relationship between air demand and individual travel behavior. To this end, domestic passenger air travel demand at Jeju Island in South Korea was examined. A multiple regression model with numerous explanatory variables was established by examining categorized household socioeconomic data that affected air demand. The air travel demand model was calibrated for 2009–2015 based on the annual average number of visits to Jeju Island by households in certain income groups. The explanatory variable was set using a dummy variable for each household income group and the proportion of airfare to GDP per capita. Higher household income meant more frequent visits to Jeju Island, which was well-represented in the model. However, the value of the coefficient for the highest income was lower than the value for the second-highest income group. This suggested that the highest income group preferred overseas travel destinations to domestic ones. The future air demand for Jeju airport was predicted as 26,587,407 passengers in 2026, with a subsequent gradual increase to approximately 33,000,000 passengers by 2045 in this study. This study proposed an air travel demand model incorporating household socioeconomic attributes to reflect individual travel behavior, which contrasts with previous studies that used aggregate data. By constructing an air travel model that incorporated socioeconomic factors as a behavioral model, more accurate and consistent projections could be obtained.

A Complex Network Methodology for Travel Demand Model Evaluation and Validation

2018 ◽  
Vol 18 (4) ◽  
pp. 1051-1073 ◽  
Author(s):  
Meead Saberi ◽  
Taha H. Rashidi ◽  
Milad Ghasri ◽  
Kenneth Ewe
Keyword(s):  
Complex Network ◽  
Model Evaluation ◽  
Travel Demand ◽  
Demand Model ◽  
Travel Demand Model

scholarly journals Creating an integrated agent-based travel demand model by combining mobiTopp and MATSim

2019 ◽  
Vol 151 ◽  
pp. 776-781 ◽  
Author(s):  
Lars Briem ◽  
Nicolai Mallig ◽  
Peter Vortisch
Keyword(s):  
Travel Demand ◽  
Demand Model ◽  
Agent Based ◽  
Travel Demand Model
Sign in / Sign up

Export Citation Format

Share Document