This report summarizes a demonstration project on the use of Total Stations in incident management. The study was conducted by the Washington State Patrol. Total Stations are intended to reduce the time of incident clearance and improve the accuracy of accident investigation. In Washington State, accidents involving fatalities, alcohol or other drugs, suspected felonies, and other accidents that may lead to court action are investigated. Responding patrol units are required to measure skid marks, road characteristics, debris patterns, intersection configurations, signs, and other relevant details. Traditionally, investigators use the coordinate method to make these measurements. This is accomplished by laying down a base tape through the accident scene. Locations are then measured perpendicular to the tape. Recreating the accident scene back at the office is then often a difficult tedious task that is usually done by hand.

Total stations are a combination of electronic distance meter and a theodolite. Basically, measurements are taken from a central point using an infra-red beam and prism. The investigator places the total station at a site where the entire site is visible. One person then holds a rod with the prism at the various points and positions to where measurements are needed. The station can then measure distance and angles to the prism relative to itself, using the average time it takes the beam to reflect from the prism. Measurements are then stored on a computer. Back at the office recreation is as simple as downloading the data to a drafting program.

One evaluation of the demonstration project consisted of measure the effects of changes in on-scene investigation times between the coordinate method and the total station. Three accidents were studied. Measurements were taken using both methods at each scene and comparisons made for measurements per hour and time to investigate. Site A (first accident) was a fatality where a vehicle left the roadway at a curve and hit a telephone pole. The accident occurred on a two-lane highway at night. The second site (Site B) involved an officer and a vehicle failing to yield right of way at an intersection. The accident occurred at night in a small town and resulted in serious injuries. Lastly, accident site C was a head on collision on a two-lane rural highway at night. The accident involved a county deputy and a vehicle that crossed the centerline and resulted in serious injuries. The following table highlights the results of the comparison between the two measurement procedures at each site.

Accident Site	Measurements/Hour (Coordinate Method)	Time to Investigate (Total Stations)	Time to Investigate (Coordinate Method)
Site A	32.4	47.01 minutes	60 minutes
Site B	30.0	50.49 minutes	75 minutes
Site C	24.0	52.42 minutes	45 minutes
Average	28.8	49.81 minutes	60 minutes

A benefit not captured by the measures used in this study was that by using the coordinate method only the minimum amount was measured, however, the entire scene was measured using the total station. From the table above it appears that total station require 46 percent of the time required to investigate a scene with the coordinate method. Also, investigators can take 70 percent more measurements per hour (See Note 1 below).

The evaluation also compared incident clearance times using data from accidents investigated 20 accidents that occurred in 1989 before the use of total stations. After data used investigations that occurred in the first seven months of 1991 and included 15 accidents, when total stations were used in all investigations. The amount of time from notification to clearance was compared. It was determined that on average accidents investigated using the total stations were cleared 51 minutes earlier then those using the coordinate method.

A benefit to cost ratio for the use of total stations was also determined. The cost equipment was compared to the value of the traveling public of reduced fuel consumption and delay. One accident was used to determine the benefit-to-cost ratio for this study, because of the difficulty in hourly traffic volumes for each accident investigated. The accident occurred during the AM Peak period on May 22, 1991. Because of computer problems exact traffic volume data was unavailable and data from the exact day in 1990 was used (See Note 2) and estimates using a four hour traffic simulation were used. The results showed that 7,000 hours of delay could be saved by the use of the total station which translates to $49,000. In addition, 7,700 gallons of fuel could be saved or $7,700 (@ $1 per gallon).