Safety Impacts of Intersection Sight Distance (2018)

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

NCHRP OVERV The proje to input b data colle 17-59 APP IEW OF M ct team use oth office a ction metho ENDIX A ETHODOL d the data co nd field data dology for Fi – DATA CO OGY llection for into the da elements in gure A-1. D A-1 LLECTIO m shown in tabase. The addition to a ata Collect N METHO Figure A-1 following su vailable ISD ion Form. DOLOGY to documen bsections c and ISD q t field data a haracterize t uality.   nd he

NCHRP OFFICE All crash in the off and Geog notewort Site Loca To link th defined i each site Figure A Function County/r team buil of the ma private ro type “rur “urban lo 17-59 COLLECT data elemen ice. Three m raphic Infor hy data colle tion Data e team-com n terms of H . -2 presents a al Class an oute/milepo t queries in jor route, an utes, which al local” wa cal” was as ION ts and seve ajor data so mation Sys ction practi piled data w SIS variable n example o F d Area Typ st was used MS Access d area type are not incl s assigned to signed to the ral roadway urces were tem (GIS) m ces are outli ith the HSI s. This equ f how the p igure A-2. e to link site d to obtain the (i.e., rural o uded in the the minor r minor road A-2 and traffic/o used, includ aps, as show ned in the fo S data, the l ated to a cou roject team NCDOT G ata with the functional r urban). Ho HSIS databa oads in rura s in urban a perations d ing the HSI n in Table llowing sec ocation of e nty-route-m extracted da IS Map. roadway in class of the wever, som se. In this s l areas and reas. ata elements S database, G 5. The deta tions. ach study in ilepost com ta from GIS ventory file minor route e minor roa ituation, the the function were collec oogle Eart ils of tersection w bination for map in NC . The projec , functional ds are local functional c al class type ted h, as . t class or lass

NCHRP Access D Access d within 0. data colle F Intersect The proje discovere chosen o selection horizonta photos co Figure A intersecti ft.) The c research angle me angle exp data colle 17-59 ensity ensity was i 25 miles of ctors additi igure A-3. ion Angle ct team use d that the m n each leg re of the distan l curve start uld vary de -4 illustrates ons. The rad enter of the team decide asurements. erienced by ction teams dentified as the subject i onally ident Using Goog d a quantitat easured ang presenting t ce has little ing near the pending on this issue w ius of the c circle is loc d to use a 75 This distan the first few to follow. the number ntersection u ified the pre le Earth R ive measure le between he end poin impact on t intersection how far alon ith Google ircle represe ated at the in ft. distance ce ensured t drivers in A-3 of intersecti sing Googl sence of sig uler Tool fo ment of ske adjacent leg ts of the arc he measured , the angle m g the leg th Earth screen nts the four tersection c from the po hat angle me a queue and ons or drive e Earth, as i nalized inter r Determin w in the ana s could vary . For legs w skew angl easuremen e data collec shots for th lengths exp enter. Based int of inters asurements provided a ways along llustrated in sections. ing Assess lysis. Howe , depending ithout horizo e. However, t extracted f tor chose as ree of the st lored (50, 75 on this init ection for a most closel consistent m the major ro Figure A-3 Density. ver, it was on the poin ntal curvatu for a leg wi rom the aer the arc poin udy , 100, and 1 ial test, the ll intersectio y match the ethod for al ute . The ts re, th a ial ts. 50 n l

NCHRP Figure A measure used the intersecti degrees, angle. 17-59 Figure -5 provides intersection ruler tool to on skew ang the differenc A-4. Measu a more detai angles. A G measure the le is the dif e between t red Angle w led example oogle Earth heading of ference betw he result and A-4 ith 50ft, 75 of how the image was each leg fro een the two 180 degree ft, 100ft, a research tea used as the b m true north headings. I s was used nd 150ft off m used the ase map. T . As a resul f the result i as the inters set. 75 ft. distan he data colle t, the s more than ection skew ce to ctor 90

NCHRP Fi FIELD M Horizont The proje level rath determin informati presence the follow 1. L 2. R 3. R le 4. N Vertical A Smart performe of six val “+” and “ respectiv 17-59 gure A-5. A EASURE al Curvatu ct team dete er than quan e a radius fo on in creatin of horizonta ing designa – Indicates – Indicates ev – Indicat ft and one to – Indicates Grade Level was u d at 100 ft., ues were re -“ indicatin ely. ngle Meas MENTS re rmined that titative, as r each horiz g a better m l curvature tions: that there is that there is es that there the right). that there is sed to meas 250 ft., and corded for e g uphill and urements U the value o it was decid ontal curve odel. Thus within the li a horizontal a horizonta are reverse no horizont ure the verti 500 ft. from ach site. In a downhill gr A-5 sing Googl f horizontal ed that the ti would not ju , each data c mits of ISD curve, and l curve, and curves (i.e., al curve wit cal grade fo the intersec ddition, the ades approa e Earth Im curve inform me and reso stify the ex ollection te looking to t it bends to t it bends to t at least two hin the limi r each site. T tion on both “+” and “-“ ching the su age and Ru ation was a urces neces pected bene am will simp he left and t he left. he right. curves—on ts of ISD. he measure approache signs were bject interse ler Tools. t the qualita sary to fits of this ly note the o the right u e bending to ments were s. Hence, a t recorded, w ction, tive sing the otal ith

NCHRP 17-59 A-6 Traffic Volume Collection The methodology for collecting and extrapolating the traffic volume varied across the three states. North Carolina The major roads for NC sites included in the study are all state-maintained roads. Thus, the project team acquired annual AADT for the major roads from the HSIS database. For sites that were missing a year of AADT, the AADT of a previous and a subsequent year were used to interpolate the missing year’s AADT. Minor road traffic volumes often required more time and effort to collect than major road traffic volumes. Generally, traffic volume data for minor roads came from one of three sources:  The NCDOT GIS AADT map, if available.  Traffic counts from local agencies, including the cities of Cary, Asheville, and Raleigh.  Two-hour traffic counts conducted by the NC data collection team during the morning or afternoon peak periods. Once the project team obtained all the short-term traffic counts, a comprehensive methodology was utilized to convert the short-term traffic counts to AADT, as discussed in the AADT Conversion section. To match these AADTs to each year of crash data, minor road AADT was extrapolated using NC annual growth factors from 2009 to 2012. Since the growth factor for 2013 is not published, and the statewide growth factors are ready from January to August, the average of the monthly growth factors was used to extrapolate the traffic volume collected in 2013 to 2012. Ohio With some exceptions, major roads for Ohio sites were state-maintained roads, while the majority of minor roadways were non-state maintained. The annual AADT for the state- maintained roadways were retrieved from the HSIS database. AADTs for non-state-maintained roadways were obtained from various sources, which included (in prioritized order):  Ohio DOT GIS AADT/VMT roadway file.  Traffic counts conducted by or on behalf of the metropolitan planning organization or local agency.  Short duration traffic counts conducted by the OH data collection team using video cameras or Jamar pneumatic tube counters. Volumes collected from local, county, and the state agencies were typically reported as AADT. The short duration data collected by the OH team and any non-AADT data provided by MPOs or local agencies was converted using a similar methodology to that described in the AADT Conversion section using the appropriate seasonal and/or hourly adjustment factors provided by

NCHRP 17-59 A-7 the OH DOT. Similar to the North Carolina data, non-state maintained AADT data was converted to the appropriate year using annual growth factors provided by the OH DOT. Washington The project team collected traffic volumes in WA using four sources:  WSDOT GIS Traffic Counts Map for state-maintained roads.  Traffic counts conducted by or on behalf of local and county agencies.  Short duration traffic counts conducted by the Washington data collection team using Pico 2500 and MetroCount 5600 pneumatic tube counters.  Contracted traffic counts conducted by Quality Counts in Portland, OR and Seattle, WA. Volumes collected from local, county, and the state agencies were assumed to be AADT. Minor volumes collected by Quality Counts or the WA data collection team were converted from short duration counts using a similar methodology described in the AADT Conversion section using growth factors provided by the WSDOT. AADT Conversion Methodology Short-term counts were conducted in NC, OH, and WA for the sites without AADTs. This appendix depicts the methodology used in NC and was also adapt by the data collection teams in OH and WA. In NC, the team obtained short-term traffic counts for 16 sites from the local agencies. For the sites without any available traffic data, the team conducted 2-hr counts during the morning or afternoon peak period. The short-term counts were then converted to AADT using a control station. There were four scenarios for converting the short counts to AADT. These scenarios are discussed individually as follows:  Scenario 1: Site with 24-hr counts for the minor road from local agencies.  Scenario 2: Site with 2-hr peak period counts for the minor road collected by the project team and 24-hr counts for the major road from local agencies.  Scenario 3: Site with 2-hr peak period counts for the minor road collected by the project team, but without available 24-hr counts for the major road.  Scenario 4: Site with 72-hr counts for the minor road from local agencies. General Formula The Traffic Monitoring Guide (FHWA 2001) provides the following formula (Equation A-1) to convert 24-hr axle counts to AADT: AADT୦୧ ൌ VOL୦୧ ൈ M୦ ൈ D୦ ൈ A୧ ൈ G୦ (A-1) where AADThi = the annual average daily travel at location i of factor group h. VOLhi = the 24-hr axle volume at location i of factor group h. Mh = the applicable seasonal (monthly) factor for factor group h.

NCHRP 17-59 A-8 Dh = the applicable day-of-week factor for factor group h (if needed). Ai = the applicable axle-correction factor for location i (if needed). Gh = the applicable growth factor for factor group h (if needed). According to the mythology provided in Traffic and Highway Engineering (Garber and Hoel 2009), which uses hourly, daily, and monthly expansion factors to covert counts of durations shorter than 24 hours and the project team collected 2-hr vehicle counts instead of axle counts, this formula was modified. In general, a 2-hr traffic volume count is converted to AADT with Equation A-2: AADT୦୧ ൌ VOL୦୧୨ ൈ H୧୨ ൈ M୦ ൈ D୦ (A-2) where AADThi = the annual average daily travel at location i of factor group h. VOLij= the 2-hr peak volume during time period j at location i of factor group h. Hij= the applicable hourly factor during time period j at location i. M h= the applicable seasonal (monthly) factor for factor group h. Dh = the applicable day-of-week factor for factor group h. Methodology For each site, the nearest Automatic Traffic Recorder (ATR) station is used as the control station to establish the volume variation pattern for the conversion. The short-term traffic counts on the major road are also used to help establish the volume variation pattern. Furthermore, NCDOT has separate factor groupings for interstate routes and non-interstate routes based on the traffic characteristic, as shown in Table A-1. In this study, only non-interstate ATR groups were utilized.

NCHRP 17-59 A-9 Table A-1. ATR Factor Groups. Non-Interstate ATR Groups Group 1: The most dominant group in the State. Mostly rural in nature and is predominantly used for count locations on nonurban primary routes and all rural and most urban secondary roads. Group 2: This group is generally applied to count locations on primary routes (US and NC routes) that provide regional access to recreational areas. These facilities are typically located in the eastern and western areas of the State. Group 3: Generally assigned to roads in the coastal and mountain regions. This group is characterized by recreational land use and areas subject to greater seasonal variation. Factors from this group are predominantly applied to primary routes (US and NC routes) located at or within recreational areas. Some local routes located in the mountain region have this group assignment. Group 4: Predominantly found in areas where land use is characterized as urban, with dense, mixed development. Factors from this group are predominantly applied to urban primary routes and higher volume secondary and local routes in large urban areas. Group 5: Primarily confined to the I-95 corridor and routes directly influenced by I-95 traffic. Group 6: Primarily found in the mountain region of the state. Used at locations strongly influenced by recreational land uses with high seasonal variation. Group 7: Limited to the extreme eastern coastal routes (Outer Banks). Characterized by very high seasonal variation. Interstate ATR Groups Group 11: Applies to urban interstate and some rural locations strongly influenced by nearby large urban areas. Group 12: The most prominent interstate group in the State. Rural in nature and is predominantly used for nonurban locations not affected by recreational travel. Group 13: Applied to the I-95 corridor due to its unique seasonal pattern. Group 14: Applied to I-77 north of I-40 and the northern end of I-85. Somewhat similar to the I-95 pattern but much more extreme due to the low volumes experienced on those interstate locations. The four different scenarios are discussed in the following sections to better illustrate methods used to convert the short counts to AADT. Scenario 1: Site with 24-hr counts for the minor road from local agencies. Site Description The major road is Millbrook Road in Raleigh, Wake County, NC, intersecting with Sweetbriar Street. Sweetbriar Street had available 24-hr traffic volume data obtained from the City of

NCHRP 17-59 A-10 Raleigh, which was 909 vehicles per day (vpd) and was collected on Wednesday, June 2, 2010. In this case, since 24-hr counts are available and there is no need to use hourly factor, the formula is simplified as shown in Equation A-3: AADT୦୧ ൌ VOL୦୧ ൈ M୦ ൈ D୦ (A-3) where AADThi = the annual average daily travel at location i of factor group h. VOLhi= the 24-hr volume at location i of factor group h. M h = the applicable seasonal (monthly) factor for factor group h. Dh = the applicable day-of-week factor for factor group h. Step 1: Determine the factor group. Group 1 was selected since Sweetbriar Street is a local route in an urban area where the land use is characterized as residential. Step 2: Identify the day-of-week factor and monthly factor. NCDOT provides a cross tabulation by day of week by month for each ART group, as shown in Table A-2. The seasonal factor (Mh × Dh) is 0.97 since the data were collected on Wednesday, June 2, 2010. Table A-2. Seasonal Factor by Day of Week by Month (Group 1). ATR Group Month Mon Tues Wed Thurs Fri Avg Wk Day 1 1 1.11 1.14 1.10 1.09 0.99 1.09 1 2 1.07 1.07 1.06 1.02 0.92 1.03 1 3 1.04 1.05 1.05 0.99 0.94 1.01 1 4 1.01 0.99 0.98 0.95 0.87 0.96 1 5 1.00 0.97 0.95 0.93 0.82 0.93 1 6 0.96 0.96 0.97 0.93 0.83 0.93 1 7 0.95 0.96 0.93 0.96 0.85 0.93 1 8 0.97 0.97 0.97 0.93 0.83 0.93 1 9 1.00 0.98 0.99 0.94 0.83 0.95 1 10 0.99 0.99 0.96 0.93 0.81 0.94 1 11 1.02 1.00 0.97 1.03 0.89 0.98 1 12 1.03 1.05 1.06 1.02 0.91 1.01 Step 3: Calculate AADT with the applicable factors. AADT୦୧ ൌ VOL୦୧ ൈ M୦ ൈ D୦ ൌ 909 ൈ 0.97 ൌ 880 vpd ∗ *The final number is rounded to the nearest 10 vpd.

NCHRP 17-59 A-11 Scenario 2: Site with 2-hr peak period counts for the minor road collected by the project team and 24-hr counts for the major road from local agencies. Site Description The major road is Millbrook Road, which is located in an urban area of Raleigh, NC. It intersects with Newberry Drive. Table A-3 shows the 2-hr traffic count conducted from 4 PM to 6 PM on Wednesday, February 27, 2013. Table A-3. Peak Hour Traffic Volume on Newberry Drive. Site 1: Newberry Dr @ E Millbrook Rd Time NB SB Total 16:00 0 4 4 16:15 9 2 11 16:30 2 1 3 16:45 4 2 6 Hourly Total 15 9 24 17:00 7 0 7 17:15 4 2 6 17:30 3 4 7 17:45 6 1 7 Hourly Total 20 7 27 Step 1: Convert peak hour traffic count to daily traffic volume. The 24-hr traffic counts on the Millbrook Road were collected by the City of Raleigh on June 13, 2012, as shown in Table A-4. The project team calculated the hourly expansion factors using Equation A-4 (Garber and Hoel 2009): Hourly Expansion Factor ൌ ୲୭୲ୟ୪ ୴୭୪୳୫ୣ ୤୭୰ ଶସି୦୰ ୮ୣ୰୧୭ୢ୴୭୪୳୫ୣ ୤୭୰ ୮ୟ୰୲୧ୡ୳୪ୟ୰ ୦୭୳୰ (A-4)

NCHRP 17-59 A-12 Table A-4. Hourly Expansion Factors on Millbrook Road. Time EB WB Total Hourly Expansion Factor 0:00 60 64 124 156.37 1:00 24 18 42 461.67 2:00 22 23 45 430.89 3:00 22 22 44 440.68 4:00 12 24 36 538.61 5:00 52 57 109 177.89 6:00 130 213 343 56.53 7:00 497 644 1141 16.99 8:00 621 591 1212 16.00 9:00 513 485 998 19.43 10:00 478 450 928 20.89 11:00 630 516 1146 16.92 12:00 782 636 1418 13.67 13:00 751 671 1422 13.64 14:00 646 673 1319 14.70 15:00 755 699 1454 13.34 16:00 767 752 1519 12.76 17:00 898 1014 1912 10.14 18:00 699 695 1394 13.91 19:00 464 505 969 20.01 20:00 360 337 697 27.82 21:00 296 281 577 33.60 22:00 162 176 338 57.37 23:00 121 82 203 95.52 Total 9762 9628 19390 1.00 The daily traffic volume on Newberry Drive was estimated using the hourly factors for 4 PM and 5 PM, as highlighted in the Table A-4: VOL୦୧ ൌ 24 ൈ 12.76 ൅ 27 ൈ 10.142 ൌ 290 vpd

NCHRP 17-59 A-13 Step 2: Repeat steps 1-3 from scenario 1 to calculate AADT. Group 1 is selected as the factor group since the route is located in an urban area, where the land use is characterized as residential. Since the data were collected on Wednesday, February, the seasonal factor (Mh × Dh) is 1.06. AADT୦୧ ൌ VOL୦୧ ൈ M୦ ൈ D୦ ൌ 290 ൈ 1.06 ൌ 310 vpd ∗ *The final number is rounded to the nearest 10 vpd. Scenario 3: Site with 2-hr peak period counts for the minor road collected by the project team, but without available 24-hr counts for the major road. Site Description US 70, which is located in a rural area in Johnston County, NC, intersects with Ole Rock Quarry Road. There are no 24-hr counts for the major and the minor road. The project team counted the traffic from 4 PM to 6 PM on Thursday, June 20, 2013, as shown in Table A-5. Table A-5. Peak Hour Traffic Volume on Ole Rock Quarry Rd. Site: Old Rock Quarry Rd Time NB SB Total 16:00 4 7 11 16:15 7 7 14 16:30 10 11 21 16:45 15 7 22 Hourly Total 36 32 68 17:00 12 10 22 17:15 10 4 14 17:30 14 9 23 17:45 16 11 27 Hourly Total 52 34 86 Step 1: Convert peak hour traffic counts to daily traffic volume. The project team used ATR station A9501, which is the closest station to the site, to obtain the hourly factor. Table A-6 shows the average hourly volume for the year of 2012 and the corresponding hourly factor.

NCHRP 17-59 A-14 Table A-6. Station A9501 Hourly Expansion Factors. Time Hourly Volume Hourly Expansion Factor 0:00 26 134.67 1:00 17 202.78 2:00 18 195.39 3:00 17 205.84 4:00 31 113.67 5:00 53 66.94 6:00 124 28.60 7:00 180 19.66 8:00 220 16.08 9:00 196 18.00 10:00 214 16.54 11:00 227 15.60 12:00 238 14.88 13:00 244 14.51 14:00 270 13.08 15:00 185 19.12 16:00 293 12.06 17:00 276 12.82 18:00 245 14.46 19:00 138 25.54 20:00 116 30.50 21:00 86 41.22 22:00 75 47.32 23:00 48 74.04 Total 3536 1.00 Then the daily traffic volume on Ole Rock Quarry Rd was estimated using the hourly factors for 4 PM and 5 PM, as highlighted in Table A-6: VOL୦୧ ൌ 68 ൈ 12.06 ൅ 86 ൈ 12.822 ൌ 961 vpd

NCHRP 17-59 A-15 Step 2: Repeat steps 1-3 from scenario 1 to calculate AADT. Group 1 is selected as the factor group since the route is located in a rural area. Because the data were collected on Thursday, June 20, 2013, the seasonal factor (Mh × Dh) is 0.93. AADT୦୧ ൌ VOL୦୧ ൈ M୦ ൈ D୦ ൌ 961 ൈ 0.93 ൌ 900 vpd ∗ *The final number is rounded to the nearest 10 vpd. Scenario 4: Site with 72-hr counts for the minor road from local agencies. Site Description The site is located in the city of Asheville, NC with mountainous terrain. The major road is Broadway Street, which intersects with Hillside Street. The project team was able to obtain the traffic data from the city, which was conducted from October 11 to 13, 2011, as shown in Table A-7. Step 1: Determine the factor group. Since the site is located in the mountainous region and the city of Asheville is influenced by recreational land uses with high seasonal variation, group 6 was selected for the day of week and monthly factors.

NCHRP 17-59 A-16 Table A-7. 72-hr Traffic Counts on Hillside Street. Time Tuesday Wednesday Thursday Average Hourly Expansion Factor 12:00 8 7 8 8 147.52 1:00 2 7 8 6 199.59 2:00 3 1 5 3 377.00 3:00 5 3 4 4 282.75 4:00 6 3 5 5 242.36 5:00 10 4 6 7 169.65 6:00 19 21 24 21 53.02 7:00 63 71 71 68 16.55 8:00 65 73 90 76 14.88 9:00 66 75 59 67 16.97 10:00 59 49 66 58 19.50 11:00 44 61 63 56 20.20 12:00 67 69 81 72 15.64 1:00 59 74 71 68 16.63 2:00 92 94 87 91 12.43 3:00 80 107 75 87 12.95 4:00 99 96 98 98 11.58 5:00 97 98 84 93 12.16 6:00 76 73 75 75 15.15 7:00 50 49 61 53 21.21 8:00 34 42 43 40 28.51 9:00 38 28 31 32 34.98 10:00 30 25 32 29 39.00 11:00 16 15 13 15 77.11 Total 1,088 1,145 1,160 1,131 1.00 Step 2: Identify the day-of-week factor and monthly factor. Table A-8 shows the seasonal factors for group 6. Since the traffic volume of the minor road was conducted for 72 hours from Tuesday to Thursday, Average Week Day Factor of October (factor=0.6) was selected for the conversion.

NCHRP 17-59 A-17 Table A-8. Seasonal Factor by Day of Week by Month (Group 6). ATR Group Month Sun. Mon. Tues. Wed. Thurs. Fri. Sat. Avg Wk Day 6 1 1.76 3.84 4.11 4.14 3.90 2.45 1.64 3.55 6 2 1.30 3.43 3.50 3.71 2.83 1.83 1.09 2.86 6 3 1.04 1.85 1.84 1.85 2.36 1.43 1.01 1.82 6 4 0.77 1.48 1.63 1.41 1.53 0.91 0.75 1.33 6 5 0.85 1.36 1.31 1.37 1.33 0.99 0.83 1.25 6 6 0.58 0.85 0.81 0.80 0.81 0.71 0.56 0.79 6 7 0.54 0.69 0.64 0.65 0.62 0.53 0.49 0.62 6 8 0.65 0.92 0.88 0.83 0.87 0.75 0.57 0.85 6 9 0.67 1.14 1.10 1.27 0.98 0.77 0.56 1.02 6 10 0.51 0.65 0.69 0.66 0.55 0.48 0.46 0.60 6 11 0.69 1.59 1.94 1.69 1.74 0.98 0.75 1.50 6 12 2.00 3.16 2.82 3.28 3.38 2.22 1.68 2.90 Step 3: Calculate AADT with the applicable factors. AADT୦୧ ൌ VOL୦୧ ൈ M୦ ൈ D୦ ൌ 1131 ൈ 0.6 ൌ 680 vpd* *The final number is rounded to the nearest 10 vpd. REFERENCES FHWA. Traffic Monitoring Guide, Federal Highway Administration, Washington, D.C., May, 2001 [Online]. Available: http://www.fhwa.dot.gov/ohim/tmguide/tmg3.htm. Garber, J. N. and L.A. Hoel. Traffic and Highway Engineering, 4th Edition, 2009.