The Executive Summary

The executive summary is a quick overview of the project and it's results.

Index

Objectives and Methodology

 Incidents (accidents, vehicle breakdowns, spilled loads or other random events) that reduce the roadway capacity account for a significant portion of the congestion delay on urban freeways, and to the increase of accidents, fuel consumption and air pollutant emissions. Freeway service patrol (FSP) is an incident management program implemented in a number of metropolitan areas to facilitate the quick removal of incidents. FSP tow truck drivers continually patrol freeway segments during commute hours and provide assistance to disabled vehicles. The California FSP program is funded by the legislature as well as federal and local monies, and is jointly administered by the California Department of Transportation (Caltrans), California Highway Patrol (CHP) and local transportation planning agencies. This report presents the findings of a comprehensive evaluation of the FSP program on a specific freeway section in the San Francisco Bay Area.

The study methodology consisted of determining the savings in incident delay and other performance measures "before" and "after" the implementation of FSP based on field data and improved analysis procedures. Data were collected during the peak periods on a 9 mile section of the I-880 freeway in the City of Hayward, Alameda County. The "before" study was conducted for 24 weekdays in the Spring of 1993, and the "after" study took place in the Fall for a total of 22 weekdays. Incident information (type, location, severity, type of assist and duration) was obtained through observations of probe vehicle drivers traveling at an average of 7 minute headways. Speeds, flows and occupancies at 1-sec intervals were collected from loop detectors spaced approximately 1/3 mile apart on the freeway mainline and on all the ramps. Vehicle trajectories and travel times were obtained from the specially instrumented probe vehicles. Supplementary information was collected from the CHP computer aided dispatch system, FSP records, and rotational and membership tow truck companies logs.

Software was developed to process the field data and create a computerized database. The I-880 database consists of 276 hours of field data that are uniquely linked to provide a complete representation of the freeway operating conditions at the test site. This is the largest and most comprehensive database on freeway operations to date and is readily available for other studies. The estimation of the incident specific delay was based on the difference in average travel speeds under normal and incident conditions using data from loop detectors and instrumented vehicles.

Findings

 A high number of incidents was observed at the study section. The average frequency was 8 incidents/hour in the peak periods, about 100 incidents per million vehicle miles of travel. The proportion of accidents was 10 percent of all the incidents. Only 4 percent of all the incidents were blocking travel lanes, most of them accidents. In addition, there was a significant number of CHP ticketing incidents, mostly citations for violations of the HOV lane usage in the study section. Time of day, day of the week, presence of shoulders and weather conditions accounted for most of the variability in incident occurrence.

The proportion of tow truck assisted incidents increased from 9 percent "before" to 24 percent "after" the implementation of the FSP service. About 80 percent of the assists were provided by FSP, mostly vehicle breakdowns with mechanical or electrical problems. About 30 percent of those breakdowns had to be towed. The response times of the FSP assisted breakdowns were reduced by 57 percent, and the response times of all assisted incidents were cut by 35 percent in the "after" study. The differences in the average clearance times "before" and "after" were not statistically significant for both accidents and breakdowns. The incident durations "before" and "after" were similar for all the incident categories. Significant reductions in durations were found only for the assisted incidents, because of the faster response times of the FSP service.

The estimation of the FSP benefits was based on the savings in the incident delay and fuel consumption due to the FSP service excluding accidents, incidents of short duration and abandoned vehicles. The estimated benefit/cost ratio of 3.4:1 shows that the program has been cost-effective at the specific test site. In addition, reductions in air pollutant emissions include 77.2 tons of carbon monoxide, 19.1 tons of oxides of nitrogen and 7.6 tons of hydrocarbons. Additional benefits that were not included in the calculation of the benefit/cost ratio include a) time and cost savings to the motorists assisted by FSP, b) improved incident detection and reporting by the FSP drivers, and c) reduction in the number of, and the time spent on incidents by CHP officers. The survey of motorists assisted by the FSP showed overwhelming approval with 93 percent rating the service as excellent. Motorists also wrote complementary comments, left voice mail and sent letters to show their support of the program. Furthermore, FSP vehicles provide a sense of security on the freeway and faster clearance of incidents reduces the chance of secondary accidents.

Recommendations

 The estimated benefit/cost ratio applies to sites with traffic and incident characteristics similar to the ones in the study area, and may not be used as a guideline for the effectiveness of the FSP elsewhere. The effectiveness of FSP would be higher on locations with similar incident patterns but higher traffic volumes, mixed lanes and narrow or no shoulders. However, the benefits would be limited on sites with few major incidents as opposed to sections with high frequency of vehicle disablements. Additional evaluation studies should be performed to quantitatively determine the range in the FSP benefits. This study provides a framework for the accurate assessment of the FSP impacts.

There is a need for comprehensive databases such the one developed in this study for other freeway sites under a range of operating conditions. Most of the studies on incident characteristics and impacts were conducted in the 60's and 70's, when urban freeways were less congested, and driver behavior and vehicle characteristics were quite different. Further work is also needed to develop improved procedures for estimating incident delays from various data sources.

A number of issues related with the operation of FSP need further investigation including:

Steps could be also taken through regulation means to reduce the number of incidents on urban freeways. Fines and citations for running out of gas and mandatory inspections for serviceability of the vehicle's electrical and mechanical systems could reduce the number of breakdowns.

FSP Project / 30 September 1994