Crash Modification Factors in the Highway Safety Manual: Resources for Evaluation (2023)

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.

Appendix M Calibration of Prediction Models of Rural and Urban Intersections for the 2nd Edition of the HSM M-1 

Table of Contents 1. Introduction ...................................................................................................................................... M-6 2. Summary of Data.............................................................................................................................. M-7 3. Methodology and Models ............................................................................................................... M-12 4. Calibration Results ........................................................................................................................ M-28 5. Validation of Adjustment Factors .................................................................................................. M-40 5.1 Rural Two-Lane, 3-Leg Stop Controlled Intersections (Rur2L-3ST) ...................................... M-41 Lighting ...................................................................................................................................... M-41 Left Turn Bay ............................................................................................................................. M-42 Right Turn Bay ........................................................................................................................... M-44 Intersection Skew ....................................................................................................................... M-45 5.2 Rural Two-Lane, 4-Leg Stop Controlled Intersections (Rur2L-4ST) ...................................... M-45 Lighting ...................................................................................................................................... M-45 Left Turn Bay ............................................................................................................................. M-46 Right Turn Bay ........................................................................................................................... M-47 Intersection Skew ....................................................................................................................... M-48 5.3 Rural Two-Lane, 4-Leg Signalized Intersections (Rur2L-4SG) .............................................. M-48 Lighting ...................................................................................................................................... M-48 Left Turn Bay ............................................................................................................................. M-49 Right Turn Bay ........................................................................................................................... M-51 5.4 Rural Multi-Lane, 3-Leg Stop Controlled Intersections (RurML-3ST) ................................... M-51 Lighting ...................................................................................................................................... M-52 Left Turn Bay ............................................................................................................................. M-52 Right Turn Bay ........................................................................................................................... M-53 Intersection Skew ....................................................................................................................... M-53 5.5 Rural Multi-Lane, 4-Leg Stop Controlled Intersections (RurML-4ST) ................................... M-55 Lighting ...................................................................................................................................... M-55 Left Turn Bay ............................................................................................................................. M-55 Right Turn Bay ........................................................................................................................... M-56 Intersection Skew ....................................................................................................................... M-57 5.6 Rural Multi-Lane, 4-Leg Signalized Intersections (RurML-4SG) ........................................... M-58 Left Turn Bay ............................................................................................................................. M-58 Right Turn Bay ........................................................................................................................... M-59 5.7 Urban Arterial, 3-Leg Stop Controlled Intersections (UrbArt-3ST) ........................................ M-59 Lighting ...................................................................................................................................... M-59 Left Turn Bay ............................................................................................................................. M-61 Right Turn Bay ........................................................................................................................... M-63 M-2 

5.8 Urban Arterial, 3-Leg Signalized Intersections (UrbArt-3SG) ................................................ M-63 Left Turn Bay ............................................................................................................................. M-64 Right Turn Bay ........................................................................................................................... M-64 Red Light Running Camera Enforcement .................................................................................. M-65 Right Turn on Red Prohibition ................................................................................................... M-66 5.9 Urban Arterial, 4-Leg Stop Controlled Intersections (UrbArt-4ST) ........................................ M-68 Lighting ...................................................................................................................................... M-68 Left Turn Bay ............................................................................................................................. M-69 Right Turn Bay ........................................................................................................................... M-70 5.10 Urban Arterial, 4-Leg Signalized Intersections (UrbArt-4SG) .............................................. M-71 Lighting ...................................................................................................................................... M-71 Left Turn Bay ............................................................................................................................. M-71 Right Turn Bay ........................................................................................................................... M-72 Red Light Running Camera Enforcement .................................................................................. M-73 Right Turn on Red Prohibition ................................................................................................... M-75 Table 1. Intersection Summary Data (Calibration Dataset) ..................................................................... M-8 Table 2. Summary of Volume Data (SPF Estimation Dataset) ................................................................ M-9 Table 3. Summary of Crash Data for Intersections on Rural Highways ................................................ M-10 Table 4. Summary of Crash for Roundabouts ........................................................................................ M-10 Table 5. Summary of Crash Data for Intersections on Urban Arterials ................................................. M-11 Table 6. Summary of Crash Data for All Way Stop Controlled Intersections ....................................... M-11 Table 7. Base Condition SPFs, Rur2L-3ST............................................................................................ M-13 Table 8. Base Condition SPFs using Poisson Distribution, Rur2L-3ST ................................................ M-14 Table 9. Base Condition SPFs, Rur2L-4ST............................................................................................ M-15 Table 10. Base Condition SPFs using Poisson Distribution, Rur2L-4ST .............................................. M-16 Table 11. Base Condition SPFs, Rur2L-4SG ......................................................................................... M-17 Table 12. Base Condition SPFs using Poisson Distribution, Rur2L-4SG .............................................. M-18 Table 13. Base Condition SPFs, RurML-3ST ........................................................................................ M-19 Table 14. Base Condition SPFs, RurML-4ST ........................................................................................ M-20 Table 15. Base Condition SPFs, RurML-4SG........................................................................................ M-21 Table 16. Base Condition SPFs, UrbArt-3ST ........................................................................................ M-22 Table 17. Base Condition SPFs, UrbArt-3SG ........................................................................................ M-23 Table 18. Base Condition SPFs, UrbArt-4ST ........................................................................................ M-24 Table 19. Base Condition SPFs, UrbArt-4SG ........................................................................................ M-25 Table 20. Base Condition SPFs, F&I Crashes, UrbArt6+ Intersections................................................. M-26 Table 21. Base Condition SPFs, PDO Crashes, UrbArt6+ Intersections ............................................... M-26 Table 22. Base Condition SPFs, F&I Crashes, UrbArtOW Intersections .............................................. M-26 Table 23. Base Condition SPFs, PDO Crashes, UrbArtOW Intersections ............................................. M-26 Table 24. Base Condition SPFs, Total Crashes, 3-Leg Rural Signalized Intersections ......................... M-27 Table 25. Base Condition SPFs, All Way Stop Controlled Intersections............................................... M-27 Table 26. Base Condition SPFs. Roundabouts ....................................................................................... M-28 Table 27. Calibration Results and Recommended Calibration Factors, Rural 3-Leg Signalized Intersections ..................................................................................................................................... M-29 Table 28. Calibration Results and Recommended Calibration Factors, Rur2L-3ST (n=208)................ M-29 M-3 

Table 29. Calibration Results and Recommended Calibration Factors, Rur2L-4ST (n=234)................ M-30 Table 30. Calibration Results and Recommended Calibration Factors, Rur2L-4SG (n=105) ............... M-31 Table 31. Calibration Results and Recommended Calibration Factors, RurML-3ST (n=14) ................ M-32 Table 32. Calibration Results and Recommended Calibration Factors, RurML-4ST (n=21) ................ M-33 Table 33. Calibration Results and Recommended Calibration Factors, RurML-4ST (n=28) ................ M-34 Table 34. Calibration Results and Recommended Calibration Factors, UrbArt-3ST (n=72) ................ M-35 Table 35. Calibration Results and Recommended Calibration Factors, UrbArt-3SG (n=9) .................. M-36 Table 36. Calibration Results and Recommended Calibration Factors, UrbArt-4ST (n=25) ................ M-37 Table 37. Calibration Results and Recommended Calibration Factors, UrbArt-4SG (n=129) .............. M-38 Table 38. Calibration Results and Recommended Calibration Factors, UrbArt6+ Intersections ........... M-39 Table 39. Calibration Results and Recommended Calibration Factors, UrbArtOW Intersections ........ M-39 Table 40. Calibration Results and Recommended Calibration Factors, All Way Stop Controlled Intersections ..................................................................................................................................... M-39 Table 41. Calibration Results and Recommended Calibration Factors, Roundabouts ........................... M-39 Table 42. Illustration of Approach 1 ...................................................................................................... M-40 Table 43. Recommended Rur2L-3ST Lighting Adjustment Factors ..................................................... M-42 Table 44. Recommended Rur2L-3ST Left Turn Bay Adjustment Factors ............................................ M-42 Table 45. Rur2L-3ST Left Turn Bay Approach 1 .................................................................................. M-43 Table 46. Rur2L-3ST Left Turn Bay Approach 2, estimate (standard error) ......................................... M-43 Table 47. Recommended Rur2L-3ST Right Turn Bay Adjustment Factors .......................................... M-44 Table 48. Rur2L-3ST Right Turn Bay Approach 1 ................................................................................ M-44 Table 49. Recommended Rur2L-3ST Skew Angle Adjustment Factors................................................ M-45 Table 50. Rur2L-3ST Skew Angle Approach 3 ..................................................................................... M-45 Table 51. Recommended Rur2L-4ST Lighting Adjustment Factors ..................................................... M-46 Table 52. Recommended Rur2L-4ST Left Turn Bay Adjustment Factors ............................................ M-46 Table 53. Recommended Rur2L-4ST Right Turn Bay Adjustment Factors .......................................... M-47 Table 54. Rur2L-4ST Right Turn Bay Approach 1 ................................................................................ M-47 Table 55. Recommended Rur2L-4ST Skew Angle Adjustment Factors................................................ M-48 Table 56. Rur2L-4ST Skew Angle Approach 3 ..................................................................................... M-48 Table 57. Recommended Rur2L-4SG Lighting Adjustment Factors ..................................................... M-49 Table 58. Recommended Rur2L-4SG Left Turn Bay Adjustment Factors ............................................ M-49 Table 59. Rur2L-4SG Left Turn Bay Approach 1 ................................................................................. M-50 Table 60. Rur2L-4SG Left Turn Bay Approach 2, estimate (standard error) ........................................ M-50 Table 61. Recommended Rur2L-4SG Right Turn Bay Adjustment Factors .......................................... M-51 Table 62. Rur2L-4SG Right Turn Bay Approach 1 ............................................................................... M-51 Table 63. Recommended RurML-3ST Lighting Adjustment Factors .................................................... M-52 Table 64. Recommended RurML-3ST Left Turn Bay Adjustment Factors ........................................... M-53 Table 65. Recommended RurML-3ST Right Turn Bay Adjustment Factors ......................................... M-53 Table 66. Recommended RurML-3ST Skew Angle Adjustment Factors .............................................. M-54 Table 67. RurML-3ST Skew Angle Approach 3.................................................................................... M-54 Table 68. Recommended RurML-4ST Lighting Adjustment Factors .................................................... M-55 Table 69. Recommended RurML-4ST Left Turn Bay Adjustment Factors ........................................... M-56 Table 70. Recommended RurML-4ST Right Turn Bay Adjustment Factors ......................................... M-56 Table 71. RurML-4ST Right Turn Bay Approach 1 .............................................................................. M-57 Table 72. Recommended RurML-4ST Skew Angle Adjustment Factors .............................................. M-58 Table 73. Recommended RurML-4SG Left Turn Bay Adjustment Factors .......................................... M-58 Table 74. RurML-4SG Left Turn Bay Approach 1 ................................................................................ M-59 Table 75. Recommended RurML-4SG Right Turn Bay Adjustment Factors ........................................ M-59 Table 76. Recommended UrbArt-3ST Lighting Adjustment Factors .................................................... M-60 Table 77. UrbArt-3ST Lighting Approach 1 .......................................................................................... M-60 Table 78. UrbArt-3ST Lighting Approach 2, estimate (standard error) ................................................. M-61 M-4 

Table 79. Recommended UrbArt-3ST Left Turn Bay Adjustment Factors ........................................... M-61 Table 80. UrbArt-3ST Left Turn Bay Approach 1 ................................................................................. M-62 Table 81. UrbArt-3ST Left Turn Bay Approach 2, estimate (standard error)........................................ M-62 Table 82. Recommended UrbArt-3ST Right Turn Bay Adjustment Factors ......................................... M-63 Table 83. UrbArt-3ST Right Turn Bay Approach 1............................................................................... M-63 Table 84. Recommended UrbArt-3SG Left Turn Bay Adjustment Factors ........................................... M-64 Table 85. Recommended UrbArt-3SG Right Turn Bay Adjustment Factors ......................................... M-64 Table 86. UrbArt-3SG Right Turn Bay Approach 1 .............................................................................. M-64 Table 87. Recommended UrbArt-3SG Red Light Camera Enforcement Adjustment Factors............... M-66 Table 88. Recommended UrbArt-3SG Right Turn on Red Prohibition Adjustment Factors ................. M-68 Table 89. Recommended UrbArt-4ST Lighting Adjustment Factors .................................................... M-69 Table 90. UrbArt-4ST Lighting Approach 1 .......................................................................................... M-69 Table 91. Recommended UrbArt-4ST Left Turn Bay Adjustment Factors ........................................... M-70 Table 92. UrbArt-4ST Left Turn Bay Approach 1 ................................................................................. M-70 Table 93. Recommended UrbArt-4ST Right Turn Bay Adjustment Factors ......................................... M-70 Table 94. Recommended UrbArt-4SG Lighting Adjustment Factors .................................................... M-71 Table 95. Recommended UrbArt-4SG Left Turn Bay Adjustment Factors ........................................... M-72 Table 96. UrbArt-4SG Left Turn Bay Approach 1 ................................................................................ M-72 Table 97. Recommended UrbArt-4SG Right Turn Bay Adjustment Factors ......................................... M-72 Table 98. UrbArt-4SG Right Turn Bay Approach 1 .............................................................................. M-73 Table 99. Recommended UrbArt-3SG Red Light Camera Enforcement Adjustment Factors............... M-74 Table 100. Recommended UrbArt-3SG Right Turn on Red Prohibition Adjustment Factors ............... M-76 M-5 

1. INTRODUCTION This Appendix documents the production of common-jurisdiction calibration factors for the rural and urban intersection base condition crash prediction models developed in NCHRP Projects 17-58, 17-62, 17- 68, and 17-70 and that are expected to be included in the 2nd edition of the HSM. For the purposes of developing the calibration factors, data were obtained from North Carolina for several intersection types for the period from 2013 to 2019: • Rural two-lane, 3-leg stop controlled intersections (Rur2L-3ST) • Rural two-lane, 3-leg signalized intersections (Rur2L-3SG) • Rural two-lane, 4-leg stop controlled intersections (Rur2L-4ST) • Rural two-lane, 4-leg signalized intersections (Rur2L-4SG) • Rural multi-lane, 3-leg stop controlled intersections (RurML-3ST) • Rural multi-lane, 3-leg signalized intersections (RurML-3SG) • Rural multi-lane, 4-leg stop controlled intersections (RurML-4ST) • Rural multi-lane, 4-leg signalized intersections (RurML-4SG) • Urban arterial, 3-leg stop controlled intersections (UrbArt-3ST) • Urban arterial, 3-leg signalized intersections (UrbArt-3SG) • Urban arterial, 4-leg stop controlled intersections (UrbArt-4ST) • Urban arterial, 4-leg signalized intersections (UrbArt-4SG) • Urban arterial (6+ lanes), 3-leg stop controlled intersections (UrbArt6+-3ST) • Urban arterial (6+ lanes), 3-leg signalized intersections (UrbArt6+-3SG) • Urban arterial (6+ lanes), 4-leg stop controlled intersections (UrbArt6+-4ST) • Urban arterial (6+ lanes), 4-leg signalized intersections (UrbArt6+-4SG) • Urban arterial, one-way, 3-leg stop controlled intersections (UrbArtOW-3ST) • Urban arterial, one-way, 3-leg signalized intersections (UrbArtOW-3SG) • Urban arterial, one-way, 4-leg stop controlled intersections (UrbArtOW-4ST) • Urban arterial, one-way, 4-leg signalized intersections (UrbArtOW-4SG) • Rural two-lane, 4-leg all way stop controlled intersections (Rur4Leg-AWSC) • Urban arterial, 3-leg all way stop controlled intersections (Urb3Leg-AWSC) • Urban arterial, 4-leg all way stop controlled intersections (Urb4Leg-AWSC) • Rural roundabouts (Rur-RndAbt) • Urban, single-lane roundabouts (Urb-RndAbtSL) • Urban, multi-lane roundabouts (Urb-RndAbtML) Due to small sample sizes that would result by limiting the sites to only those fitting the base conditions and the AADT ranges used for SPF estimation, all intersections were included (including those outside of AADT ranges), and the appropriate CMFs applied. For intersections on rural highways (Rur2L-3ST, Rur2L-4ST, Rur2L-4SG, RurML-3ST, RurML-4ST, and RurML-4SG) the base conditions, for which CMFs are available, are as follows: • No lighting. • Absence of left turn bay. • Absence of right turn bay. • No skew (i.e., intersection angle = 90 deg.) M-6 

For 3-leg signalized intersections on rural highways (Rur2L-3SG and RurML-3SG), the base conditions, for which CMFs are available, are as follows: • No lighting. • Absence of left turn lanes. • Absence of right turn lanes. For intersections on urban arterials (UrbArt-3ST, UrbArt-3SG, UrbArt-4ST, and UrbArt-4SG) the base conditions, for which CMFs are available, are as follows: • Presence of lighting. • Absence of left turn bay. • Absence of right turn bay. • No red-light running camera enforcement (for signalized intersections only). • Right turn on red permitted on all approaches (for signalized intersections only). For intersections on urban arterials with 6+ lanes (UrbArt6+-3ST, UrbArt6+-3SG, UrbArt6+-4ST, and UrbArt6+-4SG) the base conditions, for which CMFs are available, are as follows: • No lighting. • Permissive left turn signal phasing (for signalized intersections only). • Right turn on red permitted on all approaches (for signalized intersections only). • U-turn permitted at major road approaches. • Absence of right turn channelization on major road approaches. • 6-lanes on the major road and 2-lanes on the minor road (excluding turn lanes) • No schools within 1000 feet. • No bus stops within 1000 feet. • No alcohol sales establishments within 1000 feet. For intersections on urban one-way arterials (UrbArtOW-3ST, UrbArtOW-4ST, and UrbArtOW-4SG) the base conditions, for which CMFs are available, are as follows: • No lighting. • 2-lanes on the major road and 2-lanes on the minor road For all way stop controlled intersections on rural two-lane roads and urban arterials (Rur4Leg-AWSC, Urb3Leg-AWSC, Urb4Leg-AWSC) the only base condition, for which a CMF is available is as follows: • No lighting. 2. SUMMARY OF DATA Tables 1 and 2 shows the number of intersections and the AADT ranges for the calibration dataset along with the AADT ranges used for base condition SPF estimation. For some of urban arterial intersections, minor road AADTs were not available. Only using urban arterial intersections with minor road AADTs would have diminished the sample used for calibration for M-7 

some intersection types. To include all intersections in the calibration procedure, the research team interpolated the missing minor road AADTs using the following steps: 1. Calculate the minimum, maximum, average, and the standard deviation of the available minor road AADTs as a percentage of major road AADT by intersection type. 2. Use Microsoft Excel’s uniform random number generator to randomly generate the missing minor road AADTs as percentage of major road AADT with upper and lower bounds defined as the average percentage of the available minor road AADTs ±1 standard deviation. The research team conduced a sensitivity analysis to determine the effect of minor road AADTs on the calibration factors and found variations in calibration factors to be in the ±2% range for the various randomly generated minor road AADT samples. Based on this minimal variation in the effects of minor road AADT on the calibration factors, the research team went ahead with including all urban arterial intersections with interpolated minor road AADTs in cases where minor road AADTs were missing. Table 1. Intersection Summary Data (Calibration Dataset) Major Road AADT Minor Road AADT Intersection Type Count Min Max Mean Min Max Mean Rur2L-3ST 208 390 21300 4540 24 5929 1176 Rur2L-3SG 15 2364 20071 10070 491 19214 4069 Rur2L-4ST 234 151 20310 3339 40 4004 919 Rur2L-4SG 105 834 23643 9043 808 15143 4414 RurML-3ST 14 2850 36286 13435 28 2450 414 RurML-3SG 9 10193 35000 18287 2057 6864 4642 RurML-4ST 21 3747 22429 11433 51 8105 1219 RurML-4SG 28 7554 46917 19210 139 21017 5430 UrbArt-3ST 72 72 44286 10835 19 9371 2620 UrbArt-3SG 9 12328 32071 20817 291 15643 7939 UrbArt-4ST 25 335 31714 12952 17 9500 2184 UrbArt-4SG 129 3443 47643 20596 17 29464 10098 UrbArt6+-3ST 219 14714 102714 33599 193 16745 8786 UrbArt6+-3SG 12 26143 76857 51190 4957 43500 19543 UrbArt6+-4ST 34 14714 57000 33660 241 19689 5896 UrbArt6+-4SG 151 14286 76857 34776 820 45440 14094 UrbArtOW-3ST 6 17500 94500 41429 632 19000 10722 UrbArtOW-4ST 5 14286 21071 17685 534 6954 3094 UrbArtOW-4SG 14 7229 24857 17725 385 16000 7515 Rur4Leg-AWSC 72 1100 9633 3524 435 4700 2154 Urb3Leg-AWSC 2 5800 14333 10067 4800 12667 8733 Urb4Leg-AWSC 3 2850 6350 4100 1330 2650 1843 Rur-RndAbt 20 2800 16750 6982 400 7850 3968 Urb-RndAbtSL 44 1300 18667 9018 310 15667 5185 Urb-RndAbtML 5 4767 12333 8265 2567 10350 6537 M-8 

Table 2. Summary of Volume Data (SPF Estimation Dataset) Major Road AADT Minor Road AADT Intersection Type Min Max Mean Min Max Mean Rur2L-3ST 308 20092 3033 4 3064 360 Rur2L-3SG 2900 23591 11334 100 2320 4568 Rur2L-4ST 147 8461 1843 4 4740 396 Rur2L-4SG 910 14790 5345 95 11641 2477 RurML-3ST 1325 36000 11651 3 5800 760 RurML-3SG 1001 56000 15928 101 27000 5040 RurML-4ST 2422 34500 10803 25 4654 589 RurML-4SG 880 12420 4986 157 7992 1902 UrbArt-3ST 270 38460 8187 33 18460 2137 UrbArt-3SG 3050 32109 12363 110 18415 4077 UrbArt-4ST 450 37301 8251 50 13773 2088 UrbArt-4SG 1810 34960 11067 72 27228 3803 UrbArt6+-3ST 10760 66800 37072 100 8589 1247 UrbArt6+-3SG 8755 94000 38791 98 31000 5455 UrbArt6+-4ST 12668 54600 33249 118 4600 1245 UrbArt6+-4SG 7090 137550 44658 86 68343 14188 UrbArtOW-3ST 98 42631 13234 17 13340 1081 UrbArtOW-4ST 345 23365 8718 75 19192 1413 UrbArtOW-4SG 130 98826 16993 103 77000 10538 Rur4Leg-AWSC 421 12983 3357 130 9985 1873 Urb3Leg-AWSC 175 20131 5491 151 11000 3411 Urb4Leg-AWSC 1150 12955 6008 400 11982 3503 Rur-RndAbt n/a n/a 8829 n/a n/a 4059 Urb-RndAbtSL n/a n/a 7687 n/a n/a 4035 Urb-RndAbtML n/a n/a 11712 n/a n/a 5760 Table 3 shows the sum of crashes by crash and severity type for intersections on rural highways (Rur2L- 3ST, Rur2L-3SG, Rur2L-4ST, Rur2L-4SG, RurML-3ST, RurMl-3SG, RurML-4ST, and RurML-4SG) in the calibration dataset. M-9 

Table 3. Summary of Crash Data for Intersections on Rural Highways Crash Crash Type Intersection Type Severity All SV SD OD ID KABCO 1116 308 405 120 251 Rur2L-3ST KABC 377 110 105 59 96 KAB 146 60 17 30 34 KA 26 10 4 6 6 Rur2L-3SG KABCO 200 n/a n/a n/a n/a KABCO 1851 255 313 152 1077 KABC 843 66 119 67 581 Rur2L-4ST KAB 369 28 39 32 266 KA 81 8 3 8 61 KABCO 2786 224 1202 524 740 KABC 850 61 282 208 294 Rur2L-4SG KAB 241 34 37 76 95 KA 39 5 4 11 19 KABCO 42 8 22 4 8 KABC 12 1 3 2 6 RurML-3ST KAB 3 0 0 1 2 KA 0 0 0 0 0 RurML-3SG KABCO 196 n/a n/a n/a n/a KABCO 185 27 17 15 117 KABC 96 5 2 11 76 RurML-4ST KAB 40 2 0 6 32 KA 7 0 0 1 6 KABCO 990 67 484 119 276 KABC 329 20 120 58 121 RurML-4ST KAB 99 8 15 27 45 KA 9 3 0 3 3 n/a = SPF was not estimated for this crash type and crash severity combination SV = Single vehicle crashes SD = Same direction crashes OD = Opposite direction crashes ID = Intersecting direction crashes Table 4 shows the sum of crashes by severity type for roundabouts (Rur-RndAbt, Urb-RndAbtSL, and Urb-RndAbtML) in the calibration dataset. Table 4. Summary of Crash for Roundabouts Crash Crash Severity Intersection Type Type Total F&I PDO Rur-RndAbt All 436 151 275 Urb-RndAbtSL All 884 154 708 Urb-RndAbtML All 84 18 64 Table 5 shows the sum of crashes by severity type for intersections on urban arterials (UrbArt-3ST, UrbArt-3SG, UrbArt-4ST, UrbArt-4SG, UrbArt6+-3ST, UrbArt6+-3SG, UrbArt6+-4ST, UrbArt6+-4SG, UrbArtOW-3ST, UrbArtOW-4ST, and UrbArtOW-4SG) in the calibration dataset. M-10 

Table 5. Summary of Crash Data for Intersections on Urban Arterials Crash Crash Type Intersection Type Severity All SV SD OD ID KABCO 710 91 304 72 219 UrbArt-3ST KABC 234 47 83 26 73 KAB 66 24 13 12 15 KA 7 3 0 1 3 KABCO 342 15 175 36 104 KABC 98 6 36 18 36 UrbArt-3SG KAB 20 4 4 2 9 KA 2 0 0 1 1 KABCO 383 30 125 38 174 KABC 146 13 36 19 76 UrbArt-4ST KAB 40 8 5 5 22 KA 1 1 0 0 0 KABCO 10245 535 5455 1134 2702 KABC 3053 240 1293 437 992 UrbArt-4SG KAB 624 117 130 120 240 KA 58 18 6 12 19 UrbArt6+-3ST KABCO 4067 n/a n/a n/a n/a UrbArt6+-3SG KABCO 1722 n/a n/a n/a n/a UrbArt6+-4ST KABCO 1265 n/a n/a n/a n/a UrbArt6+-4SG KABCO 18232 n/a n/a n/a n/a UrbArtOW-3ST KABCO 96 n/a n/a n/a n/a UrbArtOW-4ST KABCO 94 n/a n/a n/a n/a UrbArtOW-4SG KABCO 1415 n/a n/a n/a n/a n/a = SPF was not estimated for this crash type and crash severity combination SV = Single vehicle crashes SD = Same direction crashes OD = Opposite direction crashes ID = Intersecting direction crashes Table 6 shows the sum of crashes by severity type all way stop controlled intersections on rural two- lane roads and urban arterials (Rur4Leg-AWSC, Urb3Leg-AWSC, Urb4Leg-AWSC) in the calibration dataset. Table 6. Summary of Crash Data for All Way Stop Controlled Intersections Intersection Type Crash Type/Severity Sum of Crashes Rur4Leg-AWSC All KABCO 1125 Urb3Leg-AWSC All KABCO 37 Urb4Leg-AWSC All KABCO 53 M-11 

3. METHODOLOGY AND MODELS The methodology for calculating the calibration factors is documented in the HSM 1st edition. This equation is: 𝑛𝑛sites 𝑛𝑛𝑐𝑐𝑐𝑐 ∑𝑖𝑖=1 ∑𝑗𝑗=1 𝑁𝑁𝑜𝑜,𝑤𝑤(𝑖𝑖),𝑥𝑥(𝑖𝑖),𝑦𝑦,𝑧𝑧,𝑖𝑖,𝑗𝑗 𝐶𝐶𝑤𝑤,𝑥𝑥,𝑦𝑦,𝑧𝑧 = 𝑛𝑛sites 𝑛𝑛𝑐𝑐𝑐𝑐 ∑𝑖𝑖=1 ∑𝑗𝑗=1 𝑁𝑁𝑝𝑝𝑝𝑝,𝑤𝑤(𝑖𝑖),𝑥𝑥(𝑖𝑖),𝑦𝑦,𝑧𝑧,𝑖𝑖,𝑗𝑗 where Cw, x y, z = calibration factor to adjust SPF for local conditions for site type w, cross section or control type x, crash type y, and severity z; No, w(i), x(i), y, z, i, j = observed crash frequency for site i with site type w(i) and year j (includes cross section or control type x(i) for crash type y, and severity z) (crashes/yr); Npu, w(i), x(i), y, z, i = predicted average crash frequency unadjusted by calibration factor for site i with site type w(i) (includes cross section or control type x(i) for crash type y, and severity z) (crashes/yr); nsites = number of sites in the calibration database (site); and nca = number of years in the calibration period (yr). For intersections on rural highways (Rur2L-3ST, Rur2L-4ST, Rur2L-4SG, RurML-3ST, RurML-4ST, and RurML-4SG), one of the following two model forms developed by NCHRP Project 17-62 apply: 𝑁𝑁𝑠𝑠𝑠𝑠𝑠𝑠,𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟,17−62 = 𝑒𝑒𝑒𝑒𝑒𝑒�𝑏𝑏0 + 𝑏𝑏1 × ln (𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑚𝑚𝑚𝑚𝑚𝑚 ) + 𝑏𝑏2 × ln (𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑚𝑚𝑚𝑚𝑚𝑚 )� 𝑁𝑁𝑠𝑠𝑠𝑠𝑠𝑠,𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟,17−62 = 𝑒𝑒𝑒𝑒𝑒𝑒[𝑏𝑏0 + 𝑏𝑏3 × ln (𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 )] Tables 7 to 15 present the parameter estimates (and standard errors in parentheses) for the rural highway intersections base condition SPFs developed in NCHRP 17-62. For intersections on urban arterials (UrbArt-3ST, UrbArt-3SG, UrbArt-4ST, and UrbArt-4SG), one of the following two model forms developed by NCHRP Project 17-62 apply: Model A: 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑚𝑚𝑚𝑚𝑚𝑚 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑚𝑚𝑚𝑚𝑚𝑚 𝑎𝑎 𝑏𝑏 × � � 𝑑𝑑×� � 𝑁𝑁𝑠𝑠𝑠𝑠𝑠𝑠,𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢,17−62 = e × e 10000 × (𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑚𝑚𝑚𝑚𝑚𝑚 )𝑐𝑐 × e 10000 × (𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑚𝑚𝑚𝑚𝑚𝑚 )𝑒𝑒 Model B: 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑚𝑚𝑚𝑚𝑚𝑚 𝑏𝑏 × � 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑡𝑡𝑡𝑡𝑡𝑡 � 𝑑𝑑×� � 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑚𝑚𝑚𝑚𝑚𝑚 𝑒𝑒 𝑁𝑁𝑠𝑠𝑠𝑠𝑠𝑠,𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢,17−62 = e𝑎𝑎 × e 10000 × (𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑡𝑡𝑡𝑡𝑡𝑡 )𝑐𝑐 × e 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑡𝑡𝑡𝑡𝑡𝑡 ×� � 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑡𝑡𝑡𝑡𝑡𝑡 Tables 16 to 19 present the parameter estimates (and standard errors in parentheses) for the urban arterial intersections base condition SPFs developed in NCHRP 17-62. M-12 

For the crash types for which models could not be estimated, NCHRP 17-62 recommend using the prediction for the next closest model and multiplying the prediction by the proportion of that crash type. For example, if a prediction model for KA crashes is not available, but one for KAB crashes is, the prediction for KA crashes can be obtained by the following equation: 𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁 𝑜𝑜𝑜𝑜 𝐾𝐾𝐾𝐾 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐ℎ𝑒𝑒𝑒𝑒 𝑖𝑖𝑖𝑖 𝑡𝑡ℎ𝑒𝑒 𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑 𝑠𝑠𝑠𝑠𝑠𝑠 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 𝐾𝐾𝐾𝐾 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐ℎ𝑒𝑒𝑒𝑒 = 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 𝐾𝐾𝐾𝐾𝐾𝐾 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐ℎ𝑒𝑒𝑒𝑒 × � �. 𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁 𝑜𝑜𝑜𝑜 𝐾𝐾𝐾𝐾𝐵𝐵 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐ℎ𝑒𝑒𝑒𝑒 𝑖𝑖𝑖𝑖 𝑡𝑡ℎ𝑒𝑒 𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑 𝑠𝑠𝑠𝑠𝑠𝑠 Table 7. Base Condition SPFs, Rur2L-3ST Crash Type Severity b0 b1 b2 b3 k -7.924 0.656 0.295 0.622 KABCO - (0.857) (0.099) (0.067) (0.153) -9.628 0.725 0.312 0.974 KABC - (1.301) (0.144) (0.102) (0.340) Total -10.241 0.581 0.468 1.383 KAB - (1.891) (0.214) (0.153) (0.708) -11.873 0.908 5.123 KA - - (4.477) (0.548) (4.870) -15.506 1.291 0.452 1.777 KABCO - (2.178) (0.221) (0.141) (0.600) -18.598 1.569 0.420 2.775 KABC - (3.506) (0.337) (0.231) (1.370) Same direction -16.952 1.501 5.281 KAB - - (4.337) (0.506) (4.458) KA -13.794 0.984# 0.000* - - (3 crashes) (5.693) (0.665) (0.063) -14.120 0.818 0.753 0.995 KABCO - (2.365) (0.244) (0.191) (0.653) -15.174 0.977 0.583 1.583 KABC - Intersecting (3.164) (0.320) (0.261) (1.494) direction KAB -13.383 1.017 0.000* - - (6 crashes) (4.052) (0.472) (0.005) KA -10.629 0.556# 0.000* - - (2 crashes) (6.539) (0.795) (0.000) -11.716 0.746 0.455 0.000* KABCO - (1.581) (0.181) (0.133) (0.000) KABC -15.272 1.025 0.476 1.415 - (3.605) (0.406) (0.258) (1.720) Opposite direction KAB -15.571 0.371# 1.301 1.167 (9 crashes) (4.585) (0.523) (0.511) (1.626) KA -12.867 0.875# 0.000* - - (3 crashes) (5.580) (0.658) (0.050) -5.916 0.409 0.173 0.535 KABCO - (0.907) (0.112) (0.076) (0.203) -5.398 0.302 0.787 KABC - - (1.464) (0.183) (0.569) Single vehicle KAB -5.854 0.249# 1.309 - - (20 crashes) (2.264) (0.285) (1.486) KA -7.515# 0.168# 0.000* - - (2 crashes) (6.355) (0.802) (0.000) * Poisson distribution used; scale = square root of Deviance/DOF # Not significant at 90th percentile confidence interval M-13 

Table 8. Base Condition SPFs using Poisson Distribution, Rur2L-3ST Crash Type Severity b0 b1 b2 b3 Scale Same KA -13.794 0.984 0.412 - - direction (3 crashes) (2.346) (0.274) (0) KAB -13.383 1.017 0.487 - - Intersecting (6 crashes) (1.976) (0.230) (0) direction KA -10.629 0.556 0.344 - - (2 crashes) (2.255) (0.274) (0) -11.716 0.746 0.455 0.826 KABCO - Opposite (1.307) (0.150) (0.110) (0) direction KA -12.867 0.8752 0.416 - - (3 crashes) (2.323) (0.274) (0) KA -7.513 0.1681# 0.349 Single vehicle - - (2 crashes) (2.221) (0.280) (0) # Not significant at 90th percentile confidence interval M-14 

Table 9. Base Condition SPFs, Rur2L-4ST Crash Type Severity b0 b1 b2 b3 k -6.620 0.451 0.339 0.435 KABCO - (0.805) (0.112) (0.06) (0.119) -8.747 0.825 0.929 KABC - - (1.306) (0.168) (0.312) Total -8.511 0.723 1.564 KAB - - (1.676) (0.217) (0.630) -10.539 0.799 4.683 KA - - (3.235) (0.416) (3.524) -7.914 0.364 0.399 0.000* KABCO - (1.294) (0.184) (0.101) (0.001) -7.538 0.429# 0.000* KABC - - (2.469) (0.320) (0.000) Same direction KAB -4.284# -0.087# 0.000* - - (10 crashes) (3.337) (0.448) (0.002) KA - - - - - (0 crash) -10.362 0.475 0.722 0.415 KABCO - (1.320) (0.181) (0.103) (0.219) -12.896 1.248 2.906 KABC - - (2.284) (0.292) (1.094) Intersecting direction -12.779 1.175 2.178 KAB - - (2.425) (0.306) (1.183) -15.115 1.318 3.094 KA - - (4.079) (0.508) (3.627) -10.514 0.769 0.224 0.000* KABCO - (1.776) (0.242) (0.125) (0.000) -11.702 0.881 0.000* KABC - - (3.572) (0.450) (0.002) Opposite direction KAB -9.979# 0.506# 0.000* - - (3 crashes) (6.251) (0.806) (0.000) KA -32.191# -0.117# 849.22* - - (2 crashes) (4290) (57709) (0.000) -5.533 0.415 0.256 KABCO - - (1.044) (0.136) (0.203) -6.412 0.369# 0.439 KABC - - (1.838) (0.239) (0.714) Single vehicle KAB -6.874 0.355# 0.000* - - (21 crashes) (2.337) (0.304) (0.001) KA -2.405# -0.508# 0.000* - - (3 crashes) (6.079) (0.839) (0.000) *Poisson distribution used; scale = square root of Deviance/DOF # Not significant at 90th percentile confidence interval M-15 

Table 10. Base Condition SPFs using Poisson Distribution, Rur2L-4ST Crash Type Severity b0 b1 b2 b3 Scale -7.914 0.364 0.399 0.895 KABCO - (1.159) (0.165) (0.090) (0.000) -7.538 0.429# 0.667 KABC - - Same (1.647) (0.213) (0.000) direction KAB -4.284 -0.087# 0.551 - - (10 crashes) (1.841) (0.247) (0.002) KA -8.564 0.322# 0.357 - - (3 crashes) (2.203) (0.287) (0.000) -10.514 0.769 0.224 0.803 KABCO - (1.428) (0.194) (0.101) (0.000) Opposite -11.702 0.881 0.535 KABC - - direction (1.905) (0.240) (0.000) KAB -9.979 0.506 0.355 - - (3 crashes) (2.221) (0.286) (0.000) KAB -6.874 0.355 0.688 - - (21 crashes) (1.609) (0.209) (0.000) Single vehicle KA -2.406# -0.508 0.355 - - (3 crashes) (2.160) (0.298) (0.000) # Not significant at 90th percentile confidence interval M-16 

Table 11. Base Condition SPFs, Rur2L-4SG Crash Type Severity b0 b1 b2 b3 k -8.163 0.877 1.829 KABCO - - (1.692) (0.189) (0.299) -12.337 1.028 0.231 1.403 KABC - (2.393) (0.280) (0.132) (0.470) Total -11.059 0.981 1.376 KAB - - (2.828) (0.313) (0.630) KA -8.788 0.578# 2.349 - - (16 crashes) (4.898) (0.545) (2.592) -14.523 1.509 1.613 KABCO - - (2.184) (0.242) (0.340) -15.878 1.242 0.341 0.976 KABC - (3.219) (0.372) (0.176) (0.632) Same direction -14.740 1.269 0.831 KAB - - (4.419) (0.483) (1.221) KA 0.422# -0.623# 0.000* - - (4 crashes) (7.202) (0.833) (0.003) -5.767 0.480 2.358 KABCO - - (2.209) (0.248) (0.571) -11.026 0.675 0.341 1.731 KABC - Intersecting (3.282) (0.388) (0.199) (0.926) direction -10.318 0.813 1.679 KAB - - (3.738) (0.414) (1.238) KA -14.890 1.143# 0.000* - - (6 crashes) (7.891) (0.863) (0.000) KABCO -9.404 0.861 0.000* - - (15 crashes) (2.793) (0.528) (0.000) KABC -13.000# 0.916# 0.000* - - (5 crashes) (8.371) (0.921) (0.000) Opposite direction KAB -9.041# 0.452# 0.000* - - (4 crashes) (8.695) (0.969) (0.012) KA -7.825# 0.238# 0.000* - - (2 crashes) (11.851) (1.329) (0.000) KABCO -5.325 0.325# 2.029 - - (53 crashes) (2.701) (0.303) (0.909) KABC -11.854 0.876# 0.000* - - (11 crashes) (5.610) (0.618) (0.001) Single vehicle KAB -14.053 1.083# 0.000* - - (8 crashes) (6.778) (0.743) (0.002) KA -17.692 1.405# 0.000* - - (4 crashes) (9.993) (1.087) (0.008) * Poisson distribution used; scale = square root of Deviance/DOF # Not significant at 90th percentile confidence interval M-17 

Table 12. Base Condition SPFs using Poisson Distribution, Rur2L-4SG Crash Type Severity b0 b1 b2 b3 Scale Same KA 0.422# -0.623# 0.392 - - direction (4 crashes) (2.828) (0.327) (0.000) Intersecting KA -14.890 1.143 0.448 - - direction (6 crashes) (3.540) (0.387) (0.000) KABCO -11.404 0.861 0.613 - - (15 crashes) (2.938) (0.323) (0.000) KABC -13.000 0.916 0.423 - - Opposite (5 crashes) (3.547) (0.390) (0.000) direction KAB -9.041 0.452# 0.394 - - (4 crashes) (3.431) (0.382) (0.000) KA -7.825# 0.238# 0.303 - - (2 crashes) (3.597) (0.403) (0.000) KABC (11 -11.854 0.876 0.556 - - crashes) (3.120) (0.343) (0.001) KAB -14.053 1.083 0.496 Single vehicle - - (8 crashes) (3.366) (0.369) (0.002) KA -17.692 1.405 0.384 - - (4 crashes) (3.838) (0.417) (0.000) # Not significant at 90th percentile confidence interval M-18 

Table 13. Base Condition SPFs, RurML-3ST Crash Type Severity b0 b1 b2 b3 k -9.118 0.776 0.270 0.323 KABCO - (1.259) (0.123) (0.053) (0.100) -9.392 0.659 0.346 0.261# KABC - (1.632) (0.160) (0.073) (0.159) Total -9.208 0.546 0.357 0.367# KAB - (2.264) (0.221) (0.106) (0.346) KA 2.910# -0.741# 1.913# - - (10 crashes) (5.145) (0.573) (3.120) -14.411 1.033 0.502 0.236# KABCO - (2.166) (0.209) (0.094) (0.235) -12.552 0.737 0.504 0.539# KABC - (3.234) (0.315) (0.151) (0.698) Same direction KAB -8.279# 0.510# 3.000# - - (13 crashes) (5.067) (0.542) (3.15) KA -9.971# 0.491# 5.32×10-6# - - (2 crashes) (13.798) (1.461) (0.609) -12.652 0.746 0.651 0.602 KABCO - (2.242) (0.215) (0.110) (0.321) -14.356 0.728 0.833 0.435# KABC - Intersecting (2.789) (0.268) (0.150) (0.343) direction -13.058 0.575 0.774 0.365# KAB - (3.304) (0.320) (0.182) (0.540) KA 4.137# -0.934# 8.496# - - (6 crashes) (8.354) (0.928) (10.579) KABCO -6.978# 0.345# 9.03×10-7# - - (10 crashes) (8.480) (0.902) (0.029) KABC -6.946# 0.236# 8.69×10-7# - - Opposite (4 crashes) (9.189) (0.985) (0.032) direction KAB 3.758# -1.019# 1.43×10-4# - - (2 crashes) (10.847) (1.228) (10.729) KA 42.737# -5.782# 4.734# - - (1 crash) (68.643) (8.396) (19.475) -7.259 0.663 0.826 KABCO - - (1.796) (0.192) (0.251) -7.837 0.608 0.256# KABC - - (2.281) (0.242) (0.394) Single vehicle KAB -2.295# -0.097# 2.93×10-7# - - (18 crashes) (4.036) (0.441) (0.007) KA 1.116# -0.798# 0.002# - - (1 crash) (15.288) (1.714) (77.506) * Moore-Penrose inverse matrix used. # Not significant at 90th percentile confidence interval. M-19 

Table 14. Base Condition SPFs, RurML-4ST Crash Type Severity b0 b1 b2 b3 k -9.561 0.773 0.383 0.410 KABCO - (1.353) (0.140) (0.073) (0.099) -10.411 0.711 0.475 0.433 KABC - (1.814) (0.185) (0.102) (0.162) Total -8.843 0.441 0.509 0.683 KAB - (2.404) (0.249) (0.142) (0.323) KA -13.245 1.053# 5.857# - - (14 crashes) (6.064) (0.645) (4.216) -14.343 1.158 0.345 0.362 KABCO - (2.160) (0.218) (0.115) (0.210) -13.190 1.118 0.619# KABC - - (3.713) (0.391) (0.845) Same direction KAB -9.502 0.641# 0.877# - - (12 crashes) (5.235) (0.558) (1.989) KA 9.504# -1.733# 2.10×10-5# - - (1 crash) (23.757) (2.708) (3.983) -11.531 0.496 0.939 0.942 KABCO - (2.333) (0.234) (0.148) (0.271) -8.626 0.757 1.867 KABC - - Intersecting (2.967) (0.319) (0.578) direction -9.196 0.740 3.498 KAB - - (4.102) (0.441) (1.428) -10.886 0.770 11.215 KA - - (7.863) (0.843) (8.632) KABCO -6.868# 0.383# 0.191# - - (15 crashes) (4.439) (0.476) (1.253) KABC -6.939# 0.338# 8.55×10-7# - - (9 crashes) (6.289) (0.676) (0.023) Opposite 3.638 direction KAB -6.969# 0.276# - - ×10-6# (5 crashes) (8.465) (0.911) (0.125) KA -77.373# 7.178# 0.776# - - (1 crash) (114.01) (11.168) (23.934) -9.855 0.929 0.337 KABCO - - (1.848) (0.196) (0.188) -10.416 0.876 0.154# KABC - - (2.743) (0.290) (0.449) Single vehicle KAB -7.161 0.456# 0.092# - - (22 crashes) (3.664) (0.392) (0.827) KA -27.071 2.284# 2.78×10-6# - - (2 crashes) (15.661) (1.591) (0.108) * Moore-Penrose inverse matrix used. # Not significant at 90th percentile confidence interval. M-20 

Table 15. Base Condition SPFs, RurML-4SG Crash Type Severity b0 b1 b2 b3 k -7.741 0.932 0.443 KABCO - - (2.037) (0.232) (0.151) -14.318 1.442 0.775# KABC - - (4.593) (0.515) (0.567) Total -14.662 1.399 0.499# KAB - - (5.772) (0.644) (0.829) 4.932 KA -0.930* -0.318* - - ×10-6* (4 crashes) (14.969) (1.734) (0.676) -12.709 1.391 0.443 KABCO - - (2.805) (0.316) (0.213) -17.140 1.659 0.786# KABC - - (6.798) (0.756) (1.067) Same direction -14.669# 1.242# 5.026# KAB - - (11.669) (1.305) (8.002) KA -0.050* -0.592* 3.88×10-7* - - (1 crash) (.) (0.114) (.) -9.724 1.024 0.561 KABCO - - (3.021) (0.342) (0.268) -14.965 1.412 1.925# KABC - - (7.444) (0.837) (1.787) Intersecting direction -20.048 1.921 2.105# KAB - - (9.846) (1.095) (2.486) KA -0.088* -0.575* 2.15×10-7* - - (1 crash) (0.006) (0.107) (.) -9.904 0.965 0.999# KABCO - - (4.564) (0.518) (2.342) KABC 0.813* -0.519* 1.04×10-7* - - (4 crashes) (.) (0.060) (.) Opposite direction KAB -1.476* -0.288* 1.245×10-6* - - (3 crashes) (.) (.) (.) KA 0.778* -0.682* 5.146×10-6* - - (1 crash) (12.892) (1.522) (.) KABCO 1.557# -0.378# 0.573# - - (14 crashes) (3.382) (0.393) (0.550) KABC -4.075* -0.0269* 1.623×10-7* - - (2 crashes) (0.350) (0.0611) (.) Single vehicle KAB -5.352* 0.039* 4.470×10-7* - - (1 crash) (12.126) (1.393) (.) KA -5.352* 0.039* 4.470×10-7* - - (1 crash) (12.126) (1.393) (.) * Moore-Penrose inverse matrix used. # Not significant at 90th percentile confidence interval M-21 

Table 16. Base Condition SPFs, UrbArt-3ST Crash Model Severity a b c d e k Type Form -3.1275 0.6210 0.2319 0.7280 0.8087 All KABCO A - (0.8631) (0.1209) (0.1083) (0.1754) (0.0442) -2.3919 .7690 .7615 All KABC B - - - ( .0706) ( .0514) ( .0838) -2.7900 .6705 .8031 All KAB B - - - ( .0829) ( .0593) ( .1266) -4.0506 .5272 .8594 All KA B - - - ( .1482) ( .1044) ( .4459) 1.0576 0.3861 -0.3676 1.0986 SV KABCO B - - (1.3977) (0.1658) (0.1711) (0.1316) -0.0628 0.2730 -0.3596 1.4458 SV KABC B - - (2.1687) (0.2626) (0.2659) (0.3783) -1.8441 -0.1647 1.6069 SV KAB B - - - (1.1345) (0.1252) (0.5129) -2.0986 -0.2835 4.0221 SV KA B - - - (2.1063) (0.2334) (2.6334) -14.8383 1.4636 -.1385 1.1428 SD KABCO B - - ( .6258) ( .0689) ( .0539) ( .0803) -14.2585 1.3176 -1.1784 .9478 SD KABC B - - ( .9668) ( .0989) ( .4720) ( .1552) -14.1222 1.2298 -1.2920 1.3071 SD KAB B - - ( 1.3007) ( .1327) ( .6521) ( .3124) -11.5340 .7330 1.7962 SD KA B - - - ( 2.5011) ( .2685) ( 2.0336) -3.3353 0.6177 1.1523 OD KABCO B - - - ( .1100) ( .0794) ( .2364) -4.1549 .5514 .2950 OD KABC B - - - ( .1470) ( .1003) ( .3737) -4.4870 .5270 .6168 OD KAB B - - - ( .1778) ( .1231) ( .6314) OD KA Model did not converge -11.1651 0.8094 0.2535 2.2740 ID KABCO A - - (0.8035) (0.0849) (0.0719) (0.2200) -12.7177 0.8967 0.1974 3.3635 ID KABC A - - (1.2614) (0.1316) (0.1118) (0.6204) -14.4692 0.9112 0.3412 3.0787 ID KAB A - - (1.5707) (0.1608) (0.1432) (0.8626) -17.4865 1.0812 0.3558 7.9899 ID KA A - - (3.1341) (0.3051) (0.3044) (4.2722) SV = Single vehicle crashes SD = Same direction crashes OD = Opposite direction crashes ID = Intersecting direction crashes M-22 

Table 17. Base Condition SPFs, UrbArt-3SG Crash Model Severity a b c d e k Type Form -4.5704 .6366 .1519 .4669 All KABCO B - - ( 1.0173) ( .1051) ( .0618) ( .0432) -6.7956 .4799 .2585 .5344 All KABC A - - ( 1.3109) ( .1274) ( .0725) ( .0805) -8.0554 .5062 .2814 .5745 All KAB A - - ( 1.6897) ( .1660) ( .0916) ( .1336) All KA A Model did not converge -2.3447 .3894 .9168 .4113 SV KABCO A - - ( .2106) ( .1448) ( .1964) ( .1444) SV KABC Model did not converge SV KAB Model did not converge SV KA Model did not converge -6.2255 .5414 .2390 .4615 SD KABCO A - - ( 1.0384) ( .1034) ( .0575) ( .0486) -9.1985 0.6682 0.2495 0.3761 SD KABC A - - (1.5615) (0.1517) (0.0828) (0.1029) -9.3282 .5844 .2413 .4521 SD KAB A - - ( 2.2931) ( .2243) ( .1223) ( .2344) SD KA Model did not converge -10.0017 0.9248 0.7486 OD KABCO B - - - (1.9351) (0.1991) (0.1512) -17.9744 1.3504 0.3523 1.1826 OD KABC A - - (3.5381) (0.3371) (0.1741) (0.4533) -21.2395 1.4846 .5304 1.4144 OD KAB A - - ( 4.5175) ( .4257) ( .2182) ( .6955) OD KA Model did not converge -2.3636 0.2385 1.0859 ID KABCO B - - - (1.6005) (0.1658) (0.1221) -2.9487 0.1596 1.7788 ID KABC B - - - (2.3752) (0.2460) (0.3222) -4.1873 0.1997 2.2076 ID KAB B - - - (3.2704) (0.3385) (0.6438) -8.9578 0.5014 4.3772 ID KA B - - (7.2945) (0.7521) (4.1273) SV = Single vehicle crashes SD = Same direction crashes OD = Opposite direction crashes ID = Intersecting direction crashes M-23 

Table 18. Base Condition SPFs, UrbArt-4ST Crash Model Severity a b c d e k Type Form -3.6743 .4071 .9208 1.0155 All KABCO A - - ( .6256) ( .0726) ( .3490) ( .0867) -3.9675 .3417 1.6020 All KABC B - - - ( .9735) ( .1067) ( .1834) All KAB Could not obtain useful model All KA Could not obtain useful model -2.2170 .3435 .5835 SV KABCO B - - - ( .1253) ( .0889) ( .1913) -13.8618 -1.0741 1.3021 1.5358 SV KABC B - - (5.6055) (0.6352) (0.6814) (0.8553) -12.0750 -0.9275 1.0714 2.8132 SV KAB B - - (5.9194) (0.6816) (0.7211) (1.3122) -8.3241 0.4279 1.7361 SV KA B - - - (3.5390) (0.3835) (2.8313) -12.4690 1.0633 .2661 1.1504 SD KABCO A - - ( 1.0120) ( .1013) ( .0875) ( .1471) -5.9134 .8168 .4033 1.8464 SD KABC A - - ( .9920) ( .1394) ( .1322) ( .3910) -5.8261 .3579 .3360 1.8506 SD KAB A - - ( 1.3016) ( .1865) ( .1742) ( .7743) SD KA Model did not converge -6.0829 0.4417 1.4996 OD KABCO B - - - (1.2859) (0.1399) (0.3422) -5.5548 0.2814 2.3460 OD KABC B - - - (1.9029) (0.2078) (0.9251) -5.5578 0.2462 3.0857 OD KAB B - - - (2.1901) (0.2393) (1.3915) OD KA Model did not converge ID KABCO Could not obtain useful model ID KABC Could not obtain useful model ID KAB Could not obtain useful model ID KA Could not obtain useful model SV = Single vehicle crashes SD = Same direction crashes OD = Opposite direction crashes ID = Intersecting direction crashes M-24 

Table 19. Base Condition SPFs, UrbArt-4SG Crash Model Severity a b c d e k Type Form -7.4359 .9218 .5514 All KABCO B - - - ( .6531) ( .0685) ( .0381) -10.5443 1.0989 .6386 All KABC B - - - ( .8782) ( .0915) ( .0628) -9.9857 .9535 .7440 All KAB B - - - ( 1.0898) ( .1134) ( .1031) -9.6739 .7511 .6997 All KA B - - - ( 1.8404) ( .1907) ( .3296) -4.3216 .3000 .7818 SV KABCO B - - - ( 1.2070) ( .1264) ( .1586) -7.7339 0.5209 0.9105 SV KABC B - - - (2.0833) (0.2169) (0.4749) -8.9332 0.6011 0.8532 SV KAB B - - - (2.5001) (0.2597) (0.6746) SV KA Model did not converge -8.2447 0.9424 0.6264 0.5800 SD KABCO A - (0.6774) (0.0744) (0.1243) (0.0448) -14.2230 1.2127 0.2693 0.6257 SD KABC A - - (1.0584) (0.1114) (0.0657) (0.0848) -15.2404 1.2210 0.2476 0.8485 SD KAB A - - (1.6168) (0.1695) (0.1001) (0.2019) -14.3865 .7926 .4313 1.7976 SD KA A - - ( 3.1409) ( .3312) ( .2196) ( 1.1221) -9.7053 .7364 .2867 1.1587 OD KABCO A - - ( 1.0998) ( .1173) ( .0767) ( .1190) -13.5030 1.2228 1.8372 OD KABC B - - - (1.8439) (0.1914) (0.3049) -12.7760 1.0838 2.2166 OD KAB B - - - ( 2.1772) ( .2256) ( .5129) -12.8419 .9029 3.8585 OD KA B - - - ( 4.4636) ( .4616) ( 3.0410) -1.5214 .3492 .1316 .8944 ID KABCO A - ( .4578) ( .0863) ( .0594) ( .08100) -5.6212 .2977 .1958 1.2440 ID KABC A - - ( 1.1967) ( .1319) ( .0860) ( .1843) -6.1898 .4390 1.4297 ID KAB A - - - ( 1.6067) ( .1679) ( .3125) ID KA A Model did not converge SV = Single vehicle crashes SD = Same direction crashes OD = Opposite direction crashes ID = Intersecting direction crashes For intersections on urban arterials with 6+ lanes (UrbArt6+-3ST, UrbArt6+-3SG, UrbArt6+-4ST, and UrbArt6+-4SG), the model form developed by NCHRP Project 17-58 is: 𝑁𝑁𝑠𝑠𝑠𝑠𝑠𝑠,𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢6+𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖,17−62 = 𝑒𝑒𝑒𝑒𝑒𝑒�𝑏𝑏0 + 𝑏𝑏1 × ln (𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑚𝑚𝑚𝑚𝑚𝑚 ) + 𝑏𝑏2 × ln (𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑚𝑚𝑚𝑚𝑚𝑚 )� Tables 20 and 21 present the parameter estimates (and standard errors in parentheses) for the base condition SPFs developed in NCHRP 17-58 for intersections on urban arterials with 6+ lanes. M-25 

Table 20. Base Condition SPFs, F&I Crashes, UrbArt6+ Intersections Intersection Type b0 b1 b2 k -15.033 1.087 0.532 1.536 UrbArt6+-3ST (4.353) (0.416) (0.176) (0.586) -7.107 0.650 0.156 1.927 UrbArt6+-3SG (2.816) (0.259) (0.104) (0.612) -10.078 0.579 0.603 1.667 UrbArt6+-4ST (5.064) (0.480) (0.209) (0.652) -4.631 0.358 0.273 1.771 UrbArt6+-4SG (1.278) (0.114) (0.055) (0.166) Table 21. Base Condition SPFs, PDO Crashes, UrbArt6+ Intersections Intersection Type b0 b1 b2 k -14.973 1.349 0.153 1.342 UrbArt6+-3ST (4.113) (0.393) (0.156) (0.449) -5.073 0.472 0.135 1.004 UrbArt6+-3SG (3.662) (0.337) (0.119) (0.258) -12.011 0.672 0.747 0.879 UrbArt6+-4ST (5.666) (0.541) (0.268) (0.294) -3.772 0.268 0.271 1.009 UrbArt6+-4SG (1.591) (0.143) (0.064) (0.083) For intersections on urban one-way arterials (UrbArtOW-3ST, UrbArtOW-4ST, and UrbArtOW-4SG), the model form developed by NCHRP Project 17-58 is: 𝑁𝑁𝑠𝑠𝑠𝑠𝑠𝑠,𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢6+𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖,17−62 = 𝑒𝑒𝑒𝑒𝑒𝑒�𝑏𝑏0 + 𝑏𝑏1 × ln (𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑚𝑚𝑚𝑚𝑚𝑚 ) + 𝑏𝑏2 × ln (𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑚𝑚𝑚𝑚𝑚𝑚 )� Tables 22 and 23 present the parameter estimates (and standard errors in parentheses) for the base condition SPFs developed in NCHRP 17-58 for intersections on urban arterials with 6+ lanes. Table 22. Base Condition SPFs, F&I Crashes, UrbArtOW Intersections Intersection Type b0 b1 b2 k -9.117 0.646 0.105 0.495 UrbArt6+-3ST (3.808) (0.383) (0.192) (0.193) -10.829 0.672 0.414 1.881 UrbArt6+-4ST (1.927) (0.193) (0.108) (0.794) -5.468 0.184 0.372 0.751 UrbArt6+-4SG (1.098) (0.111) (0.089) (0.080) Table 23. Base Condition SPFs, PDO Crashes, UrbArtOW Intersections Intersection Type b0 b1 b2 k -17.602 1.531 0.306 0.966 UrbArt6+-3ST (2.588) (0.263) (0.105) (0.220) -12.064 0.855 0.512 1.039 UrbArt6+-4ST (1.611) (0.167) (0.101) (0.188) M-26 

-5.917 0.381 0.362 0.496 UrbArt6+-4SG (1.199) (0.124) (0.093) (0.045) For signalized intersections on rural highways (Rur2L-3G and RuralML-3SG), the model form developed by NCHRP Project 17-68 is: 𝑁𝑁𝑠𝑠𝑠𝑠𝑠𝑠,𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟,17−68 = 𝑒𝑒𝑒𝑒𝑒𝑒�𝑏𝑏0 + 𝑏𝑏1 × ln (𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑚𝑚𝑚𝑚𝑚𝑚 ) + 𝑏𝑏2 × ln (𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑚𝑚𝑚𝑚𝑚𝑚 )� Table 24 present the parameter estimates (and standard errors in parentheses) for the base condition SPFs developed in NCHRP 17-68 for 3-leg signalized intersections on rural highways. Table 24. Base Condition SPFs, Total Crashes, 3-Leg Rural Signalized Intersections Intersection Type b0 b1 b2 k -5.88 0.54 0.23 0.31 Rur2L-3SG (1.89) (0.18) (0.07) (0.06) -6.28 0.52 0.31 0.40 RurML-3SG (1.97) (0.21) (0.08) (0.08) For all way stop controlled intersections on rural two-lane roads and urban arterials (Rur4Leg-AWSC, Urb3Leg-AWSC, Urb4Leg-AWSC), one of the following two model forms developed by NCHRP Project 17-68 apply: 𝑁𝑁𝑠𝑠𝑠𝑠𝑠𝑠,𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎,17−68 = 𝑒𝑒𝑒𝑒𝑒𝑒�𝑏𝑏0 + 𝑏𝑏1 × ln (𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑚𝑚𝑚𝑚𝑚𝑚 ) + 𝑏𝑏2 × ln (𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑚𝑚𝑚𝑚𝑚𝑚 )� 𝑁𝑁𝑠𝑠𝑠𝑠𝑠𝑠,𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎,17−68 = 𝑒𝑒𝑒𝑒𝑒𝑒[𝑏𝑏0 + 𝑏𝑏3 × ln (𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 )] Table 25 present the parameter estimates (and standard errors in parentheses) for the base condition SPFs developed in NCHRP 17-68 for 3-leg signalized intersections on rural highways. Table 25. Base Condition SPFs, All Way Stop Controlled Intersections Crash Intersection Type b0 b1 b2 b3 k Severity -9.67 1.12 0.39 Rur4Leg-AWSC Total - - (1.09) (0.12) (0.07) -8.19 0.77 0.07 F&I - - (2.78) (0.31) (0.20) Urb3Leg-AWSC -7.94 0.85 0.37 PDO - - (2.40) (0.26) (0.19) -11.62 0.92 0.32 0.66 F&I - (1.88) (0.24) (0.17) (0.14) Urb4Leg-AWSC -8.58 0.64 0.36 0.78 PDO - (1.58) (0.20) (0.15) (0.12) M-27 

For rural and urban roundabouts (Rur-RndAbt, Urb-RndAbtSL, and Urb-RndAbtML) the model form developed by NCHRP Project 17-70 is: 𝑁𝑁𝑠𝑠𝑠𝑠𝑠𝑠,𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟,17−70 = 𝑒𝑒𝑒𝑒𝑒𝑒[𝑎𝑎 + 𝑑𝑑 × NumberLegs + 𝑒𝑒 × CircLanes] × 𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 𝑏𝑏 × 𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 𝑐𝑐 𝑁𝑁𝑠𝑠𝑠𝑠𝑠𝑠,𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢,17−70 = 𝑒𝑒𝑒𝑒𝑒𝑒[𝑎𝑎 + 𝑑𝑑 × NumberLegs] × 𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 𝑏𝑏 × 𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 𝑐𝑐 where NumberLegs = 1 if a 3-leg roundabout; 0 if 4-legs. CircLanes = 1 if a single-lane roundabout; 0 if more than 1 circulating lane. Table 26 present the parameter estimates (and standard errors in parentheses) for the base condition SPFs developed in NCHRP 17-70 for rural and urban roundabouts. Table 26. Base Condition SPFs. Roundabouts Crash Intersection Type a b c d e k Severity -5.3299 0.3356 0.5142 -0.6854 -0.9375 0.6292 Total (1.5425) (0.1767) (0.1165) (0.2244) (0.2309) (0.1005) -10.4848 0.7756 0.4239 -1.0080 -0.5506 0.4424 Rur-RndAbt F&I (2.2682) (0.2356) (0.1312) (0.2827) (0.2658) (0.1288) -5.4115 0.2980 0.5463 -0.7104 -1.0192 0.7284 PDO (1.6602) (0.1939) (0.1280) (0.2430) (0.2488) (0.1192) -5.6049 0.3274 0.3960 -0.8681 0.5030 Total - (1.0533) (0.1479) (0.1157) (0.1489) (0.0727) -8.6597 0.5271 0.3505 -0.7317 0.3290 Urb-RndAbtSL F&I - (1.3337) (0.1886) (0.1374) (0.1822) (0.0908) -5.5319 0.2653 0.4294 -0.9260 0.6064 PDO - (1.1659) (0.1650) (0.1304) (0.1656) (0.0878) -5.6642 0.5210 0.2905 -0.4610 0.9263 Total - (1.2790) (0.1545) (0.1198) (0.2357) (0.1371) -10.3369 0.9134 0.1937 -0.5131 0.5611 Urb-RndAbtML F&I - (1.7505) (0.2129) (0.1248) (0.2261) (0.1398) -5.7669 0.4954 0.3098 -0.4618 1.0642 PDO - (1.3664) (0.1640) (0.1274) (0.2527) (0.1583) 4. CALIBRATION RESULTS The results of the calibration based the intersection types and the various crash types, and crash severities are shown in Tables 27 to 41. For each site type, the number of intersections as well as the sum of observed and uncalibrated predicted crashes are provided as well as the calibration factor. Generally, if the number of crashes was less than 200 or the number of intersections were less than 30, it was judged insufficient or tagged as “use with caution”. M-28 

Table 27. Calibration Results and Recommended Calibration Factors, Rural 3-Leg Signalized Intersections Intersection No. of Observed Total Predicted Total Calibration Notes Type Intersections Crashes Crashes Factors Use with Rur2L-3SG# 15 200 219.05 0.91 Caution Use with RurML-3SG# 9 196 198.87 0.99 Caution # Use with caution due to low number of intersections available for calibration Table 28. Calibration Results and Recommended Calibration Factors, Rur2L-3ST (n=208) Crash Crash Severity Observed Crashes Predicted Crashes Calibration Factor Notes Type KABCO 1116 981.31 1.14 KABC 377 360.04 1.05 All KAB* - - - Use All KABC factor KA* - - - Use All KABC factor KABCO 308 383.52 0.80 KABC* - - - Use SV KABCO factor SV KAB* - - - Use SV KABCO factor KA* - - - Use SV KABCO factor KABCO 405 389.14 1.04 KABC* - - - Use SD KABCO factor SD KAB* - - - Use SD KABCO factor KA* - - - Use SD KABCO factor KABCO 120 153.25 0.78 KABC* - - - Use OD KABCO factor OD KAB* - - - Use OD KABCO factor KA* - - - Use OD KABCO factor KABCO 251 225.13 1.11 KABC* - - - Use ID KABCO factor ID KAB* - - - Use ID KABCO factor KA* - - - Use ID KABCO factor * The number of crashes are low leading to unreliable calibration factors. Use the calibration factors recommended under notes SV = Single vehicle crashes SD = Same direction crashes OD = Opposite direction crashes ID = Intersecting direction crashes M-29 

Table 29. Calibration Results and Recommended Calibration Factors, Rur2L-4ST (n=234) Crash Crash Severity Observed Crashes Predicted Crashes Calibration Factor Notes Type KABCO 1851 820.44 2.26 KABC 843 261.71 3.22 All KAB 369 139.74 2.64 KA* - - - Use All KAB factor KABCO 255 209.55 1.22 KABC* - - - Use SV KABCO factor SV KAB* - - - Use SV KABCO factor KA* - - - Use SV KABCO factor KABCO 313 165.14 1.90 KABC* - - - Use SD KABCO factor SD KAB* - - - Use SD KABCO factor KA* - - - Use SD KABCO factor KABCO 152 103.21 1.47 KABC* - - - Use OD KABCO factor OD KAB* - - - Use OD KABCO factor KA* - - - Use OD KABCO factor KABCO 1077 338.22 3.18 KABC* 581 156.28 3.72 ID KAB* 266 93.50 2.84 KA* - - - Use ID KA Factor * The number of crashes are low leading to unreliable calibration factors. Use the calibration factors recommended under notes SV = Single vehicle crashes SD = Same direction crashes OD = Opposite direction crashes ID = Intersecting direction crashes M-30 

Table 30. Calibration Results and Recommended Calibration Factors, Rur2L-4SG (n=105) Crash Crash Severity Observed Crashes Predicted Crashes Calibration Factor Notes Type KABCO 2786 727.23 3.83 KABC 850 230.01 3.70 All KAB 241 113.39 2.13 KA* - - - Use All KAB factor KABCO 224 65.22 3.43 KABC* - - - Use SV KABCO factor SV KAB* - - - Use SV KABCO factor KA* - - - Use SV KABCO factor KABCO 1202 504.39 2.38 KABC 282 105.51 2.67 SD KAB* - - - Use SD KABC factor KA* - - - Use SD KABC factor KABCO 524 180.39 2.90 KABC* - - - Use OD KABCO factor OD KAB* - - - Use OD KABCO factor KA* - - - Use OD KABCO factor KABCO 740 181.79 4.07 KABC 294 83.28 3.53 ID KAB* - - - Use SD KABC factor KA* - - - Use SD KABC factor * The number of crashes are low leading to unreliable calibration factors. Use the calibration factors recommended under notes SV = Single vehicle crashes SD = Same direction crashes OD = Opposite direction crashes ID = Intersecting direction crashes M-31 

Table 31. Calibration Results and Recommended Calibration Factors, RurML-3ST (n=14) Crash Crash Severity Observed Crashes Predicted Crashes Calibration Factor Notes Type KABCO# 42 43.09 0.97 Use with Caution Use All KABCO KABC* - - - factor All Use All KABCO KAB* - - - factor Use All KABCO KA* - - - factor Use All KABCO KABCO* - - - factor Use All KABCO KABC* - - - factor SV Use All KABCO KAB* - - - factor Use All KABCO KA* - - - factor Use All KABCO KABCO* - - - factor Use All KABCO KABC* - - - factor SD Use All KABCO KAB* - - - factor Use All KABCO KA* - - - factor Use All KABCO KABCO* - - - factor Use All KABCO KABC* - - - factor OD Use All KABCO KAB* - - - factor Use All KABCO KA* - - - factor Use All KABCO KABCO* - - - factor Use All KABCO KABC* - - - factor ID Use All KABCO KAB* - - - factor Use All KABCO KA* - - - factor #Use with caution due to low number of intersections and low number of crashes available for calibration * The number of crashes are low leading to unreliable calibration factors. Use the calibration factors recommended under notes SV = Single vehicle crashes SD = Same direction crashes OD = Opposite direction crashes ID = Intersecting direction crashes M-32 

Table 32. Calibration Results and Recommended Calibration Factors, RurML-4ST (n=21) Crash Crash Severity Observed Crashes Predicted Crashes Calibration Factor Notes Type KABCO# 185 153.02 1.21 Use with Caution Use All KABCO KABC* - - - factor All Use All KABCO KAB* - - - factor Use All KABCO KA* - - - factor Use All KABCO KABCO* - - - factor Use All KABCO KABC* - - - factor SV Use All KABCO KAB* - - - factor Use All KABCO KA* - - - factor Use All KABCO KABCO* - - - factor Use All KABCO KABC* - - - factor SD Use All KABCO KAB* - - - factor Use All KABCO KA* - - - factor Use All KABCO KABCO* - - - factor Use All KABCO KABC* - - - factor OD Use All KABCO KAB* - - - factor Use All KABCO KA* - - - factor Use All KABCO KABCO* - - - factor Use All KABCO KABC* - - - factor ID Use All KABCO KAB* - - - factor Use All KABCO KA* - - - factor #Use with caution due to low number of intersections available for calibration * The number of crashes are low leading to unreliable calibration factors. Use the calibration factors recommended under notes SV = Single vehicle crashes SD = Same direction crashes OD = Opposite direction crashes ID = Intersecting direction crashes M-33 

Table 33. Calibration Results and Recommended Calibration Factors, RurML-4ST (n=28) Crash Crash Severity Observed Crashes Predicted Crashes Calibration Factor Notes Type KABCO# 990 850.21 1.16 Use with Caution KABC# 329 233.08 1.41 Use with Caution All KAB* - - - Use All KABC factor KA* - - - Use All KABC factor Use All KABCO KABCO* - - - factor KABC* - - - Use All KABC factor SV KAB* - - - Use All KABC factor KA* - - - Use All KABC factor KABCO# 484 599.10 0.81 Use with Caution Use SD KABCO KABC* - - - factor SD Use SD KABCO KAB* - - - factor Use SD KABCO KA* - - - factor Use All KABCO KABCO* - - - factor KABC* - - - Use All KABC factor OD KAB* - - - Use All KABC factor KA* - - - Use All KABC factor KABCO# 276 300.57 0.92 Use with Caution KABC* - - - Use ID KABCO factor ID KAB* - - - Use ID KABCO factor KA* - - - Use ID KABCO factor # Use with caution due to low number of intersections available for calibration * The number of crashes are low leading to unreliable calibration factors. Use the calibration factors recommended under notes SV = Single vehicle crashes SD = Same direction crashes OD = Opposite direction crashes ID = Intersecting direction crashes M-34 

Table 34. Calibration Results and Recommended Calibration Factors, UrbArt-3ST (n=72) Crash Crash Severity Observed Crashes Predicted Crashes Calibration Factor Notes Type KABCO 710 518.63 1.37 KABC 234 145.42 1.61 All KAB - - - Use All KABC factor KA - - - Use All KABC factor Use All KABCO KABCO* - - - factor KABC* - - - Use All KABC factor SV KAB* - - - Use All KABC factor KA* - - - Use All KABC factor KABCO 304 289.17 1.05 Use SD KABCO KABC* - - - factor SD Use SD KABCO KAB* - - - factor Use SD KABCO KA* - - - factor Use All KABCO KABCO* - - - factor KABC* - - - Use All KABC factor OD KAB* - - - Use All KABC factor KA* - - - Use All KABC factor KABCO 219 76.36 2.87 KABC* - - - Use ID KABCO factor ID KAB* - - - Use ID KABCO factor KA* - - - Use ID KABCO factor * The number of crashes are low leading to unreliable calibration factors. Use the calibration factors recommended under notes SV = Single vehicle crashes SD = Same direction crashes OD = Opposite direction crashes ID = Intersecting direction crashes M-35 

Table 35. Calibration Results and Recommended Calibration Factors, UrbArt-3SG (n=9) Crash Crash Severity Observed Crashes Predicted Crashes Calibration Factor Notes Type KABCO# 342 326.61 1.05 Use with Caution Use All KABCO KABC* - - - factor All Use All KABCO KAB* - - - factor Use All KABCO KA* - - - factor Use All KABCO KABCO* - - - factor Use All KABCO KABC* - - - factor SV Use All KABCO KAB* - - - factor Use All KABCO KA* - - - factor Use All KABCO KABCO* - - - factor Use All KABCO KABC* - - - factor SD Use All KABCO KAB* - - - factor Use All KABCO KA* - - - factor Use All KABCO KABCO* - - - factor Use All KABCO KABC* - - - factor OD Use All KABCO KAB* - - - factor Use All KABCO KA* - - - factor Use All KABCO KABCO* - - - factor Use All KABCO KABC* - - - factor ID Use All KABCO KAB* - - - factor Use All KABCO KA* - - - factor #Use with caution due to low number of intersections available for calibration * The number of crashes are low leading to unreliable calibration factors. Use the calibration factors recommended under notes SV = Single vehicle crashes SD = Same direction crashes OD = Opposite direction crashes ID = Intersecting direction crashes M-36 

Table 36. Calibration Results and Recommended Calibration Factors, UrbArt-4ST (n=25) Crash Crash Severity Observed Crashes Predicted Crashes Calibration Factor Notes Type KABCO# 383 239.41 1.60 Use with Caution Use All KABCO KABC* - - - factor All Use All KABCO KAB* - - - factor Use All KABCO KA* - - - factor Use All KABCO KABCO* - - - factor Use All KABCO KABC* - - - factor SV Use All KABCO KAB* - - - factor Use All KABCO KA* - - - factor Use All KABCO KABCO* - - - factor Use All KABCO KABC* - - - factor SD Use All KABCO KAB* - - - factor Use All KABCO KA* - - - factor Use All KABCO KABCO* - - - factor Use All KABCO KABC* - - - factor OD Use All KABCO KAB* - - - factor Use All KABCO KA* - - - factor Use All KABCO KABCO* - - - factor Use All KABCO KABC* - - - factor ID Use All KABCO KAB* - - - factor Use All KABCO KA* - - - factor #Use with caution due to low number of intersections available for calibration * The number of crashes are low leading to unreliable calibration factors. Use the calibration factors recommended under notes SV = Single vehicle crashes SD = Same direction crashes OD = Opposite direction crashes ID = Intersecting direction crashes M-37 

Table 37. Calibration Results and Recommended Calibration Factors, UrbArt-4SG (n=129) Crash Type Crash Severity Observed Crashes Predicted Crashes Calibration Factor Notes KABCO 10245 6237.96 1.64 KABC 3053 1707.87 1.79 All KAB 624 655.36 0.95 KA* - - - Use All KAB factor KABCO 535 203.27 2.63 KABC 240 63.05 3.81 SV KAB* - - - Use SV KABC factor KA* - - - Use SV KABC factor KABCO 5455 5591.60 0.98 KABC 1293 1065.29 1.21 SD KAB* - - - Use SD KABC factor KA* - - - Use SD KABC factor KABCO 1134 975.75 1.16 KABC 437 327.27 1.34 OD Use OD KABC KAB* - - - factor Use OD KABC KA* - - - factor KABCO 2702 1243.74 2.17 KABC 992 302.28 3.28 ID KAB 240 119.05 2.02 KA* - - - Use ID KAB factor * The number of crashes are low leading to unreliable calibration factors. Use the calibration factors recommended under notes SV = Single vehicle crashes SD = Same direction crashes OD = Opposite direction crashes ID = Intersecting direction crashes M-38 

Table 38. Calibration Results and Recommended Calibration Factors, UrbArt6+ Intersections Intersection No. of Observed Total Predicted Total Calibration Notes Type Intersections Crashes Crashes Factors UrbArt6+-3ST 219 4067 7522.35 0.54 Use with UrbArt6+-3SG# 12 1722 661.83 2.60 Caution UrbArt6+-4ST 34 1265 1817.21 0.70 UrbArt6+-4SG 151 18232 10527.70 1.73 # Use with caution due to low number of intersections available for calibration Table 39. Calibration Results and Recommended Calibration Factors, UrbArtOW Intersections Intersection No. of Observed Total Predicted Total Calibration Notes Type Intersections Crashes Crashes Factors UrbArtOW-3ST^ 6 - - - UrbArtOW-4ST^ 5 - - - Use with UrbArtOW-4SG# 14 1415 252.64 5.60 Caution ^ Cannot calculate reliable calibration factor due to low number of intersections and cashes available for calibration # Use with caution due to low number of intersections available for calibration Table 40. Calibration Results and Recommended Calibration Factors, All Way Stop Controlled Intersections Intersection No. of Observed Total Predicted Total Calibration Notes Type Intersections Crashes Crashes Factors Urb4Leg-AWSC^ 3 - - - Urb3Leg-AWSC^ 2 - - - Rur4Leg-AWSC 72 1125 527.99 2.13 ^ Cannot calculate reliable calibration factor due to low number of intersections and cashes available for calibration Table 41. Calibration Results and Recommended Calibration Factors, Roundabouts Intersection No. of Crash Crash Observed Predicted Calibration Notes Type Intersections Type Severity Crashes Crashes Factor Use with KABCO# 436 293.59 1.49 Caution Use All Rur-RndAbt 20 All KABC* - - - KABCO factor Use with O# 275 232.23 1.18 Caution KABCO 884 487.56 1.81 Urb- 44 All Use All KABC* - - - RndAbtSL KABCO factor O 708 390.95 1.81 KABCO^ - - - Urb- 5 All KABC^ - - - RndAbtML O^ - - - ^ Cannot calculate reliable calibration factor due to low number of intersections and cashes available for calibration * The number of crashes are low leading to unreliable calibration factors. Use the calibration factors recommended under notes # Use with caution due to low number of intersections available for calibration M-39 

5. VALIDATION OF ADJUSTMENT FACTORS This section is about the validation of SPF adjustment factors. The main purpose of this validation exercise was to validate SPF adjustment factors developed by this project (NCHRP 17-72) for the models estimated in NCHRP 17-62. The validation of adjustment factors focused on the validation of each adjustment factor individually. The validation is a cross-sectional comparison, so the optimum approach is to compare two sets of data where the only difference is in the variable of interest. The approach taken was to use sites that met all base conditions with the exception of the variable to which the adjustment factor applies. For example, where the adjustment factor for lighting is of interest, all the sites used meet the base conditions except for lighting where both sites with and without lighting are used. Three analysis approaches were explored depending on the nature of the adjustment factors. Approach 1 The first approach is to compare the observed to predicted number of crashes using the base condition model at each level of the variable, e.g., sites with lighting and those without. If that variable has an effect on safety, then the ratio of observed to predicted crashes will differ between the levels of the variable. The factor for each level, i, is the sum of observed crashes divided by the sum of predicted crashes as shown in the following equation: ∑𝑎𝑎𝑎𝑎𝑎𝑎 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐ℎ𝑒𝑒𝑒𝑒 𝐶𝐶𝑖𝑖 = ∑𝑎𝑎𝑎𝑎𝑎𝑎 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐ℎ𝑒𝑒𝑒𝑒 To do so, the following steps were undertaken: Step 1: Apply the base condition models to a dataset of sites meeting all base conditions with the exception of the variables of interest. For this variable site selection is not constrained. Step 2: Calculate the factor Ci for each level of the variable. Step 3: Adjustment factors can then be estimated for each level of the variable. Each factor is related to the base condition level by dividing the factor for level i by the factor for the base level. To illustrate, in the table below a variable has 3 levels and a factor has been estimated for each, ranging from 1.0 to 1.2. Level 1 is the base condition for the SPF. The adjustment factors for the other levels are estimated by dividing each factor Ci by the Ci for level 1. Note that in this example the Ci for level 1 is not 1.0. This may occur if the sites used did not exactly match those used for calibrating the base condition models or due to rounding the numbers of predicted crashes. Table 42. Illustration of Approach 1 Adjustment Level of Variable Ci Factors 1 0.98 1.00 2 1.20 1.22 3 1.50 1.53 M-40 

These adjustment factors were then compared to the recommended adjustment factors. Approach 2 The second approach made use of generalized linear regression modeling (GLM). In this approach the expected number of a crashes is modeled with the base condition model prediction as an offset. The variable of interest is then included in the model to estimate the adjustment factor. The equations below illustrate this approach. 𝐶𝐶𝐶𝐶𝐶𝐶𝑠𝑠ℎ𝑒𝑒𝑒𝑒 = (𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖)(𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦)(𝑆𝑆𝑆𝑆𝑆𝑆 )𝑓𝑓(𝑉𝑉𝑉𝑉𝑉𝑉) ln(𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶ℎ𝑒𝑒𝑒𝑒) − ln[(𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦)(𝑆𝑆𝑆𝑆𝑆𝑆 )] = ln(𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖) + ln (𝑓𝑓 (𝑉𝑉𝑉𝑉𝑉𝑉)) where • 𝑆𝑆𝑆𝑆𝑆𝑆 = base model prediction, • 𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖 = a constant term, and • 𝑓𝑓(𝑉𝑉𝑉𝑉𝑉𝑉) = function representing the relationship between crashes and the variable of interest Approach 3 For some variables, the recommended adjustment factors are not a single value or simple equation. In these cases, the approach to validation was to compare the sum of observed and predicted values for the base model and base model plus adjustment factor when applied to sites that did not meet the base conditions for the variable of interest. 5.1 Rural Two-Lane, 3-Leg Stop Controlled Intersections (Rur2L-3ST) Lighting The base condition is the absence of lighting. Adjustment factors were recommended for “all crash types combined” and are computed using the following equations: 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 + �1 − 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 � 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑃𝑃𝑃𝑃𝑃𝑃 = 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝑃𝑃𝑃𝑃𝑃𝑃 + �1 − 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 � and 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾|𝑎𝑎𝑎𝑎 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 + �1 − 𝑝𝑝𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾|𝑎𝑎𝑎𝑎 � × 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑃𝑃𝑃𝑃𝑃𝑃 where 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for all crash types and KABC severity, 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑃𝑃𝑃𝑃𝑃𝑃 = CMF for all crash types and PDO severity, 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for all crash types and all severity categories, 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 = proportion of all crashes that occur at night (0.260 is default), 𝑝𝑝𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾|𝑎𝑎𝑎𝑎 = proportion of all crashes that have KABC severity, 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for effect of lighting presence on nighttime crashes with KABC severity (0.80 is default); and 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝑃𝑃𝑃𝑃𝑃𝑃 = CMF for effect of lighting presence on nighttime crashes with PDO severity (0.92 is default). M-41 

Table 43. Recommended Rur2L-3ST Lighting Adjustment Factors Crash Type Crash AF Severity All KABCO Use equation above KABC Use equation above KAB Same as KABC KA Same as KABC The recommended adjustment factors could not be validated since the dataset used for calibration did not have any variation for lighting, i.e., all sites in the calibration dataset met the base condition of absence of lighting. Left Turn Bay The base condition is the absence of left turn bay. The recommended adjustment factors are shown in Table 44. Table 44. Recommended Rur2L-3ST Left Turn Bay Adjustment Factors Crash Crash Type AF Severity All KABCO 0.56 KABC 0.45 KAB 0.45 KA 0.45 Single vehicle KABCO 0.61 Same direction KABCO 0.45 Opposing direction KABCO 0.69 Intersecting direction KABCO 0.39 Approaches 1 and 2 were both applied. Table 45 shows the results of the Approach 1 analysis. There were 61 sites which met all base conditions (i.e., base conditions of all other variables) except for the left turn bay. Amongst these 61 sites, 54 sites had no left turn bay present, whereas 7 sites had left turn bay present. M-42 

Table 45. Rur2L-3ST Left Turn Bay Approach 1 Calibration Observed Crashes Predicted Crashes Factors Crash Implied Crash Type LTB LTB LTB LTB LTB LTB Severity AF Present Not Present Not Present Not Present Present Present All KABCO 50 302 49.21 249.34 1.02 1.21 0.84 KABC 22 103 18.76 93.03 1.17 1.11 1.06 KAB 7 48 9.1 44.10 0.77 1.09 0.71 KA 2 5 1.26 6.79 1.59 0.74 2.16 Single vehicle KABCO 18 82 16.8 96.11 1.07 0.85 1.26 Same KABCO 12 121 23.24 106.26 0.52 1.14 0.45 direction Opposing KABCO 3 27 1.19 38.57 2.52 0.70 3.60 direction Intersecting KABCO 16 67 13.65 59.64 1.17 1.12 1.04 direction There were no crash categories with at least 100 observed crashes (sites where left turn was present). The only crash category with at least 50 observed crashes (sites where left turn bay was present) was total crashes (i.e., All KABCO) and the implied adjustment factor of presence of left turn bay is 0.84. This does not agree with the recommended adjustment factor of 0.56, although it may not be statistically different due to the small estimation sample. Using approach 2, negative binomial models were calibrated for All KABCO crashes. Other crash types had very few crashes in total so were not analyzed. The model form is: 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶ℎ𝑒𝑒𝑒𝑒 = (𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖)(𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦)(𝑆𝑆𝑆𝑆𝑆𝑆 )𝑒𝑒𝑒𝑒𝑒𝑒𝛽𝛽∗𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 𝐵𝐵𝐵𝐵𝐵𝐵 where 𝑆𝑆𝑆𝑆𝑆𝑆 = the calibrated base condition SPF 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 𝐵𝐵𝐵𝐵𝐵𝐵 = 1 if left turn bay is present, 0 if not present 𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂 = the overdispersion parameter of the negative binomial model Table 46 provides the parameter estimates and the implied adjustment factor for All KABCO crashes. Table 46. Rur2L-3ST Left Turn Bay Approach 2, estimate (standard error) Left Turn Crash Type Crash Severity Intercept Overdispersion Implied AF Bay Present 0.0283 0.0910 All KABCO 0.1154 1.096 (0.1419) (0.4043) M-43 

In contrast to Approach 1 the adjustment factor is greater than 1 for KABCO crashes. As evident from Table 46, the statistical significance of the parameters is very poor, which would translate into a high variance for the estimated AF. Right Turn Bay The base condition is the absence of right turn bay. The recommended adjustment factors are shown in Table 47. Table 47. Recommended Rur2L-3ST Right Turn Bay Adjustment Factors Crash Crash Type AF Severity All KABCO 0.85 KABC 0.77 KAB 0.77 KA 0.77 Single vehicle KABCO 0.82 Same direction KABCO 0.79 There were 55 sites which met all base conditions (i.e., base conditions of all other variables) except for the right turn bay. Amongst these 55 sites, 54 sites had no right turn bay present, whereas only 1 site had right turn bay present. Even though a 1 site sample would not yield any conclusive results, the research team still applied Approach 1 for validation. Table 48 shows the results of the Approach 1 analysis. There were no crash categories with at least 50 observed crashes (sites where right turn bay was present). For total crashes (i.e., All KABCO), the implied adjustment factor of presence of right turn bay is 1.00. This is not very far from the recommended adjustment factor of 0.85, and it may also not be statistically different due to the small estimation sample. Table 48. Rur2L-3ST Right Turn Bay Approach 1 Calibration Observed Crashes Predicted Crashes Factors Crash Implied Crash Type RTB RTB RTB RTB RTB RTB Severity AF Present Not Present Not Present Not Present Present Present All KABCO 7 302 5.76 249.34 1.22 1.21 1.00 KABC 1 103 2.12 93.03 0.47 1.11 0.43 KAB 0 48 1.13 44.10 n/a 1.09 n/a KA 0 5 0.14 6.79 n/a 0.74 n/a Single vehicle KABCO 4 82 2.15 96.11 1.86 0.85 2.18 Same KABCO 2 121 1.96 106.26 1.02 1.14 0.90 direction M-44 

Based on the results from Approach 1, there is some evidence that the magnitude of the adjustment factor is reasonable and as such it may be cautiously considered that the adjustment factor for total crashes (i.e., All KABCO) for installation of right turn bay is validated. Intersection Skew The base condition is no skew (i.e., an intersection angel of 90 degrees). The recommended adjustment factors are shown in Table 49. Table 49. Recommended Rur2L-3ST Skew Angle Adjustment Factors Crash Crash Type AF Severity All KABCO 𝑒𝑒𝑒𝑒𝑒𝑒[0.0040 𝑥𝑥 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠] There were 193 sites which met all base conditions (i.e., base conditions of all other variables) except for the skew angle. Amongst these 193 sites, 54 sites had no skew angle, whereas 139 sites had a skew angle. Because the adjustment factor is not a simple factor for one or more levels, Approach 1 and Approach 2 are not capable of making a straightforward validation. Approach 3 is applied which compares the predictive performance of the base model with and without using the adjustment factor applied to sites with a skew angle. Table 50 shows the number of sites, observed, and predicted crashes for All KABCO crashes. Table 50. Rur2L-3ST Skew Angle Approach 3 No. of Observed Predicted Model Sites (All KABCO Crashes) (All KABCO crashes) Base Model 139 729 616.42 Base Model + AF 139 729 665.56 As expected, the application of the base models alone underpredicts the number of crashes at sites with a skew angle. When the adjustment factor is applied the predictions still underpredict crashes, but the prediction gets closer to the observed crashes. Based on the results from Approach 3, there is evidence that the direction and magnitude of the adjustment factor is reasonable and as such it may be cautiously considered that the adjustment factor for total crashes (i.e., All KABCO) for presence of skew angle is validated. 5.2 Rural Two-Lane, 4-Leg Stop Controlled Intersections (Rur2L-4ST) Lighting The base condition is the absence of lighting. Adjustment factors were recommended for “all crash types combined” and are computed using the following equations: 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 + �1 − 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 � 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑃𝑃𝑃𝑃𝑃𝑃 = 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑡𝑡 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝑃𝑃𝑃𝑃𝑃𝑃 + �1 − 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 � and M-45 

𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾|𝑎𝑎𝑎𝑎 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 + �1 − 𝑝𝑝𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾|𝑎𝑎𝑎𝑎 � × 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑃𝑃𝑃𝑃𝑃𝑃 where 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for all crash types and KABC severity, 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑃𝑃𝑃𝑃𝑃𝑃 = CMF for all crash types and PDO severity, 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for all crash types and all severity categories, 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 = proportion of all crashes that occur at night (0.244 is default), 𝑝𝑝𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾|𝑎𝑎𝑎𝑎 = proportion of all crashes that have KABC severity, 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝐾𝐾𝐴𝐴𝐴𝐴𝐴𝐴 = CMF for effect of lighting presence on nighttime crashes with KABC severity (0.80 is default); and 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝑃𝑃𝑃𝑃𝑃𝑃 = CMF for effect of lighting presence on nighttime crashes with PDO severity (0.92 is default). Table 51. Recommended Rur2L-4ST Lighting Adjustment Factors Crash Type Crash AF Severity All KABCO Use equation above KABC Use equation above KAB Same as KABC KA Same as KABC There were 10 sites which met all base conditions (i.e., base conditions of all other variables) except for the lighting. Amongst these 10 sites, 9 sites had no lighting present, whereas only 1 site had lighting present. The recommended adjustment factors cannot be validated since the 1 site with lighting present had no crashes. Left Turn Bay The base condition is the absence of left turn bay. The recommended adjustment factors are shown in Table 52. There were 10 sites which met all base conditions (i.e., base conditions of all other variables) except for the left turn bay. Amongst these 10 sites, 9 sites had no left turn bay present, whereas only 1 site had left turn bay present. The recommended adjustment factors could not be validated since the 1 site that had left turn bays only had 1 observed crash over the studied seven-year period. Table 52. Recommended Rur2L-4ST Left Turn Bay Adjustment Factors Crash Crash Type AF Severity All KABCO 0.72 KABC 0.65 KAB 0.65 KA 0.65 Single vehicle KABCO 0.67 Same direction KABCO 0.56 M-46 

Right Turn Bay The base condition is the absence of right turn bay. The recommended adjustment factors are shown in Table 53. Table 53. Recommended Rur2L-4ST Right Turn Bay Adjustment Factors Crash Crash Type AF Severity All KABCO 0.85 KABC 0.77 KAB 0.77 KA 0.77 Single vehicle KABCO 0.94 Same direction KABCO 0.74 Opposing direction KABCO 0.74 Intersecting direction KABCO 0.91 There were 10 sites which met all base conditions (i.e., base conditions of all other variables) except for the right turn bay. Amongst these 10 sites, 9 sites had no right turn bay present, whereas only 1 site had right turn bay present. Even though a 1 site sample would not yield any conclusive results, the research team still applied Approach 1 for validation. Table 54 shows the results of the Approach 1 analysis. There were no crash categories with at least 50 observed crashes (sites where right turn bay was present). For total crashes (i.e., All KABCO), the implied adjustment factor of presence of right turn bay is 0.40. This does not agree with the recommended adjustment factor of 0.85, although it may also not be statistically different due to the small estimation sample. Table 54. Rur2L-4ST Right Turn Bay Approach 1 Calibration Observed Crashes Predicted Crashes Factors Crash Implied Crash Type RTB RTB RTB RTB RTB RTB Severity AF Present Not Present Not Present Not Present Present Present All KABCO 7 60 7.46 25.48 0.94 2.35 0.40 KABC 2 25 2.09 8.05 0.95 3.11 0.31 KAB 1 11 1.05 4.41 0.96 2.49 0.38 KA 0 1 0.28 1.05 n/a 0.95 n/a Single vehicle KABCO 3 13 1.23 6.93 2.44 1.88 1.3 Same KABCO 1 15 1.59 5.25 0.63 2.86 0.22 direction Opposing KABCO 1 4 0.90 3.08 1.11 1.30 0.86 direction Intersecting KABCO 2 27 5.19 9.8 0.39 2.76 0.14 direction M-47 

Based on the results from Approach 1, there is some evidence that the magnitude of the adjustment factor is reasonable as can be seen from the calibration factor of 0.94 for All KABCO crashes at sites where the right turn bay was present. As such it may be cautiously considered that the adjustment factor for total crashes (i.e., All KABCO) for installation of right turn bay is validated. Intersection Skew The base condition is no skew (i.e., an intersection angel of 90 degrees). The recommended adjustment factors are shown in Table 55. Table 55. Recommended Rur2L-4ST Skew Angle Adjustment Factors Crash Crash Type AF Severity All KABCO 𝑒𝑒𝑒𝑒𝑒𝑒[0.0054 𝑥𝑥 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠] There were 208 sites which met all base conditions (i.e., base conditions of all other variables) except for the skew angle. Amongst these 208 sites, 9 sites had no skew angle, whereas 199 sites had a skew angle. Because the adjustment factor is not a simple factor for one or more levels, Approach 1 and Approach 2 are not capable of making a straightforward validation. Approach 3 was applied which compares the predictive performance of the base model with and without using the adjustment factor applied to sites with a skew angle. Table 56 shows the number of sites, observed, and predicted crashes for All KABCO crashes. As expected, the application of the base models alone underpredicts the number of crashes at sites with a skew angle. However, when the adjustment factor is applied the predictions still underpredict crashes to almost half of the observed crashes. Table 56. Rur2L-4ST Skew Angle Approach 3 No. of Observed Predicted Model Sites (All KABCO Crashes) (All KABCO crashes) Base Model 199 1497 620.48 Base Model + AF 199 1497 679.70 Based on the results from Approach 3, there is no evidence that the direction and magnitude of the adjustment factor is reasonable, and it can be said that the adjustment factor for skew angle is not performing well for this dataset. 5.3 Rural Two-Lane, 4-Leg Signalized Intersections (Rur2L-4SG) Lighting The base condition is the absence of lighting. Adjustment factors were recommended for “all crash types combined” and are computed using the following equations: M-48 

𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 + �1 − 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 � 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑃𝑃𝑃𝑃𝑃𝑃 = 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝑃𝑃𝑃𝑃𝑃𝑃 + �1 − 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 � and 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾|𝑎𝑎𝑎𝑎 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 + �1 − 𝑝𝑝𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾|𝑎𝑎𝑎𝑎 � × 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑃𝑃𝑃𝑃𝑃𝑃 where 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for all crash types and KABC severity, 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑃𝑃𝑃𝑃𝑃𝑃 = CMF for all crash types and PDO severity, 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for all crash types and all severity categories, 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 = proportion of all crashes that occur at night (0.286 is default), 𝑝𝑝𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾|𝑎𝑎𝑎𝑎 = proportion of all crashes that have KABC severity, 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for effect of lighting presence on nighttime crashes with KABC severity (0.80 is default); and 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝑃𝑃𝑃𝑃𝑃𝑃 = CMF for effect of lighting presence on nighttime crashes with PDO severity (0.92 is default). Table 57. Recommended Rur2L-4SG Lighting Adjustment Factors Crash Type Crash AF Severity All KABCO Use equation above KABC Use equation above KAB Same as KABC KA Same as KABC The recommended adjustment factors could not be validated since the dataset used for calibration did not have any variation for lighting, i.e., all sites in the calibration dataset (that met the base conditions for all other variables except for lighting) met the base condition of absence of lighting. Left Turn Bay The base condition is the absence of left turn bay. The recommended adjustment factors are shown in Table 58. Table 58. Recommended Rur2L-4SG Left Turn Bay Adjustment Factors Crash Crash Type AF Severity All KABCO 0.82 KABC 0.89 KAB 0.89 KA 0.89 Same direction KABCO 0.72 M-49 

There were 47 sites which met all base conditions (i.e., base conditions of all other variables) except for the left turn bay. Amongst these 47 sites, 23 sites had no left turn bay present, whereas 24 sites had left turn bay present. Approaches 1 and 2 were both applied. Table 59 shows the results of the Approach 1 analysis. Table 59. Rur2L-4SG Left Turn Bay Approach 1 Calibration Observed Crashes Predicted Crashes Factors Crash Implied Crash Type LTB LTB LTB LTB LTB LTB Severity AF Present Not Present Not Present Not Present Present Present All KABCO 653 478 211.33 138.46 3.09 3.45 0.89 KABC 204 153 66.57 35.63 3.06 4.29 0.71 KAB 60 45 31.85 19.88 1.88 2.26 0.83 KA 11 9 6.37 4.76 1.73 1.89 0.91 Same KABCO 254 199 165.62 80.92 1.53 2.46 0.62 direction The results from Approach 1 show that the implied adjustment factors are very close in effect and magnitude to the recommended calibration factors. In approach 2 negative binomial models were calibrated for All KABCO and SD KABCO crashes. The model form is: 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶ℎ𝑒𝑒𝑒𝑒 = (𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖)(𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦)(𝑆𝑆𝑆𝑆𝑆𝑆 )𝑒𝑒𝑒𝑒𝑒𝑒𝛽𝛽∗𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 𝐵𝐵𝐵𝐵𝐵𝐵 where 𝑆𝑆𝑆𝑆𝑆𝑆 = the calibrated base condition SPF 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 𝐵𝐵𝐵𝐵𝐵𝐵 = 1 if left turn bay is present, 0 if not present 𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂 = the overdispersion parameter of the negative binomial model Table 60 provides the parameter estimates and the implied adjustment factor for All KABCO crashes. Table 60. Rur2L-4SG Left Turn Bay Approach 2, estimate (standard error) Left Turn Crash Type Crash Severity Intercept Overdispersion Implied AF Bay Present All KABCO Model did not Converge 0.9637 -0.5562 SD KABCO 0.2084 0.57 (0.1601) (0.2491) In contrast to Approach 1, the implied adjustment factor from Approach 2 shows a larger benefit for SD KABCO crashes. The statistical significance of the parameters is however very poor, which would translate M-50 

into a high variance for the estimated AF. It should also be noted that the model for All KABCO crashes did not converge. Based on the results from Approach 1 and Approach 2, there is some evidence that the effect and magnitude of the adjustment factors are reasonable and as such it may be cautiously considered that the adjustment factors for all crash types and severities for installation of left turn bay is validated. Right Turn Bay The base condition is the absence of right turn bay. The recommended adjustment factors are shown in Table 61. Table 61. Recommended Rur2L-4SG Right Turn Bay Adjustment Factors Crash Crash Type AF Severity All KABCO 0.96 KABC 0.91 KAB 0.91 KA 0.91 There were 26 sites which met all base conditions (i.e., base conditions of all other variables) except for the right turn bay. Amongst these 26 sites, 23 sites had no right turn bay present, whereas 3 sites had right turn bay present. Even though this small sample may not yield any conclusive results, the research team still applied Approach 1 for validation Table 62. Rur2L-4SG Right Turn Bay Approach 1 Calibration Observed Crashes Predicted Crashes Factors Crash Implied Crash Type RTB RTB RTB RTB RTB RTB Severity AF Present Not Present Not Present Not Present Present Present All KABCO 56 478 20.02 138.46 2.80 3.45 0.81 KABC 13 153 5.04 35.63 2.58 4.29 0.60 KAB 5 45 2.87 19.88 1.74 2.26 0.77 KA 1 9 0.67 4.76 1.49 1.89 0.79 Based on the results from Approach 1, there is some evidence that the effect and magnitude of the adjustment factor is reasonable and reasonably similar to the recommended adjustment factors. As such it may be cautiously considered that the adjustment factor for all crashes for installation of right turn bay are validated. 5.4 Rural Multi-Lane, 3-Leg Stop Controlled Intersections (RurML-3ST) There were 14 rural multi-lane, 3-leg stop controlled intersections in the calibration dataset. None of these sites met the base conditions for the absence of left turn bay. Hence, for the validation analysis M-51 

presented here, sites which met all base conditions (i.e., base conditions of all other variables) except for the left turn bay and the variable of interest were considered. Lighting The base condition is the absence of lighting. Adjustment factors were recommended for “all crash types combined” and are computed using the following equations: 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 + �1 − 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 � 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑃𝑃𝑃𝑃𝑃𝑃 = 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝑃𝑃𝑃𝑃𝑃𝑃 + �1 − 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 � and 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾|𝑎𝑎𝑎𝑎 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 + �1 − 𝑝𝑝𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾|𝑎𝑎𝑎𝑎 � × 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑃𝑃𝑃𝑃𝑃𝑃 where 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for all crash types and KABC severity, 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑃𝑃𝑃𝑃𝑃𝑃 = CMF for all crash types and PDO severity, 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for all crash types and all severity categories, 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 = proportion of all crashes that occur at night (0.276 is default), 𝑝𝑝𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾|𝑎𝑎𝑎𝑎 = proportion of all crashes that have KABC severity, 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for effect of lighting presence on nighttime crashes with KABC severity (0.80 is default); and 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝑃𝑃𝑃𝑃𝑃𝑃 = CMF for effect of lighting presence on nighttime crashes with PDO severity (0.92 is default). Table 63. Recommended RurML-3ST Lighting Adjustment Factors Crash Type Crash AF Severity All KABCO Use equation above KABC Use equation above KAB Same as KABC KA Same as KABC There were 3 sites which met all base conditions (i.e., base conditions of all other variables) except for the left turn bay and lighting. Amongst these 3 sites, 2 sites had no lighting present, whereas 1 site had lighting present. The recommended adjustment factors could not be validated since the 1 site that had lighting present had no observed crashes over the studied seven-year period. Left Turn Bay The base condition is the absence of left turn bay. The recommended adjustment factors are shown in Table 64. M-52 

Table 64. Recommended RurML-3ST Left Turn Bay Adjustment Factors Crash Type Crash Severity AF All KABCO 0.56 KABC 0.45 KAB 0.45 KA 0.45 Single vehicle KABCO 0.61 Same direction KABCO 0.45 Opposing direction KABCO 0.69 Intersecting direction KABCO 0.39 The recommended adjustment factors could not be validated since the dataset used for calibration did not have any variation for left turn bay, i.e., all sites in the calibration dataset did not meet the base condition of absence of left turn bay. Right Turn Bay The base condition is the absence of right turn bay. The recommended adjustment factors are shown in Table 65. Table 65. Recommended RurML-3ST Right Turn Bay Adjustment Factors Crash Crash Type AF Severity All KABCO 0.85 KABC 0.77 KAB 0.77 KA 0.77 Single vehicle KABCO 0.82 Same direction KABCO 0.79 There were 5 sites which met all base conditions (i.e., base conditions of all other variables) except for the left turn bay and right turn bay. Amongst these 5 sites, 2 sites had no right turn bay present, whereas 3 sites had right turn bay present. The recommended adjustment factors could not be validated since both sites that met the base condition and two sites that had right turn bay present had no observed crashes over the studied seven-year period. Intersection Skew The base condition is no skew (i.e., an intersection angel of 90 degrees). Adjustment factors are computed using the following equations: 0.016 ×𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 𝐶𝐶𝐶𝐶𝐶𝐶𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = +1 0.98+(0.016×𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠) and 0.017 ×𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 𝐶𝐶𝐶𝐶𝐶𝐶𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = +1 0.52+(0.017×𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠) M-53 

where 𝐶𝐶𝐶𝐶𝐶𝐶𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for all crash types and all severity categories, 𝐶𝐶𝐶𝐶𝐶𝐶𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for all crash types and KABC severity, 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 = absolute value of the difference between 90 degrees and the actual intersection angle (degrees). The recommended adjustment factors are shown in Table 66. Table 66. Recommended RurML-3ST Skew Angle Adjustment Factors Crash Type Crash AF Severity All KABCO Use equation above KABC Use equation above KAB Same as KABC KA Same as KABC There were 6 sites which met all base conditions (i.e., base conditions of all other variables) except for left turn bay and skew angle. Amongst these 6 sites, 2 sites had no skew angle, whereas 4 sites had a skew angle. Because the adjustment factor is not a simple factor for one or more levels, Approach 1 and Approach 2 are not capable of making a straightforward validation. Approach 3 is applied which compares the predictive performance of the base model with and without using the adjustment factor applied to sites with a skew angle. There were only 7 observed crashes at the sites with a skew angle over the studied seven-year period, as such only the adjustment factor for All KABCO crashes was validated. Table 67 shows the number of sites, observed, and predicted crashes for All KABCO crashes. Table 67. RurML-3ST Skew Angle Approach 3 No. of Observed Predicted Model Sites (All KABCO Crashes) (All KABCO crashes) Base Model 4 7 24.50 Base Model + AF 4 7 17.64 The application of the base models alone overpredicts the number of crashes at sites with a skew angle. When the adjustment factor is applied the predictions still overpredict crashes, but the prediction gets closer to the observed crashes. Based on the results from Approach 3, there is no evidence that the direction and magnitude of the adjustment factor is reasonable, and it can be said that the adjustment factor for skew angle is not performing well for this dataset. M-54 

5.5 Rural Multi-Lane, 4-Leg Stop Controlled Intersections (RurML-4ST) There were 21 rural multi-lane, 4-leg stop controlled intersections in the calibration dataset. None of these sites met the base conditions for the absence of left turn bay and intersection skew. As such it was not possible to isolate for sties that met the base conditions for these two variables. Lighting The base condition is the absence of lighting. Adjustment factors were recommended for “all crash types combined” and are computed using the following equations: 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 + �1 − 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 � 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑃𝑃𝑃𝑃𝑃𝑃 = 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝑃𝑃𝑃𝑃𝑃𝑃 + �1 − 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 � and 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾|𝑎𝑎𝑎𝑎 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 + �1 − 𝑝𝑝𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾|𝑎𝑎𝑎𝑎 � × 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑃𝑃𝑃𝑃𝑃𝑃 where 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for all crash types and KABC severity, 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑃𝑃𝑃𝑃𝑃𝑃 = CMF for all crash types and PDO severity, 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for all crash types and all severity categories, 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 = proportion of all crashes that occur at night (0.273 is default), 𝑝𝑝𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾|𝑎𝑎𝑎𝑎 = proportion of all crashes that have KABC severity, 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for effect of lighting presence on nighttime crashes with KABC severity (0.80 is default); and 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝑃𝑃𝑃𝑃𝑃𝑃 = CMF for effect of lighting presence on nighttime crashes with PDO severity (0.92 is default). Table 68. Recommended RurML-4ST Lighting Adjustment Factors Crash Type Crash AF Severity All KABCO Use equation above KABC Use equation above KAB Same as KABC KA Same as KABC There were 2 sites which met all base conditions (i.e., in this case base condition for right turn bay) except for the lighting. Amongst these 2 sites, 1 site had no lighting present, and 1 site had lighting present. The recommended adjustment factors could not be validated since the 1 site had lighting present has no observed crashes. Left Turn Bay The base condition is the absence of left turn bay. The recommended adjustment factors are shown in Table 69. M-55 

Table 69. Recommended RurML-4ST Left Turn Bay Adjustment Factors Crash Crash Type AF Severity All KABCO 0.72 KABC 0.65 KAB 0.65 KA 0.65 Single vehicle KABCO 0.67 Same direction KABCO 0.56 The recommended adjustment factors could not be validated since the dataset used for calibration did not have any variation for left turn bays, i.e., all sites in the calibration dataset had at least one approach with a left turn bay. Right Turn Bay The base condition is the absence of right turn bay. The recommended adjustment factors are shown in Table 70. Table 70. Recommended RurML-4ST Right Turn Bay Adjustment Factors Crash Crash Type AF Severity All KABCO 0.85 KABC 0.77 KAB 0.77 KA 0.77 Single vehicle KABCO 0.94 Same direction KABCO 0.74 Opposing direction KABCO 0.74 Intersecting direction KABCO 0.91 There were 20 sites which met all base conditions (i.e., in this case base condition for lighting) except for the right turn bay. Amongst these 20 sites, 1 site had no right turn bay present, whereas 19 sites had right turn bay present. Although this sample with almost no variation may not yield any conclusive results, the research team still applied Approach 1 for validation. Table 71 shows the results of the Approach 1 analysis. M-56 

Table 71. RurML-4ST Right Turn Bay Approach 1 Calibration Observed Crashes Predicted Crashes Factors Crash Implied Crash Type RTB RTB RTB RTB RTB RTB Severity AF Present Not Present Not Present Not Present Present Present All KABCO 183 2 172.13 8.51 1.06 0.24 4.54 KABC 95 1 78.33 3.95 1.21 0.25 4.79 KAB 40 0 37.73 2.05 1.06 n/a n/a KA 7 0 5.18 0.21 1.35 n/a n/a Single vehicle KABCO 26 1 46.83 1.97 0.56 0.51 1.09 Same KABCO 17 0 42.28 1.83 0.40 n/a n/a direction Opposing KABCO 15 0 5.18 0.25 2.90 n/a n/a direction Intersecting KABCO 116 1 108.71 4.78 1.07 0.21 5.10 direction Based on the results from Approach 1, there is no evidence that the direction and magnitude of the implied adjustment factors is reasonable, and it can be said that the adjustment factor for right turn bay is not performing well for this dataset. Intersection Skew The base condition is no skew (i.e., an intersection angel of 90 degrees). Adjustment factors are computed using the following equations: 0.053 ×𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 𝐶𝐶𝐶𝐶𝐶𝐶𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = +1 1.43+(0.053×𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠) and 0.048 ×𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 𝐶𝐶𝐶𝐶𝐶𝐶𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = +1 0.72+(0.048×𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠) where 𝐶𝐶𝐶𝐶𝐶𝐶𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for all crash types and all severity categories, 𝐶𝐶𝐶𝐶𝐶𝐶𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for all crash types and KABC severity, 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 = absolute value of the difference between 90 degrees and the actual intersection angle (degrees). The recommended adjustment factors are shown in Table 72. M-57 

Table 72. Recommended RurML-4ST Skew Angle Adjustment Factors Crash Type Crash AF Severity All KABCO Use equation above KABC Use equation above KAB Same as KABC KA Same as KABC There was only 1 site which met all base conditions (i.e., in this case base condition for lighting and right turn bay) except for the right turn bay. Hence, the recommended adjustment factors could not be validated due to the lack of variation in the dataset. 5.6 Rural Multi-Lane, 4-Leg Signalized Intersections (RurML-4SG) Left Turn Bay The base condition is the absence of left turn bay. The recommended adjustment factors are shown in Table 73. Table 73. Recommended RurML-4SG Left Turn Bay Adjustment Factors Crash Crash Type AF Severity All KABCO 0.82 KABC 0.89 KAB 0.89 KA 0.89 Same direction KABCO 0.72 There were 6 sites which met all base conditions (i.e., base conditions of all other variables) except for the left turn bay. Amongst these 6 sites, 4 sites had no left turn bay present, whereas 2 sites had left turn bay present. Even though this small sample may not yield any conclusive results, the research team still applied Approach 1 for validation. Table 74 shows the results of the Approach 1 analysis. M-58 

Table 74. RurML-4SG Left Turn Bay Approach 1 Calibration Observed Crashes Predicted Crashes Factors Crash Implied Crash Type LTB LTB LTB LTB LTB LTB Severity AF Present Not Present Not Present Not Present Present Present All KABCO 30 74 35.21 123.13 0.85 0.60 1.42 KABC 8 22 5.60 32.06 1.43 0.69 2.08 KAB 6 8 2.66 14.56 2.26 0.55 4.11 KA 1 2 0.28 0.49 3.57 4.08 0.87 Same KABCO 11 42 17.43 94.50 0.63 0.44 1.42 direction Based on the results from Approach 1, there is no evidence that the direction and magnitude of the implied adjustment factors is reasonable, and it can be said that the adjustment factor for left turn bay is not performing well for this dataset. Right Turn Bay The base condition is the absence of right turn bay. The recommended adjustment factors are shown in Table 75. Table 75. Recommended RurML-4SG Right Turn Bay Adjustment Factors Crash Crash Type AF Severity All KABCO 0.96 KABC 0.91 KAB 0.91 KA 0.91 The recommended adjustment factors could not be validated since the dataset used for calibration did not have any variation for right turn bay, i.e., all sites in the calibration dataset that met the base condition for all other variables except for right turn bay also met the base condition of absence of right turn bay. 5.7 Urban Arterial, 3-Leg Stop Controlled Intersections (UrbArt-3ST) Lighting The base condition is the presence of lighting. Adjustment factors were recommended for “all crash types combined” and are computed using the following equations: 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 + �1 − 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 � 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑃𝑃𝑃𝑃𝑃𝑃 = 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝑃𝑃𝑃𝑃𝑃𝑃 + �1 − 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 � and 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾|𝑎𝑎𝑎𝑎 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 + �1 − 𝑝𝑝𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾|𝑎𝑎𝑎𝑎 � × 𝐶𝐶𝐶𝐶𝐹𝐹𝑎𝑎𝑎𝑎,𝑃𝑃𝑃𝑃𝑃𝑃 M-59 

where 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for all crash types and KABC severity, 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑃𝑃𝑃𝑃𝑃𝑃 = CMF for all crash types and PDO severity, 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for all crash types and all severity categories, 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 = proportion of all crashes that occur at night (0.238 is default), 𝑝𝑝𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾|𝑎𝑎𝑎𝑎 = proportion of all crashes that have KABC severity, 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for effect of lighting absence on nighttime crashes with KABC severity (1.29 is default); and 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝑃𝑃𝑃𝑃𝑃𝑃 = CMF for effect of lighting absence on nighttime crashes with PDO severity (1.12 is default). Table 76. Recommended UrbArt-3ST Lighting Adjustment Factors Crash Type Crash AF Severity All KABCO Use equation above KABC Use equation above KAB Same as KABC KA Same as KABC With the default values the Adjustment Factor for lighting is 1.23 for KABCO crashes and 1.07 for KABC crashes (Note that since default value for the proportion of all crashes that have KABC severity is not available, the default KABCO adjustment factor was calculated as a product of default adjustment factors for KABC and PDO crashes). There were 50 sites which met all base conditions (i.e., base conditions of all other variables) except for the lighting. Amongst these 50 sites, 29 sites had lighting present, whereas 21 sites had no lighting. Approaches 1 and 2 were both applied. Table 77 shows the results of the Approach 1 analysis. Table 77. UrbArt-3ST Lighting Approach 1 Observed Crashes Predicted Crashes Calibration Factors Crash Lighting Lighting Lighting Lighting Lighting Lighting Implied Crash Type Severity Not Present Not Not Present AF Present Present Present Present All KABCO 152 229 121.24 188.72 1.25 1.21 1.03 KABC 46 82 36.61 55.47 1.26 1.48 0.85 The only crash category with at least 100 observed crashes (sites where lighting was absent) was total crashes (i.e., All KABCO) and the implied adjustment factor of absence of lighting is 1.03. This does not agree with the recommended adjustment factor of 1.23, although it may not be statistically different due to the small estimation sample. In Approach 2 negative binomial models were calibrated for All KABCO crashes. M-60 

The model form is: 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶ℎ𝑒𝑒𝑒𝑒 = (𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖)(𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦)(𝑆𝑆𝑆𝑆𝑆𝑆 )𝑒𝑒𝑒𝑒𝑒𝑒𝛽𝛽∗𝐿𝐿𝐿𝐿𝐿𝐿ℎ𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 where 𝑆𝑆𝑆𝑆𝑆𝑆 = the calibrated base condition SPF 𝐿𝐿𝐿𝐿𝐿𝐿ℎ𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 = 1 if lighting is absent, 0 if present 𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂 = the overdispersion parameter of the negative binomial model Table 78 provides the parameter estimates and the implied adjustment factor for All KABCO crashes. Table 78. UrbArt-3ST Lighting Approach 2, estimate (standard error) Lighting Crash Type Crash Severity Intercept Overdispersion Implied AF Absent 0.1920 -0.0040 All KABCO 0.4184 0.99 (0.2071) (0.3198) In contrast to Approach 1, the implied adjustment factor from Approach 2 for All KABCO shows almost no effect. The statistical significance of the parameters is however very poor, which would translate into a high variance for the estimated AF. Based on the results from Approach 1 and Approach 2, there is some evidence that the effect and magnitude of the adjustment factors are reasonable for ALL KABCO and All KABC crashes, and as such it may be cautiously considered that the adjustment factors for these two crash types for presence of lighting is validated. Left Turn Bay The base condition is the absence of left turn bay. The recommended adjustment factors are shown in Table 79. Table 79. Recommended UrbArt-3ST Left Turn Bay Adjustment Factors Crash Crash Type AF Severity All KABCO 0.63 KABC 0.49 KAB 0.49 KA 0.49 Approaches 1 and 2 were both applied. Table 80 shows the results of the Approach 1 analysis. There were 34 sites which met all base conditions (i.e., base conditions of all other variables) except for the left turn bay. Amongst these 34 sites, 29 sites had no left turn bay present, whereas 5 sites had left turn bay present. M-61 

Table 80. UrbArt-3ST Left Turn Bay Approach 1 Calibration Observed Crashes Predicted Crashes Factors Crash Implied Crash Type LTB LTB LTB LTB LTB LTB Severity AF Present Not Present Not Present Not Present Present Present All KABCO 103 229 87.78 188.72 1.17 1.21 0.97 KABC 28 82 28.56 55.47 0.98 1.48 0.66 KAB 11 21 13.44 31.85 0.82 0.66 1.24 KA 1 2 2.38 7.28 0.42 0.27 1.53 The only crash category with at least 100 observed crashes (sites where left turn bay was present) was total crashes (i.e., All KABCO) and the implied adjustment factor of presence of left turn bay is 0.97. This does not agree with the recommended adjustment factor of 0.63, although it may not be statistically different due to the small estimation sample. In approach 2 negative binomial models were calibrated for All KABCO crashes. Other crash types had few crashes in total so were not analyzed. The model form is: 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶ℎ𝑒𝑒𝑒𝑒 = (𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖)(𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦)(𝑆𝑆𝑆𝑆𝑆𝑆 )𝑒𝑒𝑒𝑒𝑒𝑒𝛽𝛽∗𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 𝐵𝐵𝐵𝐵𝐵𝐵 where 𝑆𝑆𝑆𝑆𝑆𝑆 = the calibrated base condition SPF 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 𝐵𝐵𝐵𝐵𝐵𝐵 = 1 if left turn bay is present, 0 if not present 𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂 = the overdispersion parameter of the negative binomial model Table 81 provides the parameter estimates and the implied adjustment factor for All KABCO crashes. Table 81. UrbArt-3ST Left Turn Bay Approach 2, estimate (standard error) Left Turn Crash Type Crash Severity Intercept Overdispersion Implied AF Bay Present 0.1920 0.2764 All KABCO 0.4686 1.32 (0.2112) (0.5134) In contrast to Approach 1 the adjustment factor is greater than 1 for All KABCO crashes. The statistical significance of the parameters is however very poor, which would translate into a high variance for the estimated AF. M-62 

Based on the results from Approach 1, there is some evidence that the effect and magnitude of the adjustment factors are reasonable for All KABCO and All KABC crashes, and as such it may be cautiously considered that the adjustment factors for these two crash types for installation of left turn bay is validated. Right Turn Bay The base condition is the absence of right turn bay. The recommended adjustment factors are shown in Table 82. Table 82. Recommended UrbArt-3ST Right Turn Bay Adjustment Factors Crash Crash Type AF Severity All KABCO 0.85 KABC 0.77 KAB 0.77 KA 0.77 There were 35 sites which met all base conditions (i.e., base conditions of all other variables) except for the right turn bay. Amongst these 35 sites, 29 sites had no right turn bay present, whereas 6 sites had right turn bay present. There were no crash categories with at least 100 observed crashes (sites where right turn bay was present), hence only Approach 1 was applied. Table 83 shows the results of the Approach 1 analysis. Table 83. UrbArt-3ST Right Turn Bay Approach 1 Calibration Observed Crashes Predicted Crashes Factors Crash Implied Crash Type RTB RTB RTB RTB RTB RTB Severity AF Present Not Present Not Present Not Present Present Present All KABCO 40 229 66.57 188.72 0.60 1.21 0.49 KABC 15 82 20.30 55.47 0.74 1.48 0.50 KAB 2 21 10.29 31.85 0.19 0.66 0.29 KA 0 2 2.03 7.28 n/a 0.27 n/a For total crashes (i.e., All KABCO), the implied adjustment factor of presence of right turn bay is 0.49. This does not agree with the recommended adjustment factor of 0.85, although it may not be statistically different due to the small estimation sample. Based on the results from Approach 1, it can be said that the adjustment factor for right turn bay is not performing well for this dataset. 5.8 Urban Arterial, 3-Leg Signalized Intersections (UrbArt-3SG) There were 9 urban arterial, 3-leg signalized intersections in the calibration dataset. All of these sites had left turn bay present. As such it was not possible to isolate for sites that met the base conditions of absence of left turn bay. M-63 

Left Turn Bay The base condition is the absence of left turn bay. The recommended adjustment factors are shown in Table 84. Table 84. Recommended UrbArt-3SG Left Turn Bay Adjustment Factors Crash Type Crash Severity AF All KABCO 0.93 KABC 0.48 KAB 0.48 KA 0.48 The recommended adjustment factors could not be validated since the dataset used for calibration did not have any variation for left turn bay, i.e., all sites in the calibration dataset had a left turn bay present. Right Turn Bay The base condition is the absence of right turn bay. The recommended adjustment factors are shown in Table 85. Table 85. Recommended UrbArt-3SG Right Turn Bay Adjustment Factors Crash Type Crash Severity AF All KABCO 0.96 KABC 0.91 KAB 0.91 KA 0.91 There were 9 sites which met all base conditions (i.e., base conditions of all other variables plus presence of left turn bay) except for the right turn bay. Of these 9 sites, 1 site had no right turn bay present, whereas 8 sites had right turn bay present. Even though this small sample may not yield significant results, the research nonetheless applied Approach 1. Table 86 shows the results of the Approach 1 analysis. Table 86. UrbArt-3SG Right Turn Bay Approach 1 Calibration Observed Crashes Predicted Crashes Factors Crash Implied Crash Type RTB RTB RTB RTB RTB RTB Severity AF Present Not Present Not Present Not Present Present Present All KABCO 336 6 316.54 3.55 1.06 0.24 4.39 KABC 96 2 96 0.77 1.36 0.37 3.66 KAB 20 0 20 0.34 0.63 n/a n/a KA 2 0 2 0.07 0.31 n/a n/a M-64 

It can be seen that the 1 site where right turn bay was not present was a low crash site and it skews the results of the analysis. Based on the results from Approach 1, it can be said that the adjustment factor for right turn bay could not be validated due to the lack of variation in the calibration dataset. Red Light Running Camera Enforcement The base condition is the no red light running camera enforcement. Recommended adjustment factors are computed using the following equations: 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑧𝑧 = 𝑝𝑝𝑠𝑠𝑠𝑠 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑠𝑠𝑠𝑠,𝑧𝑧 + 𝑝𝑝𝑠𝑠𝑠𝑠 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑠𝑠𝑠𝑠,𝑧𝑧 + 𝑝𝑝𝑜𝑜𝑜𝑜 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑜𝑜𝑜𝑜 + 𝑝𝑝𝑖𝑖𝑖𝑖 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑖𝑖𝑖𝑖 where 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑧𝑧 = CMF for all crash types and severity category z, 𝑝𝑝𝑠𝑠𝑠𝑠 = proportion of crashes that are single vehicle, 𝑝𝑝𝑠𝑠𝑠𝑠 = proportion of crashes that are same direction, 𝑝𝑝𝑜𝑜𝑜𝑜 = proportion of crashes that are opposing direction, and 𝑝𝑝𝑖𝑖𝑖𝑖 = proportion of crashes that are intersecting direction. The CMFs for opposing direction and intersecting direction crash types are computed using the following equations: 𝐶𝐶𝐶𝐶𝐶𝐶𝑜𝑜𝑜𝑜 = 𝑝𝑝𝑎𝑎𝑎𝑎|𝑜𝑜𝑜𝑜 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎 + 𝑝𝑝𝑡𝑡𝑡𝑡|𝑜𝑜𝑜𝑜 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑡𝑡𝑡𝑡 + (1 − 𝑝𝑝𝑎𝑎𝑎𝑎|𝑜𝑜𝑜𝑜 − 𝑝𝑝𝑡𝑡𝑡𝑡|𝑜𝑜𝑜𝑜 ) 𝐶𝐶𝐶𝐶𝐶𝐶𝑖𝑖𝑖𝑖 = 𝑝𝑝𝑎𝑎𝑎𝑎|𝑖𝑖𝑖𝑖 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎 + 𝑝𝑝𝑡𝑡𝑡𝑡|𝑖𝑖𝑖𝑖 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑡𝑡𝑡𝑡 + (1 − 𝑝𝑝𝑎𝑎𝑎𝑎|𝑖𝑖𝑖𝑖 − 𝑝𝑝𝑡𝑡𝑡𝑡|𝑖𝑖𝑖𝑖 ) where 𝐶𝐶𝐶𝐶𝐶𝐶𝑜𝑜𝑜𝑜 = CMF for opposing direction crash types, 𝑝𝑝𝑎𝑎𝑎𝑎|𝑜𝑜𝑜𝑜 = proportion of opposing direction crashes that are angle (excluding left-turn-related angle crashes) (default value = 0.005), 𝑝𝑝𝑡𝑡𝑡𝑡|𝑜𝑜𝑜𝑜 = proportion of opposing direction crashes that are left turn related (default value = 0.984), 𝐶𝐶𝐶𝐶𝐶𝐶𝑖𝑖𝑖𝑖 = CMF for intersecting direction crash types, 𝑝𝑝𝑎𝑎𝑎𝑎|𝑖𝑖𝑖𝑖 = proportion of intersecting direction crashes that are angle (excluding left-turn-related angle crashes) (default value = 0.644), 𝑝𝑝𝑡𝑡𝑡𝑡|𝑖𝑖𝑖𝑖 = proportion of intersecting direction crashes that are left turn related (default value = 0.193), 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎 = CMF for angle crashes (default value = 0.8), and 𝐶𝐶𝐶𝐶𝐶𝐶𝑡𝑡𝑡𝑡 = CMF for left-turn-related crashes (default value = 0.60). M-65 

Table 87. Recommended UrbArt-3SG Red Light Camera Enforcement Adjustment Factors Crash Type Crash Severity AF All KABCO Use equation above KABC Use equation above KAB Use equation above KA Use equation above Single Vehicle KABCO 1.00 KABC 1.00 KAB 1.00 KA 1.00 Same Direction KABCO 1.155 KABC 1.234 KAB 1.240 KA 1.24 Opposing Direction KABCO Use equation above KABC Use equation above KAB Use equation above KA Use equation above Intersecting Direction KABCO Use equation above KABC Use equation above KAB Use equation above KA Use equation above The recommended adjustment factors could not be validated since the dataset used for calibration did not have any variation for red light running camera enforcement, i.e., all sites in the calibration dataset had no red-light running camera enforcement. Right Turn on Red Prohibition The base condition is right turn on red allowed on all approaches. Recommended adjustment factors are computed using the following equations: 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑧𝑧 = 𝑝𝑝𝑠𝑠𝑠𝑠 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑠𝑠𝑠𝑠,𝑧𝑧 + 𝑝𝑝𝑠𝑠𝑠𝑠 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑠𝑠𝑠𝑠,𝑧𝑧 + 𝑝𝑝𝑜𝑜𝑜𝑜 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑜𝑜𝑜𝑜 + 𝑝𝑝𝑖𝑖𝑖𝑖 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑖𝑖𝑖𝑖 where 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑧𝑧 = CMF for all crash types and severity category z, 𝑝𝑝𝑠𝑠𝑠𝑠 = proportion of crashes that are single vehicle, 𝑝𝑝𝑠𝑠𝑠𝑠 = proportion of crashes that are same direction, 𝑝𝑝𝑜𝑜𝑜𝑜 = proportion of crashes that are opposing direction, and 𝑝𝑝𝑖𝑖𝑖𝑖 = proportion of crashes that are intersecting direction. M-66 

The CMFs for same, opposing, and intersecting direction crash types with KABC severity are computed using the following equations: 𝑛𝑛 𝐶𝐶𝐶𝐶𝐶𝐶𝑠𝑠𝑠𝑠,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝑡𝑡𝑡𝑡|𝑠𝑠𝑠𝑠 × �𝐶𝐶𝐶𝐶𝐶𝐶𝑡𝑡𝑡𝑡,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 � + (1 − 𝑝𝑝𝑡𝑡𝑡𝑡|𝑠𝑠𝑠𝑠 ) 𝑛𝑛 𝐶𝐶𝐶𝐶𝐶𝐶𝑜𝑜𝑜𝑜,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝑡𝑡𝑡𝑡|𝑜𝑜𝑜𝑜 × �𝐶𝐶𝐶𝐶𝐶𝐶𝑡𝑡𝑡𝑡,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 � + (1 − 𝑝𝑝𝑡𝑡𝑡𝑡|𝑜𝑜𝑜𝑜 ) 𝑛𝑛 𝐶𝐶𝐶𝐶𝐶𝐶𝑖𝑖𝑖𝑖,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝑡𝑡𝑡𝑡|𝑖𝑖𝑖𝑖 × �𝐶𝐶𝐶𝐶𝐶𝐶𝑡𝑡𝑡𝑡,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 � + (1 − 𝑝𝑝𝑡𝑡𝑡𝑡|𝑖𝑖𝑖𝑖 ) where 𝐶𝐶𝐶𝐶𝐶𝐶𝑠𝑠𝑠𝑠,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for same direction crash types with KABC severity, 𝑝𝑝𝑡𝑡𝑡𝑡|𝑠𝑠𝑠𝑠 = proportion of same direction crashes that are right turn related (default value = 0.099), 𝐶𝐶𝐶𝐶𝐶𝐶𝑜𝑜𝑜𝑜,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for opposing direction crash types with KABC severity, 𝑝𝑝𝑡𝑡𝑡𝑡|𝑜𝑜𝑜𝑜 = proportion of opposing direction crashes that are right turn related (default value = 0.029), 𝐶𝐶𝐶𝐶𝐶𝐶𝑖𝑖𝑖𝑖,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for intersecting direction crash types with KABC severity, 𝑝𝑝𝑡𝑡𝑡𝑡|𝑖𝑖𝑖𝑖 = proportion of intersecting direction crashes that are right turn related (default value = 0.191), 𝐶𝐶𝐶𝐶𝐶𝐶𝑡𝑡𝑡𝑡,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for right-turn-related crashes with KABC severity (default value = 0.625 = 1/1.6), and 𝑛𝑛 = number of signalized intersection approaches for which right-turn-on-red is prohibited. The CMFs for same, opposing, and intersecting direction crash types with KABCO severity are computed using the following equations: 𝑛𝑛 𝐶𝐶𝐶𝐶𝐶𝐶𝑠𝑠𝑠𝑠,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝑠𝑠𝑠𝑠|𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑠𝑠𝑠𝑠,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 + �1 − 𝑝𝑝𝑠𝑠𝑑𝑑|𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 � × �𝑝𝑝𝑡𝑡𝑡𝑡|𝑠𝑠𝑠𝑠 × �𝐶𝐶𝐶𝐶𝐶𝐶𝑡𝑡𝑡𝑡,𝑂𝑂 � + (1 − 𝑝𝑝𝑡𝑡𝑡𝑡|𝑠𝑠𝑠𝑠 )� 𝑛𝑛 𝐶𝐶𝐶𝐶𝐶𝐶𝑜𝑜𝑜𝑜,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝑜𝑜𝑜𝑜|𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑜𝑜𝑜𝑜,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 + �1 − 𝑝𝑝𝑜𝑜𝑜𝑜|𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 � × �𝑝𝑝𝑡𝑡𝑡𝑡|𝑜𝑜𝑜𝑜 × �𝐶𝐶𝐶𝐶𝐶𝐶𝑡𝑡𝑡𝑡,𝑂𝑂 � + (1 − 𝑝𝑝𝑡𝑡𝑡𝑡|𝑜𝑜𝑜𝑜 )� 𝑛𝑛 𝐶𝐶𝐶𝐶𝐶𝐶𝑖𝑖𝑖𝑖,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝑖𝑖𝑖𝑖|𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑖𝑖𝑖𝑖,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 + �1 − 𝑝𝑝𝑖𝑖𝑖𝑖|𝐾𝐾𝐾𝐾𝐾𝐾𝐶𝐶 � × �𝑝𝑝𝑡𝑡𝑡𝑡|𝑖𝑖𝑖𝑖 × �𝐶𝐶𝐶𝐶𝐶𝐶𝑡𝑡𝑡𝑡,𝑂𝑂 � + (1 − 𝑝𝑝𝑡𝑡𝑡𝑡|𝑖𝑖𝑖𝑖 )� where 𝐶𝐶𝐶𝐶𝐶𝐶𝑠𝑠𝑠𝑠,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for same direction crash types with KABCO severity, 𝑝𝑝𝑠𝑠𝑠𝑠|𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = proportion of KABC crashes that are same direction, 𝐶𝐶𝐶𝐶𝐶𝐶𝑜𝑜𝑜𝑜,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for opposing direction crash types with KABCO severity, 𝑝𝑝𝑜𝑜𝑜𝑜|𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = proportion of KABC crashes that are opposing direction, 𝐶𝐶𝐶𝐶𝐶𝐶𝑖𝑖𝑖𝑖,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for intersecting direction crash types with KABCO severity, 𝑝𝑝𝑖𝑖𝑖𝑖|𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = proportion of KABC crashes that are intersecting direction, and 𝐶𝐶𝐶𝐶𝐶𝐶𝑡𝑡𝑡𝑡,𝑂𝑂 = CMF for right-turn-related crashes with PDO severity (default value = 0.909 = 1/1.1); M-67 

Table 88. Recommended UrbArt-3SG Right Turn on Red Prohibition Adjustment Factors Crash Type Crash Severity AF All KABCO Use equation above KABC Use equation above KAB Use equation above KA Use equation above Single Vehicle KABCO 1.00 KABC 1.00 KAB 1.00 KA 1.00 Same Direction KABCO Use equation above KABC Use equation above KAB Use equation above KA Use equation above Opposing Direction KABCO Use equation above KABC Use equation above KAB Use equation above KA Use equation above Intersecting Direction KABCO Use equation above KABC Use equation above KAB Use equation above KA Use equation above The recommended adjustment factors could not be validated since the dataset used for calibration did not have any variation for right turn on red prohibition, i.e., all sites in the calibration dataset had right turn n red allowed on all approaches. 5.9 Urban Arterial, 4-Leg Stop Controlled Intersections (UrbArt-4ST) Lighting The base condition is the presence of lighting. Adjustment factors were recommended for “all crash types combined” and are computed using the following equations: 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝐾𝐾𝐴𝐴𝐴𝐴𝐴𝐴 + �1 − 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 � 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑃𝑃𝑃𝑃𝑃𝑃 = 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝑃𝑃𝑃𝑃𝑃𝑃 + �1 − 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 � and 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾|𝑎𝑎𝑎𝑎 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 + �1 − 𝑝𝑝𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾|𝑎𝑎𝑎𝑎 � × 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑃𝑃𝑃𝑃𝑃𝑃 where 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for all crash types and KABC severity, 𝐶𝐶𝑀𝑀𝑀𝑀𝑎𝑎𝑎𝑎,𝑃𝑃𝑃𝑃𝑃𝑃 = CMF for all crash types and PDO severity, 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for all crash types and all severity categories, 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 = proportion of all crashes that occur at night (0.229 is default), 𝑝𝑝𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾|𝑎𝑎𝑎𝑎 = proportion of all crashes that have KABC severity, M-68 

𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for effect of lighting absence on nighttime crashes with KABC severity (1.29 is default); and 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝑃𝑃𝑃𝑃𝑃𝑃 = CMF for effect of lighting absence on nighttime crashes with PDO severity (1.12 is default). Table 89. Recommended UrbArt-4ST Lighting Adjustment Factors Crash Type Crash AF Severity All KABCO Use equation above KABC Use equation above KAB Same as KABC KA Same as KABC With the default values the Adjustment Factor for lighting is 1.22 for KABCO crashes and 1.07 for KABC crashes (Note that since default value for the proportion of all crashes that have KABC severity is not available, the default KABCO adjustment factor was calculated as a product of default adjustment factors for KABC and PDO crashes). There were 19 sites which met all base conditions (i.e., base conditions of all other variables) except for the lighting. Amongst these 19 sites, 12 sites had lighting present, whereas 7 sites had no lighting. There were no crash categories with at least 100 observed crashes (sites where lighting was absent), hence only Approach 1 was applied. Table 90 shows the results of the Approach 1 analysis. Table 90. UrbArt-4ST Lighting Approach 1 Observed Crashes Predicted Crashes Calibration Factors Crash Lighting Lighting Lighting Lighting Lighting Lighting Implied Crash Type Severity Not Present Not Not Present AF Present Present Present Present All KABCO 73 128 78.26 108.29 0.93 1.18 0.79 KABC 32 47 23.38 38.43 1.37 1.22 1.12 For total crashes (i.e., All KABCO) the implied adjustment factor for the absence of lighting is 0.79 (counterintuitive to the recommended adjustment factor of 1.22), whereas, for total injury crashes (i.e., All KABC) the implied adjustment factor of absence of lighting is 1.12 (very close to the recommended adjustment factor of 1.07). Based on the results of the Approach 1 analysis, it can be seen that the implied adjustment factor for KABC crashes is very similar to the recommended adjustment factor, and it may not be statistically different due to the small estimation sample. Hence, it can be said that the adjustment factor for the absence of lighting for KABC crashes has been validated. Left Turn Bay The base condition is the absence of left turn bay. The recommended adjustment factors are shown in Table 91. M-69 

Table 91. Recommended UrbArt-4ST Left Turn Bay Adjustment Factors Crash Crash Type AF Severity All KABCO 0.73 KABC 0.71 KAB 0.71 KA 0.71 There were 13 sites which met all base conditions (i.e., base conditions of all other variables) except for the left turn bay. Amongst these 13 sites, 12 sites had no left turn bay present, whereas only 1 site had left turn bay present. Even though a 1 site sample may not yield significant results, the research nonetheless applied Approach 1. Table 92 shows the results of the Approach 1 analysis. Table 92. UrbArt-4ST Left Turn Bay Approach 1 Calibration Observed Crashes Predicted Crashes Factors Crash Implied Crash Type LTB LTB LTB LTB LTB LTB Severity AF Present Not Present Not Present Not Present Present Present All KABCO 84 128 7.49 108.29 11.21 1.18 9.49 KABC 36 47 3.08 38.43 11.69 1.22 9.56 KAB 4 17 0.84 10.5 4.76 1.62 2.94 KA 0 1 0.02 0.28 n/a 3.57 n/a It can be seen that the 1 site where left turn bay was present was a high crash site and skews the results of the analysis. Based on the results from Approach 1, it can be said that the adjustment factors for left turn bay could not be validated due to the lack of variation in the calibration dataset. Right Turn Bay The base condition is the absence of right turn bay. The recommended adjustment factors are shown in Table 93. Table 93. Recommended UrbArt-4ST Right Turn Bay Adjustment Factors Crash Crash Type AF Severity All KABCO 0.85 KABC 0.77 KAB 0.77 KA 0.77 M-70 

There were 12 sites which met all base conditions (i.e., base conditions of all other variables) except for the right turn bay. All these 12 sites had no right turn bay present. Hence, the recommended adjustment factors could not be validated since the dataset used for calibration did not have any variation for right turn bay, i.e., all sites in the calibration dataset met the base condition of absence of right turn bay. 5.10 Urban Arterial, 4-Leg Signalized Intersections (UrbArt-4SG) Lighting The base condition is the presence of lighting. Adjustment factors were recommended for “all crash types combined” and are computed using the following equations: 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 + �1 − 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 � 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑃𝑃𝑃𝑃𝑃𝑃 = 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝑃𝑃𝑃𝑃𝑃𝑃 + �1 − 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 � and 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾|𝑎𝑎𝑎𝑎 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 + �1 − 𝑝𝑝𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾|𝑎𝑎𝑎𝑎 � × 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑃𝑃𝑃𝑃𝑃𝑃 where 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for all crash types and KABC severity, 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑃𝑃𝑃𝑃𝑃𝑃 = CMF for all crash types and PDO severity, 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for all crash types and all severity categories, 𝑝𝑝𝑛𝑛𝑛𝑛|𝑎𝑎𝑎𝑎 = proportion of all crashes that occur at night (0.235 is default), 𝑝𝑝𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾|𝑎𝑎𝑎𝑎 = proportion of all crashes that have KABC severity, 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝐾𝐾𝐾𝐾𝐵𝐵𝐶𝐶 = CMF for effect of lighting absence on nighttime crashes with KABC severity (1.29 is default); and 𝐶𝐶𝐶𝐶𝐶𝐶𝑛𝑛𝑛𝑛,𝑃𝑃𝑃𝑃𝑃𝑃 = CMF for effect of lighting absence on nighttime crashes with PDO severity (1.12 is default). Table 94. Recommended UrbArt-4SG Lighting Adjustment Factors Crash Type Crash AF Severity All KABCO Use equation above KABC Use equation above KAB Same as KABC KA Same as KABC The recommended adjustment factors could not be validated since the dataset used for calibration did not have any variation for lighting, i.e., all sites in the calibration dataset (that met the base conditions for all other variables) met the base condition of presence of lighting. Left Turn Bay The base condition is the absence of left turn bay. The recommended adjustment factors are shown in Table 95. M-71 

Table 95. Recommended UrbArt-4SG Left Turn Bay Adjustment Factors Crash Crash Type AF Severity All KABCO 0.90 KABC 0.91 KAB 0.91 KA 0.91 There were 28 sites which met all base conditions (i.e., base conditions of all other variables) except for the left turn bay. Amongst these 28 sites, 3 sites had no left turn bay present, whereas 25 sites had left turn bay present. Even though this small sample may not yield significant results, the research team nonetheless applied Approach 1. Table 96 shows the results of the Approach 1 analysis. Table 96. UrbArt-4SG Left Turn Bay Approach 1 Calibration Observed Crashes Predicted Crashes Factors Crash Implied Crash Type LTB LTB LTB LTB LTB LTB Severity AF Present Not Present Not Present Not Present Present Present All KABCO 1203 48 1092.77 76.02 1.10 0.63 1.74 KABC 442 11 292.88 18.27 1.51 0.60 2.51 KAB 91 2 117.53 8.05 0.77 0.25 3.12 KA 7 0 20.79 1.61 0.34 n/a n/a It can be seen that the 3 sites where left turn bay was not present were low crash sites and skew the results of the analysis. Based on the results from Approach 1, it can be said that the adjustment factors for left turn bay cannot be validated due to the lack of variation in the calibration dataset. Right Turn Bay The base condition is the absence of right turn bay. The recommended adjustment factors are shown in Table 97. Table 97. Recommended UrbArt-4SG Right Turn Bay Adjustment Factors Crash Crash Type AF Severity All KABCO 0.96 KABC 0.91 KAB 0.91 KA 0.91 M-72 

There were 5 sites which met all base conditions (i.e., base conditions of all other variables) except for the right turn bay. Of these 5 sites, 3 sites had no right turn bay present, whereas 2 sites had right turn bay present. Even though this small sample may not yield significant results, the research team nonetheless applied Approach 1. Table 98 shows the results of the Approach 1 analysis. Table 98. UrbArt-4SG Right Turn Bay Approach 1 Calibration Observed Crashes Predicted Crashes Factors Crash Implied Crash Type RTB RTB RTB RTB RTB RTB Severity AF Present Not Present Not Present Not Present Present Present All KABCO 45 48 73.57 76.02 0.61 0.63 0.97 KABC 10 11 19.32 18.27 0.52 0.60 0.86 KAB 3 2 7.91 8.05 0.38 0.25 1.53 KA 0 0 1.4 1.61 n/a n/a n/a For total crashes (i.e., All KABCO) the implied adjustment factor for the presence of right turn bay is 0.97 (very similar to the recommended adjustment factor of 0.96), whereas, for total injury crashes (i.e., All KABC) the implied adjustment factor for the presence of right turn bay is 0.86 (very close to the recommended adjustment factor of 0.91). Based on the results of the Approach 1 analysis, it can be seen that the implied adjustment factors for All KABCO and All KABC crashes are very similar to the recommended adjustment factor, and they may not be statistically different due to the small estimation sample. Hence, it can be said with caution (due to the small sample size) that the adjustment factor for presence of right turn bay for All KABCO and All KABC crashes have been validated. Red Light Running Camera Enforcement The base condition is the no red light running camera enforcement. Recommended adjustment factors are computed using the following equations: 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑧𝑧 = 𝑝𝑝𝑠𝑠𝑠𝑠 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑠𝑠𝑠𝑠,𝑧𝑧 + 𝑝𝑝𝑠𝑠𝑠𝑠 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑠𝑠𝑠𝑠,𝑧𝑧 + 𝑝𝑝𝑜𝑜𝑜𝑜 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑜𝑜𝑜𝑜 + 𝑝𝑝𝑖𝑖𝑖𝑖 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑖𝑖𝑖𝑖 where 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑧𝑧 = CMF for all crash types and severity category z, 𝑝𝑝𝑠𝑠𝑠𝑠 = proportion of crashes that are single vehicle, 𝑝𝑝𝑠𝑠𝑠𝑠 = proportion of crashes that are same direction, 𝑝𝑝𝑜𝑜𝑜𝑜 = proportion of crashes that are opposing direction, and 𝑝𝑝𝑖𝑖𝑖𝑖 = proportion of crashes that are intersecting direction. M-73 

The CMFs for opposing direction and intersecting direction crash types are computed using the following equations: 𝐶𝐶𝐶𝐶𝐶𝐶𝑜𝑜𝑜𝑜 = 𝑝𝑝𝑎𝑎𝑎𝑎|𝑜𝑜𝑜𝑜 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎 + 𝑝𝑝𝑡𝑡𝑡𝑡|𝑜𝑜𝑜𝑜 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑡𝑡𝑡𝑡 + (1 − 𝑝𝑝𝑎𝑎𝑎𝑎|𝑜𝑜𝑜𝑜 − 𝑝𝑝𝑡𝑡𝑡𝑡|𝑜𝑜𝑜𝑜 ) 𝐶𝐶𝐶𝐶𝐶𝐶𝑖𝑖𝑖𝑖 = 𝑝𝑝𝑎𝑎𝑎𝑎|𝑖𝑖𝑖𝑖 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎 + 𝑝𝑝𝑡𝑡𝑡𝑡|𝑖𝑖𝑖𝑖 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑡𝑡𝑡𝑡 + (1 − 𝑝𝑝𝑎𝑎𝑎𝑎|𝑖𝑖𝑖𝑖 − 𝑝𝑝𝑡𝑡𝑡𝑡|𝑖𝑖𝑖𝑖 ) where 𝐶𝐶𝐶𝐶𝐶𝐶𝑜𝑜𝑜𝑜 = CMF for opposing direction crash types, 𝑝𝑝𝑎𝑎𝑎𝑎|𝑜𝑜𝑜𝑜 = proportion of opposing direction crashes that are angle (excluding left-turn-related angle crashes) (default value = 0.005), 𝑝𝑝𝑡𝑡𝑡𝑡|𝑜𝑜𝑜𝑜 = proportion of opposing direction crashes that are left turn related (default value = 0.984), 𝐶𝐶𝐶𝐶𝐶𝐶𝑖𝑖𝑖𝑖 = CMF for intersecting direction crash types, 𝑝𝑝𝑎𝑎𝑎𝑎|𝑖𝑖𝑖𝑖 = proportion of intersecting direction crashes that are angle (excluding left-turn-related angle crashes) (default value = 0.644), 𝑝𝑝𝑡𝑡𝑡𝑡|𝑖𝑖𝑖𝑖 = proportion of intersecting direction crashes that are left turn related (default value = 0.193), 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎 = CMF for angle crashes (default value = 0.8), and 𝐶𝐶𝐶𝐶𝐶𝐶𝑡𝑡𝑡𝑡 = CMF for left-turn-related crashes (default value = 0.60). Table 99. Recommended UrbArt-3SG Red Light Camera Enforcement Adjustment Factors Crash Type Crash AF Severity All KABCO Use equation above KABC Use equation above KAB Use equation above KA Use equation above Single Vehicle KABCO 1.00 KABC 1.00 KAB 1.00 KA 1.00 Same Direction KABCO 1.155 KABC 1.234 KAB 1.240 KA 1.24 Opposing Direction KABCO Use equation above KABC Use equation above KAB Use equation above KA Use equation above Intersecting Direction KABCO Use equation above KABC Use equation above KAB Use equation above KA Use equation above The recommended adjustment factors could not be validated since the dataset used for calibration did not have any variation for red light running camera enforcement, i.e., all sites in the calibration dataset had no red light running camera enforcement. M-74 

Right Turn on Red Prohibition The base condition is right turn on red allowed on all approaches. Recommended adjustment factors are computed using the following equations: 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑧𝑧 = 𝑝𝑝𝑠𝑠𝑠𝑠 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑠𝑠𝑠𝑠,𝑧𝑧 + 𝑝𝑝𝑠𝑠𝑠𝑠 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑠𝑠𝑠𝑠,𝑧𝑧 + 𝑝𝑝𝑜𝑜𝑜𝑜 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑜𝑜𝑜𝑜 + 𝑝𝑝𝑖𝑖𝑖𝑖 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑖𝑖𝑖𝑖 where 𝐶𝐶𝐶𝐶𝐶𝐶𝑎𝑎𝑎𝑎,𝑧𝑧 = CMF for all crash types and severity category z, 𝑝𝑝𝑠𝑠𝑠𝑠 = proportion of crashes that are single vehicle, 𝑝𝑝𝑠𝑠𝑠𝑠 = proportion of crashes that are same direction, 𝑝𝑝𝑜𝑜𝑜𝑜 = proportion of crashes that are opposing direction, and 𝑝𝑝𝑖𝑖𝑖𝑖 = proportion of crashes that are intersecting direction. The CMFs for same, opposing, and intersecting direction crash types with KABC severity are computed using the following equations: 𝑛𝑛 𝐶𝐶𝐶𝐶𝐶𝐶𝑠𝑠𝑠𝑠,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝑡𝑡𝑡𝑡|𝑠𝑠𝑠𝑠 × �𝐶𝐶𝐶𝐶𝐶𝐶𝑡𝑡𝑡𝑡,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 � + (1 − 𝑝𝑝𝑡𝑡𝑡𝑡|𝑠𝑠𝑠𝑠 ) 𝑛𝑛 𝐶𝐶𝐶𝐶𝐶𝐶𝑜𝑜𝑜𝑜,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝑡𝑡𝑡𝑡|𝑜𝑜𝑑𝑑 × �𝐶𝐶𝐶𝐶𝐶𝐶𝑡𝑡𝑡𝑡,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 � + (1 − 𝑝𝑝𝑡𝑡𝑡𝑡|𝑜𝑜𝑜𝑜 ) 𝑛𝑛 𝐶𝐶𝐶𝐶𝐶𝐶𝑖𝑖𝑖𝑖,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝑡𝑡𝑡𝑡|𝑖𝑖𝑖𝑖 × �𝐶𝐶𝐶𝐶𝐶𝐶𝑡𝑡𝑡𝑡,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 � + (1 − 𝑝𝑝𝑡𝑡𝑡𝑡|𝑖𝑖𝑖𝑖 ) where 𝐶𝐶𝐶𝐶𝐶𝐶𝑠𝑠𝑠𝑠,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for same direction crash types with KABC severity, 𝑝𝑝𝑡𝑡𝑡𝑡|𝑠𝑠𝑠𝑠 = proportion of same direction crashes that are right turn related (default value = 0.099), 𝐶𝐶𝐶𝐶𝐶𝐶𝑜𝑜𝑜𝑜,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for opposing direction crash types with KABC severity, 𝑝𝑝𝑡𝑡𝑡𝑡|𝑜𝑜𝑜𝑜 = proportion of opposing direction crashes that are right turn related (default value = 0.029), 𝐶𝐶𝐶𝐶𝐶𝐶𝑖𝑖𝑖𝑖,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for intersecting direction crash types with KABC severity, 𝑝𝑝𝑡𝑡𝑡𝑡|𝑖𝑖𝑖𝑖 = proportion of intersecting direction crashes that are right turn related (default value = 0.191), 𝐶𝐶𝐶𝐶𝐶𝐶𝑡𝑡𝑡𝑡,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for right-turn-related crashes with KABC severity (default value = 0.625 = 1/1.6), and 𝑛𝑛 = number of signalized intersection approaches for which right-turn-on-red is prohibited. The CMFs for same, opposing, and intersecting direction crash types with KABCO severity are computed using the following equations: 𝑛𝑛 𝐶𝐶𝐶𝐶𝐶𝐶𝑠𝑠𝑠𝑠,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝑠𝑠𝑠𝑠|𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑠𝑠𝑠𝑠,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 + �1 − 𝑝𝑝𝑠𝑠𝑠𝑠|𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 � × �𝑝𝑝𝑡𝑡𝑡𝑡|𝑠𝑠𝑠𝑠 × �𝐶𝐶𝐶𝐶𝐶𝐶𝑡𝑡𝑡𝑡,𝑂𝑂 � + (1 − 𝑝𝑝𝑡𝑡𝑡𝑡|𝑠𝑠𝑠𝑠 )� 𝑛𝑛 𝐶𝐶𝐶𝐶𝐶𝐶𝑜𝑜𝑜𝑜,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝑜𝑜𝑜𝑜|𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑜𝑜𝑜𝑜,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 + �1 − 𝑝𝑝𝑜𝑜𝑜𝑜|𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 � × �𝑝𝑝𝑡𝑡𝑡𝑡|𝑜𝑜𝑜𝑜 × �𝐶𝐶𝐶𝐶𝐶𝐶𝑡𝑡𝑡𝑡,𝑂𝑂 � + (1 − 𝑝𝑝𝑡𝑡𝑢𝑢|𝑜𝑜𝑜𝑜 )� 𝑛𝑛 𝐶𝐶𝐶𝐶𝐶𝐶𝑖𝑖𝑖𝑖,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = 𝑝𝑝𝑖𝑖𝑖𝑖|𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 × 𝐶𝐶𝐶𝐶𝐶𝐶𝑖𝑖𝑖𝑖,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 + �1 − 𝑝𝑝𝑖𝑖𝑖𝑖|𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 � × �𝑝𝑝𝑡𝑡𝑡𝑡|𝑖𝑖𝑖𝑖 × �𝐶𝐶𝐶𝐶𝐶𝐶𝑡𝑡𝑡𝑡,𝑂𝑂 � + (1 − 𝑝𝑝𝑡𝑡𝑡𝑡|𝑖𝑖𝑖𝑖 )� where 𝐶𝐶𝐶𝐶𝐶𝐶𝑠𝑠𝑠𝑠,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for same direction crash types with KABCO severity, 𝑝𝑝𝑠𝑠𝑠𝑠|𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = proportion of KABC crashes that are same direction, 𝐶𝐶𝐶𝐶𝐶𝐶𝑜𝑜𝑜𝑜,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for opposing direction crash types with KABCO severity, 𝑝𝑝𝑜𝑜𝑜𝑜|𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = proportion of KABC crashes that are opposing direction, 𝐶𝐶𝐶𝐶𝐶𝐶𝑖𝑖𝑖𝑖,𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = CMF for intersecting direction crash types with KABCO severity, M-75 

𝑝𝑝𝑖𝑖𝑖𝑖|𝐾𝐾𝐾𝐾𝐾𝐾𝐾𝐾 = proportion of KABC crashes that are intersecting direction, and 𝐶𝐶𝐶𝐶𝐶𝐶𝑡𝑡𝑡𝑡,𝑂𝑂 = CMF for right-turn-related crashes with PDO severity (default value = 0.909 = 1/1.1); Table 100. Recommended UrbArt-3SG Right Turn on Red Prohibition Adjustment Factors Crash Type Crash AF Severity All KABCO Use equation above KABC Use equation above KAB Use equation above KA Use equation above Single Vehicle KABCO 1.00 KABC 1.00 KAB 1.00 KA 1.00 Same Direction KABCO Use equation above KABC Use equation above KAB Use equation above KA Use equation above Opposing Direction KABCO Use equation above KABC Use equation above KAB Use equation above KA Use equation above Intersecting Direction KABCO Use equation above KABC Use equation above KAB Use equation above KA Use equation above The recommended adjustment factors could not be validated since the dataset used for calibration did not have any variation for right turn on red prohibition, i.e., all sites in the calibration dataset (that met the base conditions for all other variables) met the base condition of right turn on red allowed on all approaches. M-76