National Academies Press: OpenBook

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

Chapter: APPENDIX E. RELATIONSHIP AMONG URBAN/SUBURBAN ROADWAY CHARACTERISTICS, POSTED SPEED LIMIT, AND CRASHES IN WASHTENAW COUNTY, MICHIGAN

« Previous: APPENDIX D. RELATIONSHIP AMONG URBAN/SUBURBAN ROADWAY CHARACTERISTICS, OPERATING SPEED, AND CRASHES IN AUSTIN, TEXAS
Page 107
Suggested Citation:"APPENDIX E. RELATIONSHIP AMONG URBAN/SUBURBAN ROADWAY CHARACTERISTICS, POSTED SPEED LIMIT, AND CRASHES IN WASHTENAW COUNTY, MICHIGAN." National Academies of Sciences, Engineering, and Medicine. 2021. Development of a Posted Speed Limit Setting Procedure and Tool. Washington, DC: The National Academies Press. doi: 10.17226/26200.
×
Page 107
Page 108
Suggested Citation:"APPENDIX E. RELATIONSHIP AMONG URBAN/SUBURBAN ROADWAY CHARACTERISTICS, POSTED SPEED LIMIT, AND CRASHES IN WASHTENAW COUNTY, MICHIGAN." National Academies of Sciences, Engineering, and Medicine. 2021. Development of a Posted Speed Limit Setting Procedure and Tool. Washington, DC: The National Academies Press. doi: 10.17226/26200.
×
Page 108
Page 109
Suggested Citation:"APPENDIX E. RELATIONSHIP AMONG URBAN/SUBURBAN ROADWAY CHARACTERISTICS, POSTED SPEED LIMIT, AND CRASHES IN WASHTENAW COUNTY, MICHIGAN." National Academies of Sciences, Engineering, and Medicine. 2021. Development of a Posted Speed Limit Setting Procedure and Tool. Washington, DC: The National Academies Press. doi: 10.17226/26200.
×
Page 109
Page 110
Suggested Citation:"APPENDIX E. RELATIONSHIP AMONG URBAN/SUBURBAN ROADWAY CHARACTERISTICS, POSTED SPEED LIMIT, AND CRASHES IN WASHTENAW COUNTY, MICHIGAN." National Academies of Sciences, Engineering, and Medicine. 2021. Development of a Posted Speed Limit Setting Procedure and Tool. Washington, DC: The National Academies Press. doi: 10.17226/26200.
×
Page 110
Page 111
Suggested Citation:"APPENDIX E. RELATIONSHIP AMONG URBAN/SUBURBAN ROADWAY CHARACTERISTICS, POSTED SPEED LIMIT, AND CRASHES IN WASHTENAW COUNTY, MICHIGAN." National Academies of Sciences, Engineering, and Medicine. 2021. Development of a Posted Speed Limit Setting Procedure and Tool. Washington, DC: The National Academies Press. doi: 10.17226/26200.
×
Page 111
Page 112
Suggested Citation:"APPENDIX E. RELATIONSHIP AMONG URBAN/SUBURBAN ROADWAY CHARACTERISTICS, POSTED SPEED LIMIT, AND CRASHES IN WASHTENAW COUNTY, MICHIGAN." National Academies of Sciences, Engineering, and Medicine. 2021. Development of a Posted Speed Limit Setting Procedure and Tool. Washington, DC: The National Academies Press. doi: 10.17226/26200.
×
Page 112
Page 113
Suggested Citation:"APPENDIX E. RELATIONSHIP AMONG URBAN/SUBURBAN ROADWAY CHARACTERISTICS, POSTED SPEED LIMIT, AND CRASHES IN WASHTENAW COUNTY, MICHIGAN." National Academies of Sciences, Engineering, and Medicine. 2021. Development of a Posted Speed Limit Setting Procedure and Tool. Washington, DC: The National Academies Press. doi: 10.17226/26200.
×
Page 113
Page 114
Suggested Citation:"APPENDIX E. RELATIONSHIP AMONG URBAN/SUBURBAN ROADWAY CHARACTERISTICS, POSTED SPEED LIMIT, AND CRASHES IN WASHTENAW COUNTY, MICHIGAN." National Academies of Sciences, Engineering, and Medicine. 2021. Development of a Posted Speed Limit Setting Procedure and Tool. Washington, DC: The National Academies Press. doi: 10.17226/26200.
×
Page 114
Page 115
Suggested Citation:"APPENDIX E. RELATIONSHIP AMONG URBAN/SUBURBAN ROADWAY CHARACTERISTICS, POSTED SPEED LIMIT, AND CRASHES IN WASHTENAW COUNTY, MICHIGAN." National Academies of Sciences, Engineering, and Medicine. 2021. Development of a Posted Speed Limit Setting Procedure and Tool. Washington, DC: The National Academies Press. doi: 10.17226/26200.
×
Page 115
Page 116
Suggested Citation:"APPENDIX E. RELATIONSHIP AMONG URBAN/SUBURBAN ROADWAY CHARACTERISTICS, POSTED SPEED LIMIT, AND CRASHES IN WASHTENAW COUNTY, MICHIGAN." National Academies of Sciences, Engineering, and Medicine. 2021. Development of a Posted Speed Limit Setting Procedure and Tool. Washington, DC: The National Academies Press. doi: 10.17226/26200.
×
Page 116
Page 117
Suggested Citation:"APPENDIX E. RELATIONSHIP AMONG URBAN/SUBURBAN ROADWAY CHARACTERISTICS, POSTED SPEED LIMIT, AND CRASHES IN WASHTENAW COUNTY, MICHIGAN." National Academies of Sciences, Engineering, and Medicine. 2021. Development of a Posted Speed Limit Setting Procedure and Tool. Washington, DC: The National Academies Press. doi: 10.17226/26200.
×
Page 117
Page 118
Suggested Citation:"APPENDIX E. RELATIONSHIP AMONG URBAN/SUBURBAN ROADWAY CHARACTERISTICS, POSTED SPEED LIMIT, AND CRASHES IN WASHTENAW COUNTY, MICHIGAN." National Academies of Sciences, Engineering, and Medicine. 2021. Development of a Posted Speed Limit Setting Procedure and Tool. Washington, DC: The National Academies Press. doi: 10.17226/26200.
×
Page 118

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 95 APPENDIX E. RELATIONSHIP AMONG URBAN/SUBURBAN ROADWAY CHARACTERISTICS, POSTED SPEED LIMIT, AND CRASHES IN WASHTENAW COUNTY, MICHIGAN OVERVIEW In order to provide guidance for the setting of speed limits consistent with the stated objectives of NCHRP Project 17-76, a safety analysis was conducted specific to urban and suburban roadway segments located in Washtenaw County, Michigan (see Figure 17). This included the collection of historical traffic and crash data, in addition to the collection of other relevant roadway characteristics. Ultimately, the relative safety performance of urban and suburban non-freeway roadway segments was compared with the posted speed limit and other roadway features to provide additional qualitative guidance for practitioners. Figure 17. Location of study roadway segments in Washtenaw County, Michigan. DATABASE DEVELOPMENT Initially, roadway inventory data for all public highways in Washtenaw County, including those under the jurisdiction of the Michigan DOT, Washtenaw County, and the City of Ann Arbor, were collected and merged with daily traffic volume estimates and posted speed limit data, where available, to develop a database of potential study highways. These data were Washtenaw County

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 96 then reduced to include only non-freeway and non-local roadways with urban designations where the posted speed limit ranged between 25 to 50 mph. These candidate roadways were then segmented such that each roadway segment’s end points were intersections controlled via either signalization, stop control, a roundabout (yield control), or the route otherwise ending (such as the county line). A manual review of each segment was then conducted to determine additional roadway features (such as the presence of sidewalks, lane width, etc.). This process resulted in a total of 586 distinct roadway segments encompassing approximately 312 mi. The following subsections describe the process to develop the database for subsequent analysis. Roadway Inventory Data Roadway inventory data for Washtenaw County were collected via the Michigan Geographic Framework (MGF), which includes all public roadways in the state of Michigan (216). The MGF provides a linear referencing system that is used by several agencies in the state of Michigan and allows for merging data from other sources with these roadway inventory data (such as traffic crashes or the posted speed limit). Figure 18 shows a map of the approximately 2,954 mi of public roadways in Washtenaw County along with the adjusted census urban boundaries (ACUBs), as indicated by the pink shading. The major urban areas in Washtenaw County include the cities of Ann Arbor, Ypsilanti, Saline, and Chelsea. Figure 18. Washtenaw County roadway network and urban boundaries. Identification of Study Roadway Segments Given the roadway inventory data shown in Figure 18 for Washtenaw County, daily traffic volume estimates and posted speed limits for each roadway were assigned to each roadway via ArcGIS. AADT volume estimates were acquired from the FHWA Highway Performance Monitoring System (HPMS) shapefile as well as from the Southeast Michigan Ypsilanti Ann Arbor Chelsea Saline Washtenaw County

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 97 Council of Government’s (SEMCOG’s) open data portal (217, 218). Posted speed limit data were also acquired from SEMCOG’s open data portal. These data were used to identify all public roadways that met the criteria for inclusion as a potential study roadway, as shown in Table 38. Table 38. Summary of criteria for inclusion as potential study roadway. Characteristic Criteria Posted Speed Limit 25 to 50 mph per SEMCOG’s open data portal National Functional Class Includes other principal arterial, minor arterial, major collector, minor collector Excludes interstates, other freeways, and local Historical Traffic Volume Must include recent AADT estimate from either HPMS or SEMCOG Urban Boundary Includes roadways that fall within or extend from ACUB per MGF Segmentation of Roadway Data Given these potential study roadways, a spatial analysis was performed in ArcGIS to identify the location of all public roadway intersections along these highways. The traffic control for each intersection (signal, all-way stop, minor route stop, or roundabout) was then determined via a manual review of satellite imagery. These data were then used to segment the roadway inventory data such that each roadway segment’s end points were intersections controlled via either signalization, stop control, a roundabout (yield control), or the route otherwise ending (such as the county line). This ensured that there was no traffic control along the major route within the bounds of each segment. All segments less than 0.06 mi were excluded from further analysis. All segments that ranged from 0.06 to 0.10 mi in length were manually reviewed via satellite imagery to determine if the roadway inventory data were representative of actual field conditions, and atypical segments were screened from further analysis. An example of output from this segmentation process is shown in Figure 19. Figure 19. Segmentation of study road segments in Washtenaw County, Michigan. After the identification and segmentation process was complete, a total of 586 distinct study roadway segments encompassing approximately 312.7 mi were selected for further analysis. A map of the study roadway segments in Washtenaw County is provided in Figure 20.

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 98 A map of study segments by posted speed limit is presented in Figure 21, and a map of study segments by AADT is presented in Figure 22. Figure 20. Map of study road segments in Washtenaw County, Michigan. Figure 21. Study road segments by posted speed limit. Washtenaw County

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 99 Figure 22. Study road segments by AADT estimate. Historical Traffic Crash Data Historical traffic crash data were collected from annual databases maintained by the Michigan State Police (MSP), including a 5-year period from 2013 to 2017. These data were assigned to each study segment based upon the location along the MGF associated with each crash record. Each crash was also determined to be intersection related if it occurred within 250 ft of an intersection at either end point of the segment. All remaining records were considered to be midblock segment crashes for the purposes of this evaluation. Only those crashes occurring on the subject roadway were counted, and crashes occurring on the side streets and driveways were excluded. This allocation scheme was used in order to be consistent with data collected in Austin, Texas, specific to NCHRP Project 17-76. Figure 23 shows an example of the crash data allocation process. Figure 23. Example of traffic crash locations along study road segments (2013–2017). Intersection Crashes Midblock Crashes

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 100 Additional Geometric and Roadway Characteristics After the identification of the study road segments, an additional manual review of satellite imagery was performed in order to collect a variety of geometric and other roadway characteristics specific to each segment. These characteristics are summarized in Table 39. Table 39. Roadway and traffic control device variables for Washtenaw County, Michigan. Column Heading Description AADTR AADT estimate from FHWA HPMS or SEMCOG system-wide data rounded to nearest 10 AccessDen Data for DrvUsigPerMileBoth regrouped into three levels: LT40=less than 40 driveways/unsignalized intersections per mile, 40to60=between 40 and 60 driveways/unsignalized intersections per mile, and GT60=greater than 60 driveways/unsignalized intersections per mile Bike_Lane Bike lane presence: 1=yes, 0=no Bus Bus stop presence collected via “Places” in Google Earth: 1=yes, 0=no Crosswalk Presence of a midblock crosswalk in between end points (do not include crosswalks located adjacent to end points): 1=yes, 0=no Curb_1yes Is curb present (for at least 50% of segment): yes=1, no=0 Develop Development (revised to match Austin variable): Residential, RuralParks, or ComRetInd (for commercial, retail, or industrial) DrvUsigPerMileBoth Per mile rate for driveways (residential and commercial) and public intersections along the segment in both directions FuncClass Revised National Functional Class Code Horz Horizontal alignment: 0=straight, 1=some horizontal curvature Int_Type1 Description of End Point #1: signal, all-way stop, stop, roundabout, or other break point Int_Type2 Description of End Point #2: signal, all-way stop, stop, roundabout, or other break point LnWdG Lane width (ft) for the segment grouped into N=Narrow (7, 8, 9, or 10 ft), T=Typ (11 or 12 ft), W=Wide (13 ft or more) Median Median type: none (undivided), raised (also depressed), TWLTL (also delineated) MedWidth Median width (ft) Miles Total length in miles (accounts for divided roadways in two pieces) Num_Lanes Number of through lanes (not including exclusive turn lanes) NumSigInt Number of signalized intersections along segment, including any signals at the beginning or end of the segment OnStreet_Parking On-street parking: yes=1 or no=0 PedAuto The typical or average distance between the sidewalk (or where the pedestrian could be walking) and the automobile lane for the corridor (ft) POSTED_SPE Posted speed limit per SEMCOG system-wide data RoadSurf Distance between the driving surface edges, including consideration of travel lanes, median width, on-street parking lane width, and bike lane width School School zone presence collected via “Places” in Google Earth: 1=yes, 0=no Sidewalk Sidewalk presence, 0=none, 1=one side, 2=both sides (based on majority of segment)

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 101 DATA ANALYSIS—METHODOLOGY The research team compiled the dataset consisting of 530 sites to conduct the analysis. Table 40 provides the descriptive statistics for the variables considered in the analysis. Table 40. Descriptive statistics of variables for Washtenaw County, Michigan. Variablea Variable Typeb Minimum Maximum Mean Std. Dev. AADTR Numerical 800 44,700 11,726.83 7,964.25 DrvUsigPerMileBoth Numerical 0 142.9 39.40 31.31 Miles Numerical 0.06 6.839 0.57 0.63 PedAuto Numerical 0 75 10.67 10.15 RoadSurf Numerical 20 108 38.78 15.31 MedWidth Numerical 0 60 4.72 7.97 Num_Lanes Numerical 2 4 2.58 0.91 NumSigInt Numerical 0 2 1.49 0.70 POSTED_SPE Numerical 25 50 35.55 8.31 Crosswalk Dichotomous 0 1 0.19 0.40 Curb_1yes Dichotomous 0 1 0.75 0.44 Horz Dichotomous 0 1 0.18 0.38 OnStreet_Parking Dichotomous 0 1 0.17 0.38 School Dichotomous 0 1 0.05 0.21 Bike_Lane Dichotomous 0 1 0.22 0.42 Bus Dichotomous 0 1 0.43 0.50 Sidewalk_1yes Dichotomous 0 1 0.82 0.38 Develop Polychotomous ComRetInd (221), Residential (261), Rural/Parks (48) FuncClass Polychotomous Coll (164), MinArt (220), PrinArt (146) Int_Type1 Polychotomous AllWay (81), Break (4), Roundabout (13), Signal (419), Stop (13) Int_Type2 Polychotomous AllWay (49), Break (39), Roundabout (19), Signal (369), Stop (54) LnWdG Polychotomous N (46), T (460), W (24) Median Polychotomous Raised (16), TWLTL (183), None (331) Sidewalk Polychotomous None (94), One Side (89), Both Sides (347) AccessDen Polychotomous LT40 (327), 40to60 (78), GT60 (125) a Variable descriptions are in Table 39. b For dichotomous variables, “1” indicates the presence of the feature, and “0” indicates its absence. For polychotomous variables, the numbers in parentheses represent frequencies of the corresponding categories. Regression Model Overview The research team applied NB regression models to the site-level data to investigate the relationship between crashes and posted speed limit while controlling for the effects of other variables including the natural logarithm of AADT and roadway geometry variables. The research team examined several NB regression models. The models included the log of segment length as an offset variable.

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 102 RESULTS AND DISCUSSION Variable Relationships with Crashes The investigation into the relationships among the roadway characteristics, posted speed limit, and traffic crashes focused on segment crashes rather than all crashes that included intersections. The research team considered both crashes with fatal and injuries (i.e., KABC) and all severity level crashes (i.e., KABCO). Injury Segment Crashes (KABC Midblock) Multiple models were developed to identify the model containing several variables that were significant at α=0.1 or smaller. Table 41 provides the significance per variable, while Table 42 provides the estimate along with the p-value for each variable/level combination for the selected model. Key observations include the following:  As expected, more vehicle volume was associated with more crashes.  Higher posted speed limits were associated with more crashes.  Minor arterials and collectors, compared to principal arterials, were associated with fewer crashes.  The presence of a midblock crosswalk was associated with more vehicle crashes.  Contrary to what was found in Austin, the presence of a sidewalk was found to be associated with fewer crashes.  A greater number of access points was associated with more crashes. For the model shown in Table 41 and Table 42, the access points were grouped into the levels used in USLIMITS2. Roads with more than 60 access points per mile had more crashes than roads with between 40 to 60 access points. Roads with between 40 and 60 access points had more crashes than roads with fewer than 40 access points.  Greater distances between pedestrians and autos were associated with more crashes. The research team currently does not have a theory as to why pedestrians being a greater distance away would be associated with more vehicle crashes. Table 41. Score statistics for model shown in Table 42. Variable DF Chi-Square Pr > ChiSq LnVol 1 65.20 <.0001 POSTED_SPE 1 12.25 0.0005 FuncClass 2 12.91 0.0016 AccessDen 2 11.50 0.0032 Crosswalk 1 6.66 0.0099 PedAuto 1 4.96 0.0259 Sidewalk_1yes 1 4.09 0.0431 Note: LR statistics for Type 3 analysis. Variable descriptions are in Table 39. LnVol=Log(AADTR).

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 103 Table 42. Variables with significant effects on KABC segment crashes. Variable Level DF Estimate Standard Error Wald 95% Confidence Limits Wald Chi- Square Pr > ChiSq Intercept 1 −4.9187 0.8432 −6.5713 −3.2660 34.03 <.0001 FuncClass Coll 1 −0.5925 0.1736 −0.9328 −0.2522 11.65 0.0006 FuncClass MinArt 1 −0.3516 0.1111 −0.5693 −0.1338 10.01 0.0016 FuncClass PrinArt 0 0.0000 0.0000 0.0000 0.0000 . . Crosswalk 1 0.2588 0.1013 0.0602 0.4574 6.53 0.0106 PedAuto 1 0.0103 0.0047 0.0011 0.0195 4.85 0.0276 POSTED_SPE 1 0.0227 0.0065 0.0100 0.0354 12.22 0.0005 Sidewalk_1yes 1 −0.2930 0.1453 −0.5778 −0.0081 4.06 0.0439 AccessDen LT40 0 0.0000 0.0000 0.0000 0.0000 . . AccessDen 40to60 1 0.2009 0.1164 −0.0272 0.4291 2.98 0.0843 AccessDen GT60 1 0.3669 0.1109 0.1496 0.5843 10.95 0.0009 LnVol 1 0.6812 0.0805 0.5234 0.8389 71.65 <.0001 Dispersion 1 0.3813 0.0487 0.2969 0.4897 NR NR Notes: Analysis of maximum likelihood parameter estimates. Variable descriptions are in Table 39. LnVol=Log(AADTR). . = value is not relevant since this level is the base for the variable. {blank} = value not relevant because the variable is not a multicategory variable. NR = value not relevant for the dispersion variable. All Segment Crashes (KABCO Midblock) Similar to the effort for midblock injury crashes, the research team developed several NB regression models to understand the relationship between all segment crashes and roadway characteristics. Table 43 provides the significance per variable, while Table 44 provides the estimate along with the p-value for each variable/level combination for the selected model. The following observations are similar to the findings for midblock injury crashes:  As expected, more vehicle volume was associated with more crashes.  Minor arterials and collectors, compared to principal arterials, were associated with fewer crashes.  A greater number of access points was associated with more crashes.  Contrary to what was found in Austin, the presence of a sidewalk was found to be associated with fewer crashes.  Greater distances between pedestrians and autos were associated with more crashes. The research team currently does not have a theory as to why pedestrians being a greater distance away would be associated with more vehicle crashes. Findings that were either not present or were different from those found with KABC midblock include the following:  Wider median widths were found to be associated with more crashes, which is contrary to the finding for the Austin data.  More signals were associated with more crashes, which agrees with the finding from Austin, but the relationship was not present when examining Washtenaw’s injury crashes.  Posted speed limit was significant for injury crashes (see Table 41) but not all crashes.

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 104  Median type was significant for all severity level crashes, with the presence of a raised median being associated with fewer crashes than either a TWLTL or no median, which is consistent with the findings for data from Austin. The presence of a TWLTL was also associated with fewer crashes when compared to no median, a finding that is contradictory in this case to the KABC NID crash data from Austin. Table 43. Score statistics for model shown in Table 46. Variable DF Chi-Square Pr > ChiSq Sidewalk_1yes 1 17.14 <.0001 LnVol 1 67.92 <.0001 Crosswalk 1 12.22 0.0005 FuncClass 2 14.89 0.0006 MedWidth 1 11.16 0.0008 NumSigInt 1 7.43 0.0064 Median 2 10.00 0.0067 PedAuto 1 7.11 0.0077 AccessDen 2 7.71 0.0211 Notes: LR statistics for Type 3 analysis. Variable descriptions are in Table 39. LnVol=Log(AADTR). Table 44. Variables with significant effects on KABCO segment crashes. Variable Level DF Estimate Standard Error Wald 95% Confidence Limits Wald Chi- Square Pr > ChiSq Intercept 1 −2.2141 0.6991 −3.5844 −0.8439 10.03 0.0015 FuncClass Coll 1 −0.5590 0.1462 −0.8455 −0.2725 14.63 0.0001 FuncClass MinArt 1 −0.3256 0.1027 −0.5268 −0.1243 10.05 0.0015 FuncClass 0 0.0000 0.0000 0.0000 0.0000 . . Crosswalk 1 0.3108 0.0903 0.1338 0.4879 11.84 0.0006 Median Raised 1 −1.3487 0.4191 −2.1701 −0.5272 10.35 0.0013 Median TWLTL 1 −0.2722 0.1410 −0.5485 0.0041 3.73 0.0535 Median None 0 0.0000 0.0000 0.0000 0.0000 . . MedWidth 1 0.0330 0.0101 0.0132 0.0529 10.63 0.0011 PedAuto 1 0.0112 0.0043 0.0028 0.0196 6.80 0.0091 Sidewalk_1yes 1 −0.5020 0.1223 −0.7418 −0.2622 16.84 <.0001 NumSigInt 1 0.1844 0.0671 0.0529 0.3159 7.56 0.0060 AccessDen LT40 0 0.0000 0.0000 0.0000 0.0000 . . AccessDen 40to60 1 0.1879 0.1046 −0.0171 0.3928 3.23 0.0725 AccessDen GT60 1 0.2376 0.0939 0.0535 0.4217 6.40 0.0114 LnVol 1 0.6337 0.0729 0.4908 0.7766 75.56 <.0001 Dispersion 1 0.5145 0.0421 0.4382 0.6040 NR NR Notes: Analysis of maximum likelihood parameter estimates. Variable descriptions are in Table 39. LnVol=Log(AADTR). . = value is not relevant since this level is the base for the variable. {blank} = value not relevant because the variable is not a multicategory variable. NR = value not relevant for the dispersion variable.

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 105 Variable Relationships with Posted Speed Limit A better understanding of relationships between roadway characteristics and the posted speed limit may assist with identifying how the look and feel of a road could help to communicate the appropriate operating speed along with the posted speed for the facility. Table 45 provides a summary of three models that explore the significance for several variables with posted speed limit. These results illustrate how certain roadway characteristics tend to be associated with different posted speed limits. Key observations include the following:  Certain variables were always significant at a 0.05 level or better regardless of the method used to select the variables. The variables with a strong association with posted speed limit included: o Number of driveways/unsignalized intersections per mile. o Functional classification. o Presence of horizontal curvature within the segment. o Presence of median. o Presence of on-street parking. o Distance between vehicles and pedestrians (PedAuto).  Variables that were not significant at the 0.05 level included: o Develop, which indicates that a range of posted speed limits exist for the three development levels used: residential; commercial, retail, or industrial; and rural/parks. o Lane width for the segment. o Number of signalized intersections. While the number of signalized intersections was found to not be significant with respect to posted speed limit, it was with respect to crashes. o Road surface width. o Presence of school. FINDINGS’ IMPACT ON SLS-TOOL The findings from this effort support the following decision rules for the SLS-Tool:  Inclusion of the following variables: o Signal density (represented by the variable NumSigInt in the models). o Access density, along with the break points at 40 and 60 access points per mile.  Addition of the following variable: o Median type (raised medians were associated with fewer KABCO crashes compared to none or TWLTL, and TWLTLs were associated with fewer KABCO crashes compared to no median).

NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool 106 Table 45. Models exploring the significance of variables for posted speed limit. Variable DF Model 1 Chi- Square Model 1 Pr > ChiS q Model 2 Chi- Square Model 2 Pr > ChiSq Model 3 Chi- Square Model 3 Pr > ChiS q AADTR 1 3.6 0.0579 3.54 0.0598 8.15 0.0043 Crosswalk 1 3.73 0.0534 3.17 0.075 20.11 <.0001 Curb_1yes 1 18.18 <.0001 23.27 <.0001 3.52 0.1717 Develop 2 4.28 0.1176 NI NI NI NI DrvUsigPerMileBoth 1 11 0.0009 10.88 0.001 11.89 0.0006 FuncClass 2 28.38 <.0001 41.1 <.0001 31.85 <.0001 Horz 1 6.32 0.0119 6.86 0.0088 8.19 0.0042 LnWdG 2 4.24 0.1202 NI NI 5.28 0.0715 Median 2 39.69 <.0001 44.19 <.0001 47.61 <.0001 Miles 1 35.91 <.0001 39.43 <.0001 35 <.0001 Num_Lanes 1 5.16 0.0232 6.75 0.0094 NI NI NumSigInt 1 1.85 0.1736 NI NI NI NI OnStreet_Parking 1 6.51 0.0107 6.02 0.0141 20.73 <.0001 PedAuto 1 28.04 <.0001 30.26 <.0001 24.13 <.0001 RoadSurf 1 0.87 0.3515 1.41 0.2355 NI NI School 1 1.43 0.2325 NI NI NI NI Sidewalk_1yes 1 23.9 <.0001 24.88 <.0001 23.11 <.0001 Note: LR statistics for Type 3 analysis. Variable descriptions are in Table 39. Model 1: Variables selected based on P-value threshold (probability to enter: 0.25, probability to leave: 0.1). Model 2: Variables selected based on AIC. Model 3: Variables selected based on BIC. NI = variable not included in the model.

Next: APPENDIX F. DECISION RULE DEVELOPMENT »
Development of a Posted Speed Limit Setting Procedure and Tool Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Several types of speed limits exist, including statutory speed limit, posted speed limit, school zone speed limit, work zone speed limit, variable speed limit, and advisory speed.

The TRB National Cooperative Highway Research Program's NCHRP Web-Only Document 291: Development of a Posted Speed Limit Setting Procedure and Tool documents the research efforts and findings from an NCHRP Project 17-76 to identify factors that influence a driver’s operating speed and the development of a Speed Limit Setting Procedure and Tool.

The document is supplemental to NCHRP Research Report 966: Posted Speed Limit Setting Procedure and Tool: User Guide.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!