Development of a Posted Speed Limit Setting Procedure and Tool (2021)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 137 APPENDIX G. WORKSHOP SLIDES This appendix includes the slides and presenter notes used in one of the workshops presented during the project. 1. This presentation will give you an overview of a new tool that is being developed in NCHRP 17-76 for setting speed limits. 2. The workshop will cover the following: - Existing practices with regards to posting speed limits. - The activities within the NCHRP 17-76 project. - Background on the development of the user guide and the associated Speed Limit Setting Tool.

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 138 - A brief demonstration of the SLS-Tool. - Opportunity for discussion and feedback on the user guide and the SLS-Tool. 3. Several factors are considered when selecting a speed limit for a speed zone. 4. The type of roadway influences the selection. For example, appropriate speed limits vary for freeways as compared to residential streets.

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 139 5. Human factors are another consideration. When using 85th percentile speed as a starting point, you are considering the driver’s (human’s) interpretation of what is appropriate for the facility. 6. Engineering judgment is used when selecting the roadway factor or the safety factors that will influence your decision along with how much those factor will affect your recommendation.

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 140 7. An agency’s policy can also be a major factor in the setting of speed limits. Some locations have laws that restrict the posted speed limit to not exceed a certain value for a facility. The personal experience from some transportation engineers is that politics can influence decisions. Engineers may be under considerable pressure to propose speed limits below the 85th percentile, while a strict, objective, engineering approach might suggest a higher speed limit. While not a desirable situation, this can be the political reality of a job. 8. The user guide developed in NCHRP 17-76 focuses on posted speed limits within speed zones. Other speed limits illustrated on this slide such as school zones or advisory speed limits were outside the scope of the 17-76 project.

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 141 9. As part of the NCHRP 17-76 project, the research team reviewed state websites to identify materials that document how to set a posted speed limit. We found 31 documents on the web. Of those documents, all stated that 85th percentile speed is to be used, and most stated that crash history should be considered. Several various factors are also to be considered. For example, about half of the states reviewed noted that roadside development should be considered. The bottom of the slide lists other factors being considered by at least three states, such as alignment or cross section. 10. While statutory speed limits form the basis for many posted speeds limits, non-statutory speed limits are established by engineering studies. The Manual on Uniform Traffic Control Devices (MUTCD) provide some uniformity in setting of non-statutory speed limits. A guiding

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 142 principle of the MUTCD is that through uniformity, consistent driver expectations can improve safety. Within Section 2B.13 of the MUTCD are several paragraphs related to the selection of the posted speed limit value including Paragraph 1 (standard), Paragraph 12 (guidance), and Paragraph 16 (option). Other paragraphs within Section 2B.13 focus on statutory speed limits, need for engineering study, requiring limits to be multiples of 5 mph, placement of signs, use of warning signs with speed limit signs, where to conduct speed studies, special speed limits, changeable message signs, and school zones. Relative to setting speed limits, the key paragraphs in Section 2B.13 are: “01 Speed zones (other than statutory speed limits) shall only be established on the basis of an engineering study that has been performed in accordance with traffic engineering practices. The engineering study shall include an analysis of the current speed distribution of free-flowing vehicles.” “12 When a speed limit within a speed zone is posted, it should be within 5 mph of the 85th percentile speed of free-flowing traffic.” “16 Other factors that may be considered when establishing or reevaluating speed limits are the following: • Road characteristics, shoulder condition, grade, alignment, and sight distance, • The pace, • Roadside development and environment, • Parking practices and pedestrian activity, and • Reported crash experience for at least a 12-month period.” 11. Recently, the speed limit debate has increased with two publications. In March 2017, the National Association of City Transportation Officials (NACTO) released a policy statement. One of the action items in that statement would “permit local control of city speed limits.” They recommend “state rules or laws that set speed limits at the 85th percentile speed should be repealed.”

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 143 In July 2017, the National Transportation Safety Board (NTSB) published a report on speeding (Reducing Speeding-Related Crashes Involving Passenger Vehicles). That document included several recommendations for reducing speed-related crashes including two recommendations directed to the Federal Highway Administration for changes to the MUTCD (p. 57): • Revise Section 2B.13 of the Manual on Uniform Traffic Control Devices so that the factors currently listed as optional for all engineering studies are required, require that an expert system such as USLIMITS2 be used as a validation tool, and remove the guidance that speed limits in speed zones should be within 5 mph of the 85th percentile speed (H-17-27). • Revise Section 2B.13 of the Manual on Uniform Traffic Control Devices to, at a minimum, incorporate the safe system approach for urban roads to strengthen protection for vulnerable road users (H-17-28). Documents referenced: • NACTO (2017). “Creating Safe, Sustainable, Multi-modal Urban Transportation.” National Association of City Transportation Officials. www.nacto.org. • NTSB (2017). Reducing Speeding-Related Crashes Involving Passenger Vehicles. NTSB/SS-17/01 PB2017-102341. National Transportation Safety Board. 12. Other new approaches to setting speed limits have been implemented recently, including citywide or default speed limits and implementing slow zones. These approaches are outside of the NCHRP 17-76 scope but are being providing as an example of how posting speed limits is evolving.

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 144 13. To address the NTSB recommendations, the National Committee on Uniform Traffic Control Devices (NCUTCD), Regulatory and Warning Signs Technical Committee established a task force to explore current and potential approaches to the setting of posted speed limits. The task force started the process by creating a survey to gather an understanding of how practitioners actually applied current practices in setting speed limits and queried their opinions on the topic. The findings from the surveys were presented to several groups, and discussions were held on the findings and the potential direction that should be taken. The task force’s suggested changes to the MUTCD are shown on this slide and include: • Changing the MUTCD to reinforce the stated understanding that other factors have a role in setting speed limits (in addition to 85th percentile). Refine the factors in Paragraph 16 and group the paragraphs that speak to setting of speed limits. • Retaining reference to 85th percentile as a factor that should be considered, particularly for freeways, expressways, and rural areas. • Keeping the MUTCD broad. While it could be reorganized to better present the material for setting speed limits (by minor reorganization), it should not be expanded into greater detail. The detail should be provided from national research and state/local procedures and promoted by FHWA. • Not including reference of specific processes, such as USLIMITS2, but promoting this level of detail in state/local procedures and investigating more deeply the reasons why after more than 20 years a small fraction of practitioners use this expert system.

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 145 14. One of the questions from the NCUTCD task force survey was “How would you set speed limits if given the choice?” The provided responses included rounding to nearest 5 mph of 85th, or rounding up or down, etc. Half of the survey participants selected “other” and typed a response. To the right of the graph on this slide is a word cloud where the size of the word reflects the use of that word in the responses. As you can see, the word context was used more than any other word. The research team incorporated context into the SLS-Tool. 15. The objectives for the NCHRP Project 17-76 were to identify and describe factors that influence operating speed, and to create a user guide and a tool that will provide guidance in making informed decisions related to establishing speed limits on roadways.

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 146 16. When developing the user guide and the SLS-Tool, the research team established guiding principles. We wanted the procedure to be easy to explain—at least relatively easy to explain because the setting of speed limit is a complex and involved process. The use of decision rules will provide the ability to have consistent results. How those decision rules are established must be defendable, and the research report will document those sources. We wanted the tool to be transparent so the user can trace how a suggested speed limit was reached. We also wanted the tool to be flexible so future knowledge can update the decision rules and the procedure. 17. Another guiding principle is that we wanted the SLS-Tool to be able to be used for all roadway types and contexts, rather than forcing a user to obtain different tools. Given the number of roadway types and context, the research team recognized that the decision rules may apply to a groups of roadway types and contexts.

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 147 18. Additional guiding principles for the spreadsheet tool are listed on this slide. The SLS- Tool will have most of the data on one screen and will use colors to show what data needs to be provided versus warnings or the suggested speed limit. Per requirement of the contract, the tool was to be spreadsheet based. 19. At the time of this workshop, we anticipate that the user guide will be published within the NCHRP series and the research report will be a web-only document. The user guide will be accompanied by the SLS-Tool.

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 148 20. Overall, the development of the suggested speed limit follows the four steps shown on this slide. You start with identifying the roadway segment’s context and type. The user then enters the existing operating speed to consider the speeds drivers are selecting for the segment. The crash experience or potential is the next step within the procedure, resulting in a final suggested speed limit for the segment. The next slides will go into details for each step. 21. The roadway context uses the Expanded Functional Classification System developed as part of NCHPR Report 855. It includes five roadway contexts as shown on this slide.

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 149 22. The user identifies the roadway type, again using the definitions provided in NCHRP Research Report 855. A brief description is provided on this slide. 23. The combination of five roadway types and five roadway contexts results in a matrix with 25 cells. When reviewing these cells, it is reasonable that the decisions rules used for some of the cells would be the same. This slide shows the Speed Limit Setting Groups identified by the research team. The groups include Limited Access for freeways and Undeveloped for all rural roadway contexts except freeways. How to groups the remaining cells into categories led to several discussions within the research team, the panel, and other subject matter experts. This slide shows the final recommendation where two additional groups, Developed and Full Access, are established.

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 150 24. The driver’s operating speed, as measured by the 85th percentile speed and the 50th percentile speed, is used as starting points for identifying the suggested speed limit. How to round the appropriate speed is based on decisions rules that consider roadway characteristics and crashes. The procedure also includes a maximum speed limit check, where appropriate. Similar to USLIMITS2, the research suggested that most roadway context/type combinations consider three percentile/rounding groups: closest 85th, rounded down 85th, and closest 50th. The closest 85th would be used when roadway conditions and safety are acceptable. When roadway conditions or safety is not favorable to all users or a concern, the other extreme of closest 50th is suggested. The rounded down 85th would be used for between conditions. Full Access as a roadway context/type group only uses the 50th percentile speed distribution. Full Access would have speed limits of 35 mph or less. 25. Speed distribution is obtained by measuring the speed of free-flow vehicles.

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 151 26. Safety is considered through the provision of crashes. It is also considered in the decision rules when a geometric variable has been shown through other research to be associated with more crashes. 27. The research team has been asked why the SLS-Tool uses speed distribution. The research team included speed distribution to retain a connection with drivers. The speed distribution is adjusted to consider roadway characteristics and crashes via the decision rules.

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 152 28. The graph shows the relative rate of involvement of being in a crash when compared to the average speed. It is based on six studies (1968-1990). The lowest point on the curve is where the relative involvement rate is lowest: some studies found that point to be about 7 mph higher than the average speed (near the 85th percentile), and others found it to be closest to the average speed of the traffic stream. Involvement rate is relatively low at 85th percentile speed, which has been the key variable for many methods of setting speed limits. Many guidelines suggest starting with 85th and rounding down depending upon conditions. The gray u-shape curve reflects work by Hauer that considers the rate at which motorists experience overtakings. You minimize the number of motorists who catch up or overtake you and the number of vehicles that you catch up with or overtake by traveling at the median speed. There appears to be a 10-15 mph band of speeds where accident risk is minimal. To effectively manage crash risk, the speed limit should be set within the low-risk region. It should be noted that the speed of the crash-involved vehicles is rarely known very accurately. The study by West and Dunn is probably the most reliable; turning vehicles were excluded; they relied on multidisciplinary teams to investigate the crashes and had speed sensors on some of the roads. There are numerous factors that can confound the results. For example, younger drivers, because of their inexperience, tend to have higher crash rates but also tend to drive faster than average. Older drivers, who as a group also have higher than average crash rates, tend to driver slower. Sources: J. A. Cirillo (1968). “Interstate System Accident Research Study II, Interim Report II.” Public Roads, Vol. 35, No. 3. D. Solomon (1964). Accidents on Main Rural Highways Related to Speed, Driver, and Vehicle. Federal Highway Administration, July 1964 (reprinted 1974). L. B. West, Jr., and J. W. Dunn (1971). “Accidents, Speed Deviation and Speed Limits.” Traffic Engineering, Vol. 41, No. 10,

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 153 D. L. Harkey, H. D. Robertson, and S. E. Davis (1990). “Assessment of Current Speed Zoning Criteria.” Transportation Research Record, No. 1281. E. Hauer (July 1971). “Accidents, Overtaking and Speed Control.” Accident Analysis and Prevention, Vol. 3, No. 1. J. Stuster, Z. Coffman, and D. Warren (1998). Synthesis of Safety Research Related to Speed and Speed Management. FHWA-RD-98-154. https://www.fhwa.dot.gov/publications/research/safety/98154/speed.cfm. 29. This graph shows the relationship between crash rate and the difference between posted speed limit and average speed for 649 sites in Austin, Texas. The dataset represents urban and suburban streets. Speed data were obtained from traffic counters either installed by the research team or installed by the city when responding to requests for traffic calming or changes in posted speed limits or other needs. A trend line was added to aid in seeing the relationship between crash rate and the difference between posted speed limit and average speed. The minimum crash rate is occurring near the point where the majority of drivers are driving at the posted speed limit. While other research has suggested setting the posted speed limit at the 85th percentile, this study indicates that setting urban/suburban streets closer to the 50th percentile will result in minimizing crashes. The study is also supporting the goal of minimizing the variation in speeds. We prefer drivers to be driving a uniform speed. The trend line demonstrates that when speeds are notably difference from the average speed—either higher or lower—more crashes can be expected. There are several cautions with this finding. It represents data for one city. Additional studies should be conducted to validate this finding for urban/suburban streets. The speed limit range included residential streets and higher-speed arterials. It is preferred to have unique results for each roadway type; however, we do not have an adequate number of sites to permit such an analysis. A larger study is needed.

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 154 30. In addition to the hundreds of operating speed data available from the Austin, Texas, dataset, the research team has been provided speed limit studies for several other sites. As shown on this slide, slightly less than half of our sample is from Texas, and more than half is from other states. 31. This slide shows the 85th and average speed data for 875 speed studies. The yellow squares are the average speed on the roadway for the given posted speed limit, and the red dot is the 85th percentile speed for that site. The graph also shows trendlines for the 85th percentile data with a red dotted line and the average speed with a yellow dashed line. The solid black line represents the condition when the operating speed would equal the posted speed limit. The graph illustrates several relationships. First, since all the trend lines have an upward slope, it demonstrates that overall higher posted speed limits are associated with higher operating

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 155 speeds. Stated in another manner, the number on the speed limit sign does matter, and it does have a relationship with the operating speed. A second relationship that can be seen is that there is a wide range of operating speeds for a given posted speed limit, which indicates that there are other factors that are influencing the operating speed of drivers in addition to the posted speed limit number. 32. The decision rules used in the SLS-Tool were developed based on research findings, consideration of information in the AASHTO Green Book and Highway Safety Manual, the research team’s expert opinions, and feedback from experts including the project panel. 33. As part of 17-76, the research team reviewed the literature and key reference documents like the Highway Safety Manual. Because fewer insights are available for urban/suburban roads,

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 156 NCHRP 17-76 focused Phase II research efforts on those functional class roads. Two databases were created. The one using Washtenaw, Michigan, data (e.g., Ann Arbor) included roadway characteristics, posted speed limits, and crash data. The other database used Austin, Texas, data. The City of Austin had posted several of its speed studies online, and the research team obtained that data along with collecting additional speed data, especially on four-lane streets. 34. This slide shows examples of the key findings from the NCHRP 17-76 Phase II research efforts. USLIMTS2 included decisions rules based on several variables. Based on NCHRP 17-76 research efforts, the research team recommended keeping for the Developed SLSG signal density, access density, and on-street parking variables. The research team suggested the inclusion of median type when the cross section has four or more lanes and the median type is undivided (i.e., none; note TWLTL is not considered to be in the undivided median type group). The NCHRP 17-76 research also expanded upon the use of the 50th percentile speed.

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 157 35. The rest of the presentation showed the decision rules and examples of the tool interface (they were removed from this publication because they were dated material; see Appendix F for discussion of decision rules and the user guide for examples of the SLS-Tool). 36. Thank you for your time.