Publication Detail
Theoretical Development of an Emission-Optimized Trip Assignment
|
UCD-ITS-RR-98-10 Research Report |
Suggested Citation:
Sugawara, Satoshi (1998) Theoretical Development of an Emission-Optimized Trip Assignment. Institute of Transportation Studies, University of California, Davis, Research Report UCD-ITS-RR-98-10
Urban air quality remains a potentially serious health problem and mobile emissions represent one of the major source contributors. One way to reduce mobile source pollutant emissions is to control urban traffic movement using ITS technologies such as ramp metering and route guidance. Currently there aren't many traffic management planning evaluation tools to assess the effectiveness of traffic control alternatives associated with reduction of mobile emissions. One advance toward such a planning tool would be a trip assignment model that estimates emission-optimized traffic flows. Currently, commonly used trip assignment models, such as user-equilibrium models, yield time-dependent traffic flows, i.e., they assume that drivers choose routes based on optimizing travel time, not pollutant emissions. Vehicle emissions are then calculated as byproducts of the time-dependent traffic flows. As a result, the equilibrium conditions for estimated traffic flow by no means minimize vehicle emissions.
This study presents the development of an emission-optimized trip assignment model (named XXEEO) that estimates traffic flows which minimize total network emissions. The modeling process requires defining a mathematical relationship between traffic flows and emissions using the typical travel time-flow relationship developed by the Bureau of Public Road and speed-emission relationship developed by the California Air Resources Board. Similar to the conventional trip assignment models, the emission-optimized model relies on the Frank-Wolfe algorithm but substitutes a global optimization method, Simulated Annealing, for the bisection method typically applied in the Frank-Wolfe algorithm.
The model is tested on two networks. The performance results demonstrate that the XXEEO modeling concept can be successfully applied. The results also suggest substantial differences arise with respect to vehicle emissions between the emission-optimized model and the conventional time-dependent user-equilibrium and system-optimized models. Thus, the modeling concept that is developed is shown to be potentially a very valuable planning evaluation tool for assessing the effects of traffic management measures associated with emission reductions.
This study presents the development of an emission-optimized trip assignment model (named XXEEO) that estimates traffic flows which minimize total network emissions. The modeling process requires defining a mathematical relationship between traffic flows and emissions using the typical travel time-flow relationship developed by the Bureau of Public Road and speed-emission relationship developed by the California Air Resources Board. Similar to the conventional trip assignment models, the emission-optimized model relies on the Frank-Wolfe algorithm but substitutes a global optimization method, Simulated Annealing, for the bisection method typically applied in the Frank-Wolfe algorithm.
The model is tested on two networks. The performance results demonstrate that the XXEEO modeling concept can be successfully applied. The results also suggest substantial differences arise with respect to vehicle emissions between the emission-optimized model and the conventional time-dependent user-equilibrium and system-optimized models. Thus, the modeling concept that is developed is shown to be potentially a very valuable planning evaluation tool for assessing the effects of traffic management measures associated with emission reductions.
Master's Thesis