Local traffic measures such as controlling traffic demand, banning heavy duty vehicles or restricting speeds activated only during periods of peak pollution can contibute to significant reductions in air quality measures.

A simulation of an intersection in Sweden shows that measures applied temporarily can help locales meet air quality standards.

Date Posted
11/14/2011
Identifier
2011-B00754
TwitterLinkedInFacebook

Modeling Reduced Traffic Emissions in Urban Areas: The Impact of Demand Control, Banning Heavy Duty Vehicles, Speed Restriction and Adaptive Cruise Control

Summary Information

Traffic emissions are a main source of air pollution in urban areas, and with the proliferation of vehicles, often exceed air quality standards. A general finding is that emissions increase with increases in traffic volume, the number of heavy vehicles, traffic dynamics (i.e., acceleration and deceleration), and when traffic moves at either very high or very low speeds. For example, many European cities breach European Union air quality standards at specific locations (hot-spots) and times (pollution peaks). Reducing emissions can be achieved by activating local measures during periods of poor air quality. Measures including traffic demand control, banning Heavy Duty Vehicles (HDVs), speed restriction, and Adaptive Cruise Control (ACC) could reduce emissions at hot-spots and/or during pollution peaks. The intention of implementing these measures is to reduce traffic volume, reduce the number of heavy vehicles, lower the number of acceleration and decelerations, and obtain optimal speed.

This report used a traffic model and an emissions model to examine the impact of these example traffic measures applied at a single intersection.

Traffic measures aiming to reduce traffic emissions should focus on the following four factors:

1) reducing traffic volume

2) reducing the number of heavy vehicles

3) limiting traffic dynamics (acceleration and deceleration)

4) obtaining optimal speed.

These factors can be controlled by roadside and in-vehicle treatments, such as traffic re-routing, banning heavy duty vehicles, implementing speed restrictions, and using Adaptive Cruise Control (for optimal speed). These measures can be activated at locations with unacceptably high emissions or when there are periods of peak pollution.

METHODOLOGY

A simulation study examined the impact of these measures on emissions and the extent to which they improved upon signal timing. The study used VISSIM (a microscopic model of individual vehicle movements) and EnViVer (a model for predicting emissions) to evaluate the effects of the measures as applied to a single, four-way intersection at Bentinckplein in the city of Rotterdam, The Netherlands. The intersection carries up to 45,000 vehicles per day, and consists of left and right turn lanes in the approaching roads, public transportation lines for buses and trams, and pedestrian and cyclist crossing facilities. Traffic data were taken including traffic counts and average vehicle speed per 15 minutes was collected for two classes of vehicles (shorter and longer than 3.5 meters) and at two directions for a little more than a month, from September 12 to October 16, 2008.

FINDINGS

The results included the impact of each of the measures on total emissions and emissions per vehicle types. Highlights include the following.

  • Traffic demand control. Reducing traffic demand by 20 percent resulted in traffic volume reductions of about 14 percent, and delay and number of stops per vehicle reductions of 4.8 percent and 6.4 percent respectively. The total emissions reduction (i.e., of all pollutants) was about 23 percent. The same reduction percentage was obtained per pollutant for HDVs was more than 20 percent for all pollutants because HDVs produce seven-to-eight times more emissions per kilometer compared to light duty vehicles (LDVs).
  • Banning HDVs. Eliminating HDVs from the model reduced CO2 by 25.8 percent, NOx by 50 percent, and PM10 by 30.9 percent for emissions. The banning of HDVs, however, increased the number of LDVs in the traffic network by 3 percent, resulting in a 10 percent increase of LDV emissions of CO2, NOx, and PM10. The higher number of LDVs in the network also increased vehicle dynamics (more maximum acceleration and speed variability), which contributed to the LDV emissions. More LDVs resulted in an increase in average speed by 1.6 percent, reduced delay by 0.1 percent, and reduced the number of stops per vehicle by 0.2 percent.
  • Speed Restriction. Applying a speed limit of 30 km/hour reduced CO2 by 7.8 percent but increased PM10 by 31.4 percent and NOx by 1.8 percent. The reduction in CO2 was to a lower number of vehicles entering the network in the speed restriction scenario during the warning time. For LDVs, speed restrictions of 30 km/hour reduced CO2 and NOx were reduced by 10.7 percent and 3.2 percent, respectively. For HDVs, CO2 was reduced by 5.6 percent but NOx was increased by 1.9 percent. PM10 increased for LDVs (25.4 percent) and HDVs (38.2 percent).
  • Adaptive Cruise Control. A penetration rate of 40 percent ACC vehicles (only for LDVs) led to a reduction in CO2 and NOx by 3 percent but an increase in PM10 of 2.9 percent. The ACC lowered vehicle dynamics, as shown by velocity-acceleration distribution plots from EnViVer, which resulted in reduced CO2 and NOx emissions. In contrast, PM10 does not vary with vehicle dynamics, and PM10 emissions actually increased in the ACC scenario due to the increase of the average speed by 4.3 percent.
The conclusions of this research included the following.
  • The best measures for reducing all pollutants are those that reduce traffic demand, either total traffic volume or the number of HDVs.
  • HDVs significantly impact NOx and PM10 emissions.
  • The effects of speed restrictions on emissions vary by vehicle type. The overall effect was that a speed restriction of 30 km/hour reduced CO2 emissions by 7 percent but increased NOx and PM10 emissions.
  • ACC lowered vehicle dynamics, but the reductions in emissions was not significant. This result may be due to the heavy traffic volumes at the test site and the complexity of the intersection.
Deployment Locations