Contract No. 89-1204
Michigan Department of Transportation
Traffic and Safety Division
425 West Ottawa
Lansing, Michigan 48184-1073
Martin R. Parker & Associates, Inc.
38549 Laurenwood Drive
Wayne, Michigan 48184-1073
This report describes the findings of a study conducted to determine if including factors in addition to the 85th percentile speed could increase the effectiveness of Michigan’s speed zoning procedure as measured by improved safety and increased driver compliance.
The study included an examination of speed zoning methods used in other states, including Michigan; an assessment of using selected quantitative methods on Michigan highways; a before and after accident analysis of speed zones implemented on nonlimited access highways in Michigan; and an assessment of how time and location of the speed survey stations affect the 85th percentile speed.
To improve safety and driver compliance, it is recommended that speed limits be posted within 5 mi/h of the 85th percentile speed.
Appreciation is given to the traffic officials who returned the survey and provided background information on their speed zoning procedure.
The information contained in this report was compiled exclusively for the use of the Michigan Department of Transportation. Recommendations contained herein are based upon the research data obtained and the expertise of the author. The contents do not necessarily reflect the views or policy of the Michigan Department of Transportation.
Establishment of speed zones on Michigan state highways is primarily based on the 85th percentile speed obtained by measuring a sample of free-flow vehicle speeds traveling under favorable traffic and roadway conditions. While every state uses the 85th percentile speed as a primary factor in setting speed limits, some states use quantitative methods which include factors such as roadside development, pedestrian activity, and accident history.
This study was conducted to determine if including other factors in addition to the 85th percentile speed could improve safety and increase driver compliance. In addition, the effects that time of day and location of the speed survey station have on the 85th percentile speed were examined.
Based on a survey of highway officials in 50 states, the District of Columbia, Guam, and Puerto Rico, background information on speed zoning methods was obtained. Very few evaluations have been conducted of any speed zoning method. An assessment of selected quantitative methods used in other states was conducted at Michigan speed zone sites established on the state trunkline (excluding limited access highways) between 1982 and 1986.
To assess the safety impact of the current Michigan speed zoning method, a before and after design with a comparison group and a check for comparability was employed on speed zones established on state highways during the period 1982 through 1986. To determine if any other quantitative method was superior to the Michigan procedure in improving safety and driver compliance, data needed to calculate the recommended speed were collected for each procedure at each Michigan site. An assessment of selected speed zoning methods was made based on safety, compliance, cost-effectiveness, and other criteria.
The effects of time, location, and other factors were examined by collecting speed data at 80 locations on 28 selected sections of Michigan trunkline which were zoned during the period 1982 through 1986. Speed survey stations within each zone were located based on an analysis of accidents and the geometry in each zone. Automated equipment was used to collect speed data for a 24-hour period at each station.
Finally, data were collected at 13 speed zone locations to validate the recommended procedure.
SUMMARY OF FINDINGS
- The 85th percentile speed is the primary factor states use in setting speed limits.
- Engineering judgement is the primary tool used to set speed limits. Frequently, the process is quite subjective which leads to arbitrarily posted limits.
- The available evidence suggests that posting limits in the region of the 85th percentile speed minimizes accident involvements and provides acceptable driver compliance. There is no information that suggests including other factors in setting speed limits would provide additional safety or compliance benefits.
- An analysis of accidents at 68 Michigan sites where speed limits were changed and 86 comparison sites revealed that the current speed zoning method practiced in Michigan reduce total accidents by 2.2 percent. The level of confidence of this estimate is 62 percent. The 95 percent confidence interval for this estimate ranges from an accident reduction of 7 percent to an accident increase of 3 percent. The analysis revealed that this effect was not consistent from site to site. Accidents did not increase when speed limits were raised, and accidents did not decrease when speed limits were lowered.
- The most beneficial safety effect occurred when speed limits were posted within 5 mi/h of the 85th percentile speed. At sites posted near the 85th percentile speed, accidents were reduced by 3.5 percent. The level of confidence of this estimate is 73 percent. At sites where the speed limit was posted more than 5 mi/h below the 85th percentile speed, there was a 0.47 percent increase in accidents; however, this result is not statistically significant.
- Speed limits posted at approximately 31 percent of the Michigan sites were not within 5 mi/h of the 85th percentile speed.
- At a typical Michigan site, a 5-mi/h difference in posted speed has a dramatic effect on driver compliance. If limits are set within 5 mi/h of the 85th percentile speed, at a minimum, 67 percent of the motorists would be in voluntary compliance. When limits are set within 7 mi/h, it is possible that only 40 percent compliance would be achieved.
- An assessment of selected quantitative speed zoning methods used in other states was made based on safety, driver compliance, cost-effectiveness, and other criteria. Based on the assessment, the current Michigan procedure of posting limits within 5 mi/h of the 85th percentile speed was found to be superior to the other speed zoning methods examined.
- Significant differences in hourly 85th percentile speeds were found at the survey stations on Michigan roadways examined in this study. The average difference for all monitoring stations, between the lowest and highest hourly 85th percentile speed, was 5.7 mi/h. The lowest variation in hourly 85th percentile speeds occurred between the hours of 8:00 a.m. and 5:00 p.m. When data are collected between 8:00 a.m. and 5:00 p.m., the hourly variations due to time of day can produce an error of approximately 1.5 mi/h above or below the 24-hour 85th percentile speed.
- The method used by the Michigan Department of Transportation to collect speed data appears to have a significant effect on the 85th percentile speed. Based on selected samples, it appears that the Department’s estimate of the 85th percentile speed is approximately 3 mi/h lower than the speed recorded by automated equipment.
- The current Michigan practice of posting speed limits within 5 mi/h of the 85th percentile speed has a beneficial effect, although small, on reducing total accidents, but has a major beneficial effect on providing improved driver compliance.
- Posting speed limits more than 5 mi/h below the 85th percentile speed does not reduce accidents and ahs an adverse effect on driver compliance.
- The accident analysis revealed that the speed limit changes on Michigan roadways produced a small effect on total accidents, and these effects varied from location to location. Consequently, speed zoning should not be used as the only corrective measure at high-accident locations in lieu of other safety improvements.
- The quantitative speed zoning methods or other factors used by the other states examined in this study would not improve safety and driver compliance if implemented on Michigan roadways.
- The 85th percentile speed varies by hour of the day. Speed samples taken for a short period at a survey site can overestimate or underestimate the 24-hour 84th percentile speed by 1.5 mi/h or more.
- The use of radar to collect speed data in Michigan appears to underestimate the 85th percentile speed by approximately 3 mi/h.
- Field studies conducted at 13 selected Michigan speed zone sites illustrate the validity of setting speed limits within 5 mi/h of the 85th percentile speed.
- The speed zoning procedure recommended in this study is not dramatically different from the speed zoning method currently practiced by the Michigan Department of Transportation and the Michigan State Police.
- The use of automated equipment is strongly recommended to minimize errors associated with time of day effects and current speed data collection methodology.
1. The following speed zoning procedure is recommended for implementation in Michigan.
- Speed limits should be posted within 5 mi/h of the 85th percentile speed.
- An accident analysis should be conducted as a routine part of speed zone investigations. The analysis should identify abnormally high accident characteristics and problem locations. A field review should be conducted to identify possible causes and develop recommendations for improvements. Speed zoning, per se, should not be used as a countermeasure to address abnormally high-accident situations.
- To minimize time of day effects and data collection errors, 85th percentile speeds should be determined by using automated equipment to collect data for a 24-hour period.
- The location of the survey stations should be based on the geometry in each zone and roadside development. Stations should not be placed within 500 feet of isolated major intersections or horizontal curves.
- The data should be analyzed in accordance with the guidelines listed below to determine the appropriate speed limit.
2. The following guidelines should be used for setting speed limits.
- The posted speed limit should be set within 5 mi/h of the 85th percentile speed.
- The beginning and ending of each speed zone should be at a point obvious to the motorist such as a change in geometry, roadside development, etc. Jurisdictional boundaries such as city or township lines may be an inappropriate location for a speed zone change.
- The use of short (less than 0.20 mile) speed zones and transition zones is discouraged. The majority of reasonable motorists adjust their speed based on environmental and traffic conditions and not on artificially low or high posted speed limits.
- Within each zone it is desirable that features such as design, roadside development, etc. be consistent, as homogeneous sections tend to encourage similar operating speeds. It is not always practical to subdivide a roadway section into homogeneous zones because this could result in a number of short sections with various speed limits.
- The speed limit on the entire zone should not be based on one special condition such as an isolated horizontal curve or intersection. When appropriate, advisory speeds should be used at these locations.
- Combining individual 85th percentile speeds in a zone to arrive at an average or composite figure is discouraged. It is also not necessary to collect speed data for both directions of travel at the same survey station. A more representative sample can be obtained by spreading the stations throughout the zone.
- The 85th percentile speed at each individual survey station should be compared to speeds at other stations in the zone. If the individual 85th percentile speeds vary by more than 5 mi/h, consideration should be given to providing separate zones if this does not result in short section lengths.
- Michigan law and congressional directives establish a 55 mi/h maximum speed limit on nonlimited access highways. On some rural highways, 85th percentile speeds exceed 55 mi/h. This creates a problem when using the 85th percentile speed to set speed limits in areas that transition from rural to urban conditions. Until realistic zoning is used on all highways, engineering judgement must be employed to set speed limits in transition areas.
3. To improve public understanding of the safety impacts and other benefits of using the 85th percentile speed to set speed limits, a public informational brochure should be developed for distribution.