Multimodal Level of Service Analysis for Urban Streets: Users Guide (2008)

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.

Multimodal Level of Service for Urban Streets Page 40 VI. EXAMPLE PROBLEMS This section provides example problems illustrating the application of the above described methodologies for estimating the performance and level of service for urban streets. EXAMPLE PROBLEM 1 – DETERMINE LOS OF STREET Objective: To determine the perceived level of service (LOS) for all modal users of the urban street with a given configuration and demand. Data: The study length of the urban street is 1 mile long. Within this study section, the street is generally a 4 lane arterial with parking and sidewalk (see below). Traffic signals are present at each intersection. The street carries between 10,000 and 15,000 average annual daily traffic (AADT). The specific segment data Exhibit 18. Example Problem 1 Plan and Cross-section 11’ 11’11’ 11’8’5’ 5’8’ 70 ft ROW Bus Stop varies Key Issues: This sample problem illustrates the computation of level of service for a given street design and a given demand level. Approach: 1. Determine starting and ending points of study section of street. –For convenience, the beginning intersection for each direction of analysis will be excluded from the directional analysis (this causes the analysis procedure for each segment to be identical, only with different data for each segment). 2. Start evaluation of LOS for the Eastbound Direction. 3. Divvy up Eastbound Study Section into Analysis Segments – Since the demand varies between intersections, while the geometry is constant the length of the facility, the facility is divided into 5 segments. Each analysis segment is defined to start immediately

Multimodal Level of Service for Urban Streets Page 41 downstream of the beginning traffic signal and to end immediately downstream of the ending traffic signal. Thus, each analysis segment consists of a segment and a traffic signal. 4. Assemble Data – The required data is shown for each segment in the exhibits. 5. Compute Auto LOS – The required inputs and computation results are shown in Exhibit 19: Example 1 - Computation of Auto LOS. The analysis begins with the eastbound direction (Since the procedures for evaluating the westbound direction are identical, they are not shown here in this example problem). a. Compute Demand - The eastbound hourly flow rate for the peak 15 minutes of the weekday peak hour is computed by applying a peaking factor (k), a directional factor (d), and a peak hour factor (PHF) to the average weekday traffic for each segment. This step can be skipped if peak 15 minute flow rates are available. b. Compute Capacity – The eastbound through movement capacity for each segment is computed at the downstream signal for each segment. The adjusted saturation flow rate (vehicles per lane per hour of green) are computed using the procedures given in Chapter 16, Signalized Intersections of the Highway Capacity Manual (2000). The volume/capacity (v/c) ratio is computed for each segment. If the v/c ratio exceeds 1.00 then the segment is defined to be operating at LOS F. c. Compute Mean Through Speed – Although the automobile level of service is not dependent on the auto speed, this piece of information is required for the transit, pedestrian, and bicycle level of service analyses. The mean speed is computed using equations 31 and 32. d. Compute Stops and Left Lanes – The two required pieces of information for estimating auto level of service are the mean number of stops per vehicle per mile and the proportion of intersections with exclusive left turn lanes. The number of stops per vehicle is computed for each segment using equation 34. The number of stops per vehicle is divided by the segment length to obtain average stops per vehicle per mile. The proportion of intersections with left turn lanes is computed by summing the number of intersections with left turn lanes in the eastbound direction and dividing by the total number of intersections. e. Compute Auto LOS – The probability of people rating the segment according to each level of service is computed using the auto LOS model. The average level of service for each segment is computed using the special weights shown just below the table. These weights place a heavier weight on the percentage of the population that considers the street segment to operating at inferior levels of service. Once the segment LOS values are known, they are used to compute the average level of service for the eastbound direction of the street by weighting the LOS results for each segment by the relative length of each segment within the study section.

Multimodal Level of Service for Urban Streets Page 42 6. Compute Pedestrian LOS – The pedestrian LOS is computed next because it is a required input for the transit LOS. The pedestrian LOS is computed for pedestrian travel on the right-hand side of the eastbound direction of auto travel for the street. The computations are shown in Exhibit 20: Example 1 - Computation of Pedestrian LOS. a. Compute Pedestrian Density LOS – The counted (or estimated) pedestrian flow rate and sidewalk width are used to compute the pedestrian flow rate per foot width of sidewalk. The result is compared to the level of service flow rate thresholds given in Chapter 18, Pedestrians, of the Highway Capacity Manual (2000), for sidewalks. If there is no sidewalk, this step is skipped. b. Compute Pedestrian Segment LOS – The pedestrian segment LOS is computed based on the geometry for the right-hand side of the street looking in the eastbound direction. Vehicle traffic volumes and speeds are obtained from the auto LOS calculation step. The average of the mid-block free-flow speed and the overall average auto travel speed is used for the purpose of the pedestrian segment LOS computation. c. Compute Pedestrian Intersection LOS – This computation requires additional data on the number of right turns on red (RTOR) and permitted left turns made by traffic leaving the subject street (in either the westbound or eastbound directions) and crossing the pedestrian crosswalk on the south side of the street while the “walk” indication is displayed for the crosswalk (or the signal is green, in the absence of pedestrian displays for the crosswalk). The average pedestrian delay waiting for the green light or “walk” indication at the intersection crosswalk is estimated using the procedure given in Chapter 18, Pedestrians of the Highway Capacity Manual (2000). d. Compute Roadway Crossing Difficulty Factor (RCDF) – The analyst inputs additional information on the distance required to cross to the opposite curb for the subject street, and the green/cycle (“walk”) ratio for pedestrians crossing the subject street at the downstream signalized intersection. The “Flashing Don’t Walk” period should not be included in the computed g/c ratio for pedestrians crossing the street. If there is a median in the center of the street at least 6 feet wide, then the walking distance to the median can be used for the crossing distance, rather than the full curb-to-curb street width. The traffic volume used for this step includes both direction of travel on the subject street (unless a wide median is present, in which case only traffic in the eastbound direction should be used). e. Compute Pedestrian Facility LOS – The minimum of the roadway crossing wait delay, or the divert to signal delay is used to computed the crossing LOS. The crossing delay is compared delay thresholds in Chapter 18, Pedestrians, of the Highway Capacity Manual (2000), to obtain the crossing LOS. The pedestrian non-density LOS is computed (without the RCDF factor) based on the previously computed segment and intersection LOS. The difference between these the crossing LOS and the non-density LOS is used to determine the value of the RCDF. The RCDF multiplied by the initial non-density LOS become the final pedestrian non-density LOS. The worst of the pedestrian density and non-density LOS’s becomes the pedestrian facility LOS for each segment. The pedestrian facility LOS’s for each of the segments are averaged based on relative segment lengths, to obtain overall pedestrian facility LOS for the eastbound direction.

Multimodal Level of Service for Urban Streets Page 43 7. Compute Transit LOS – The data and computation steps are shown in Exhibit 21: Example 1 - Computation of Transit LOS. a. Input Data – The frequency of transit service, on-time performance, bus stop amenities (percent with shelters, percent with benches), and load factor are entered for the eastbound direction for each segment. Each segment must be characterized as central business district (CBD) or not. A CBD is a segment of street where the base travel rate of bus service (against which passengers measure their perceived LOS) is 15 mph. This generally is the case for dense street networks with high rise development. b. Compute Mean Bus Speed – The average speed of bus service is computed based on the mean auto speed and expected bus stop dwell time for each segment, following the procedure provided in Chapter 27, Transit of the Highway Capacity Manual (2000). c. Compute Perceived Travel Time Rates – The various perceived travel time rates are computed following the formulae given in this users guide. d. Compute Transit LOS – The transit level of service for the eastbound direction of travel for each segment is computed following the formulae given in this users guide. The segment LOS results are averaged based on relative lengths to obtain the facility LOS for the eastbound direction of travel. 8. Compute Bicycle LOS a. Geometric Input Data – Much of the same geometric input data required for the auto and pedestrian LOS analyses are used for the bicycle LOS analysis. New data items consist of whether or not the street has a median (is divided), the width of cross- streets, and the number of unsignalized intersections and major driveways per mile. b. Performance and Other Input Data – The auto volumes and speed are obtained from the auto LOS analysis. The on-street parking percentage is the same as was used in the pedestrian analysis. The percent heavy vehicles and the pavement rating are new data items required for the bicycle LOS analysis. c. Compute Bicycle LOS – The effective width (between the bicyclist and vehicle traffic) is computed first based upon the vehicle volume and then based on other factors to obtain the final “effective width”. The mean vehicle speed is converted to a “speed factor” (The speed factor is fixed at 0.8103 for vehicle speeds less than 21 mph). Note that the parking lane is treated as a shoulder lane (for bicycle LOS computation purposes) only in segment #5 where the parking occupancy is 0%. The segment, intersection, and combined bicycle LOS for each segment are computed using the formulae given in this users guide. The length weighted average bicycle LOS for the eastbound direction of the facility is computed based on the relative segment lengths. 9. Summary Table of LOS Results – The summary table of modal LOS results is given in Exhibit 23: Example 1 – Summary of Results 10. The above analysis is then repeated for the reverse direction of travel on the street.

Multimodal Level of Service for Urban Streets Page 44 Exhibit 19: Example 1 - Computation of Auto LOS 1. Compute Eastbound Hourly Demand (v) Weekday Peak Factor Dir. Factor Pk.Hr.Fac. Demand ADT k d PHF v Segment (vpd) (#) (#) (#) (vph) 1 10,000 0.080 0.55 0.92 478 2 15,000 0.080 0.55 0.92 717 3 10,000 0.080 0.55 0.92 478 4 15,000 0.080 0.55 0.92 717 5 10,000 0.080 0.55 0.92 478 2. Compute Eastbound Hourly Capacity and V/C Adjusted Thru Lanes Thru Capacity V/c V/c Saturation One-Dir. (g/C) Check Segment (vphgl) (#) (#) (vph) 1 1500 2 0.50 1500 0.32 OK 2 1500 2 0.50 1500 0.48 OK 3 1650 2 0.45 1485 0.32 OK 4 1650 2 0.45 1485 0.48 OK 5 1650 2 0.44 1452 0.33 OK Adjusted Sat. Flows computed per Chapter 16, Signalized Intersections. 3. Compute Mean Through Speed Free Segment Cycle Progress. Intersect Average Speed Length Length Quality Delay Speed Segment (mph) (ft) (sec) (#) (sec) (mph) 1 35 600 60 5 3.5 26.9 2 35 600 60 5 4.2 25.7 3 35 1200 90 3 16.5 20.5 4 35 1200 120 3 24.2 17.2 5 35 1680 120 3 22.6 20.7 Total/Ave. 5280 20.7 4. Compute Stops & % Left Lane Progress. Stops Segment Stops Left Ln Quality Per Veh. Length Per Mile (Yes=1) Segment (#) (stps/veh) (ft) (stps/mi) (0,1) 1 5 0.41 600 3.65 0 2 5 0.44 600 3.88 0 3 3 0.61 1200 2.71 0 4 3 0.66 1200 2.88 0 5 3 0.62 1680 1.94 0 Total/Ave 2.74 5280 2.74 0.00 5. Compute Auto LOS LOS A LOS B LOS C LOS D LOS E LOS F Weight. Auto Segment (%) (%) (%) (%) (%) (%) Ave. LOS 1 11.1% 31.5% 26.8% 16.2% 9.1% 5.3% 2.97 C 2 10.5% 30.7% 26.9% 16.8% 9.5% 5.6% 3.01 C 3 13.6% 34.8% 25.7% 14.1% 7.5% 4.2% 2.80 C 4 13.1% 34.2% 25.9% 14.5% 7.8% 4.4% 2.83 C 5 16.1% 37.2% 24.4% 12.4% 6.3% 3.5% 2.66 B Average 13.5% 34.7% 25.7% 14.2% 7.5% 4.3% 2.80 C weights: 1 2 3 4 5 6

Multimodal Level of Service for Urban Streets Page 45 Exhibit 20: Example 1 - Computation of Pedestrian LOS 1. Compute Pedestrian Density LOS Sidewalk Ped Ped. Ped. Density LOS Lookup Seg. Width Flow Density Density Ped/hr/ft LOS (ft) (pph) LOS # LOS 0 A 1 5 4000 4.12 D 300 B 2 5 1500 1.84 B 420 C 3 5 1000 1.27 A 600 D 4 5 500 0.65 A 900 E 5 5 50 0.07 A 1380 F 2. Compute Pedestrian Segment LOS Outside Bike/Shlder On-Street Barrier Buffer Dir. Traffic Traf. Lnes Midblock Ped. Seg. Lane Lane Width Parking Width Volume One-Dir Traf Spd Seg. (ft) (ft) (%) (Y/N) (ft) (vph) (lanes) (mph) LOS # 1 12 8 100% No 0 478 2 31.0 1.97 2 12 8 50% No 0 717 2 30.3 2.48 3 12 8 25% No 0 478 2 27.8 2.24 4 12 8 25% No 0 717 2 26.1 2.51 5 12 8 5% No 0 478 2 27.9 2.35 Midblock traffic speed = average of auto free-flow speed, and mean auto speed with intersection delay. 3. Compute Pedestrian Intersection LOS RTOR+ X-Street X-Street X-Street X-Street Ped. Right Trn Ped. Seg. Perm LT Volume PHF Speed Lanes Delay Channel Intersect (vph) (vph) (#) (mph) (#) (sec) Islands (#) LOS # 1 50 500 0.92 25 2 7.5 0 2.48 2 75 500 0.92 25 2 7.5 0 2.52 3 50 750 0.92 35 4 13.6 0 3.03 4 50 1200 0.92 40 4 18.2 2 4.12 5 50 1500 0.92 45 6 18.8 2 4.85 Pedestrian Delay computed per Chapter 18 method. 4. Compute Roadway Crossing Difficulty Factor (RCDF) Signal Signal Cross St. Divert Crossing Vehicle Vehicle Ave. Seg. Spacing Cycle g/C Delay Distance Speed Vol 2-Dir Wait (ft) (sec) (#) (sec) (ft) (mph) (vph) (sec) 1 600 60 0.17 135 64 31.0 800 421 2 600 60 0.17 135 64 30.3 1,200 2943 3 1200 90 0.11 264 64 27.8 800 425 4 1200 120 0.08 279 64 26.1 1,200 3009 5 1680 120 0.08 370 64 27.9 800 425 5. Compute Pedestrian Facility LOS Min. Crossing No Cross RCDF Ped. Ped. Ped. Ped. Seg. Wait,Divrt LOS LOS NDLOS Density Fac. LOS Facility (sec) (#) (#) (#) (#) LOS # (#) LOS 1 135 6.00 2.78 1.20 3.33 4.12 4.12 E 2 135 6.00 2.95 1.20 3.54 1.84 3.54 E 3 264 6.00 2.98 1.20 3.58 1.27 3.58 E 4 279 6.00 3.31 1.20 3.97 0.65 3.97 E 5 370 6.00 3.42 1.20 4.10 0.07 4.10 E Ave 3.89 E

Multimodal Level of Service for Urban Streets Page 46 Exhibit 21: Example 1 - Computation of Transit LOS 1. Input Data Transit On-Time Stops with Stops with Load Central Busi. Frequency Performance Shelter Bench Factor District Segment (bus/h) (%) (%) (%) (p/seat) (Y/N) 1 18 48% 100% 100% 1.1 Yes 2 18 48% 100% 100% 1.2 Yes 3 9 50% 0% 100% 0.8 No 4 4 65% 0% 0% 0.7 No 5 2 70% 0% 0% 0.5 No 2. Compute Mean Bus Speed Length Auto Spd Bus Stops Delay/Stop Ave Bus Segment (ft) (mph) (#) (sec) (mph) 1 600 26.9 1 20 11.6 2 600 25.7 1 20 11.4 3 1200 20.5 1 15 14.9 4 1200 17.2 1 15 13.1 5 1680 20.7 1 10 17.5 Total/Ave 5280 20.7 14.2 Average Bus Speed computed per Chapter 27 method. 3. Compute Transit Perceived Travel Time and Headway Factors a1 IVTTR EWTTR ATR PTTR Fptt Fh Segment factor min/mi min/mi min/mi 1 1.41 5.16 1.83 0.41 10.52 0.70 3.75 2 1.60 5.27 1.83 0.41 11.66 0.67 3.75 3 1.00 4.02 1.69 0.05 7.35 0.92 3.46 4 1.00 4.59 0.83 0.00 6.24 0.98 2.83 5 1.00 3.42 0.61 0.00 4.64 1.11 1.97 IVTTR = In-Vehicle Travel Time Rate EWTTR = Equivalent Wait Travel Time Rate ATR = Amenity Time Rate PTTR = Perceived Travel Time Rate Fptt = Perceived Travel Time Factor Fh = Headway Factor 4. Compute Transit LOS Wait/Ride Ped LOS Transit Segment Score LOS Score LOS 1 2.61 4.12 2.71 B 2 2.52 3.54 2.75 B 3 3.19 3.58 1.75 A 4 2.79 3.97 2.42 B 5 2.19 4.10 3.34 C Average 2.63 B Note: Shaded entries are input data fields

Multimodal Level of Service for Urban Streets Page 47 Exhibit 22: Example 1 - Computation of Bicycle LOS 1. Geometric Input Data Outside Bike/Shldr Through Divided/ Sig. Int Unsig. Conf Lane Width Lane Width Lanes Undivided Cross-Dist Per Mile Segment (ft) (ft) (lanes) (D/UD) (ft) (conf/mi) 1 12 0 2 UD 40 0.0 2 12 0 2 UD 40 10.0 3 12 0 2 UD 64 5.0 4 12 0 2 UD 64 2.0 5 12 0 2 UD 88 1.0 2. Performance and Other Input Data Traffic Heavy Midblock On-Street Pavement Volume Vehicle Traffic Spd Parking Rating Segment (vph) (%) (mph) (%) (#) 1 478 3% 31.0 100% 4.0 2 717 6% 30.3 50% 4.0 3 478 12% 27.8 25% 3.5 4 717 12% 26.1 25% 4.0 5 478 6% 27.9 5% 4.0 Pavement Rating: 1=Poor, 5=Excellent Midblock traffic speed = average of auto free-flow speed and mean auto speed with intersection delay. 3. Compute Bicycle LOS Prelim. Effective Speed Segment Intersect Bicycle Bike Eff. Width Width Factor LOS LOS Score LOS Segment (Wv) (We) (#) (#) (#) (#) 1 12.0 2.0 3.49 4.52 2.62 3.72 E 2 12.0 7.0 3.43 5.10 2.85 4.21 E 3 12.0 9.5 3.11 6.14 2.99 4.23 E 4 12.0 9.5 2.84 5.94 3.21 4.14 E 5 12.0 11.5 3.12 4.32 3.36 3.89 E Average 4.04 E Note: Shaded entries are input data fields Exhibit 23: Example 1 – Summary of Results Direction = Eastbound Auto Transit Bicycle Pedestrian Segment LOS LOS LOS LOS 1 C B E E 2 C B E E 3 C A E E 4 C B E E 5 B C E E Facility C B E E

Multimodal Level of Service for Urban Streets Page 48 EXAMPLE PROBLEM 2 – DETERMINE LOS IMPACTS OF CONVERTING FROM 4-LANE TO 3-LANE CROSS-SECTION Objective: For a given set of modal demands, determine the impacts of converting a street from 4 auto lanes to 2 lanes plus two-way-left-turn-lane. Exhibit 24: Example Problem 3 Plan and Cross-Section Views 10’ 12’5’ 5’8’5’ 5’8’ 70 ft ROW 12’ Bus Stop varies Key Issues: The key issue in this example is the tradeoff between the benefits to the bicycles of providing the bicycle lane and the reduction of through lanes for autos and buses from 4 lanes to 2 lanes plus a two-way left turn lane. Since this configuration puts more vehicle traffic closer to the sidewalk (by concentrating two-lanes worth of traffic into a single lane in each direction), a 5- foot wide landscaped buffer strip with trees has been added between the curb and the sidewalk to mitigate this potential impact. The parking lanes on both sides were eliminated on both sides to make room for the landscaped buffer strips. The addition of bicycle lanes benefits bicyclists, however; the concentration of all vehicle traffic in a single lane in each direction, closer to bicyclists, is a disbenefit for them. A level of service analysis is required to see if the proposed new cross-section has a positive, negative, or neutral impact on auto drivers, transit passengers, bicyclists and pedestrians. Approach: The same analysis approach is used as was used in Example Problem #1. The points below highlight the changes in the inputs and results. 1. Determine starting and ending points of study section of street. – Use the same limits as for Example Problem 1. 2. Start evaluation of LOS for the Eastbound Direction. 3. Divvy up Eastbound Study Section into Analysis Segments – Use same segments as for Example Problem 1. 4. Assemble Data – The same demand data is used as in Example Problem #1, however; the geometric cross section data is changed to reflect the new proposed cross section. 5. Compute Auto LOS – The required inputs and computation results are shown in Exhibit 26: Example 2 - Computation of Auto LOS.

Multimodal Level of Service for Urban Streets Page 49 a. Compute Demand – Same as Problem 1. b. Compute Capacity –The adjusted saturation flow rate (vehicles per lane per hour of green) are computed using the procedures given in Chapter 16, Signalized Intersections, of the Highway Capacity Manual (2000), but this time with no left turn adjustment, because left turns can now be made in the left turn lane. The number of through lanes is reduced to 1 lane. The v/c check identifies one segment that will be over capacity. c. Compute Mean Through Speed – Same procedure with different input data as in Problem #1. d. Compute Stops and Left Lanes – Same procedure with different input data as in Problem #1. e. Compute Auto LOS – Same procedure with different input data as in Problem #1. The overall facility LOS is “F” because one of the segments had a v/c ratio greater than 1. This overrides the normal averaging of segment LOS (which would have resulted in an average LOS D). 6. Compute Pedestrian LOS – The computations are shown in Exhibit 27: Example 2 - Computation of Pedestrian LOS. a. Compute Pedestrian Density LOS – Same procedure with different input data as in Problem #1. b. Compute Pedestrian Segment LOS – Same procedure with different input data as in Problem #1. c. Compute Pedestrian Intersection LOS – Same procedure with different input data as in Problem #1. d. Compute Roadway Crossing Difficulty Factor (RCDF) – Same procedure with different input data as in Problem #1. Note that pedestrian crossing distance is only to the median, because the two-way-left-turn is considered (in this example jurisdiction) to function as a de-facto median in this case. Other jurisdictions may consider the two-way-left-turn-lane to not provide a refuge for crossing pedestrians, in which case the full curb-to-curb street width would be used to compute the RCDF. e. Compute Pedestrian Facility LOS – Same procedure with different input data as in Problem #1. 7. Compute Transit LOS – The data and computation steps are shown in Exhibit 28: Example 2 - Computation of Transit LOS. a. Input Data – Same input data as in Problem #1. b. Compute Mean Bus Speed – Same procedure as in Problem #1 with different input data (auto speed from the auto LOS computation). c. Compute Perceived Travel Time Rates – Same procedure as in Problem #1 with different input data (bus speed and pedestrian LOS). d. Compute Transit LOS – Same procedure as in Problem #1 with different input data. 8. Compute Bicycle LOS – The data and computation steps are shown in Exhibit 29: Example 2 - Computation of Bicycle LOS. a. Geometric Input Data – Same data as in Problem #1, except for the road is now considered “divided”. b. Performance and Other Input Data – Same procedure as in Problem #1 with different input data (auto speed from auto LOS computation). c. Compute Bicycle LOS – Same procedure as in Problem #1 with different input data. 9. Summary Table of LOS Results – The summary table of modal LOS results is given in Exhibit 30. Example Problem 2 Results Summary

Multimodal Level of Service for Urban Streets Page 50 10. The above analysis is then repeated for the reverse direction of travel on the street. • Conclusions: The results are summarized in Converting 4 traffic lanes into 2 bike lanes, 2 traffic lanes, and a two-way-left-turn- lane (TWLTL) (and adding a pedestrian buffer zone between the sidewalk and the curb) caused auto LOS to deteriorate on two of the five segments, but overall facility LOS for auto remained unchanged by the conversion. • The conversion improved bicycle LOS on one of the five segments but did not significantly affect overall facility LOS for bicycles. This is because the benefits of adding a bike lane and prohibiting parking were partially balanced out by the increase in vehicle traffic in the right lane adjacent to the bicycle lane. • The conversion worsened transit LOS for two of the five segments, but had no significant impact on overall transit LOS. • The conversion significantly improved pedestrian LOS. The addition of a landscaped buffer strip with trees more than made up for the loss of on-street parking (from the point of view of pedestrians walking on the street). Exhibit 25: Impacts of Lane Conversion. The key conclusions are highlighted below. • Converting 4 traffic lanes into 2 bike lanes, 2 traffic lanes, and a two-way-left-turn-lane (TWLTL) (and adding a pedestrian buffer zone between the sidewalk and the curb) caused auto LOS to deteriorate on two of the five segments, but overall facility LOS for auto remained unchanged by the conversion. • The conversion improved bicycle LOS on one of the five segments but did not significantly affect overall facility LOS for bicycles. This is because the benefits of adding a bike lane and prohibiting parking were partially balanced out by the increase in vehicle traffic in the right lane adjacent to the bicycle lane. • The conversion worsened transit LOS for two of the five segments, but had no significant impact on overall transit LOS. • The conversion significantly improved pedestrian LOS. The addition of a landscaped buffer strip with trees more than made up for the loss of on-street parking (from the point of view of pedestrians walking on the street). Exhibit 25: Impacts of Lane Conversion Auto LOS Transit LOS Bicycle LOS Pedestrian LOS Segment Before After Before After Before After Before After 1 C C B C E D E E 2 C D B C E E E D 3 C C A A E E E C 4 C C B B E E E E 5 B C C C E E E D Facility C C B B E E E D

Multimodal Level of Service for Urban Streets Page 51 Exhibit 26: Example 2 - Computation of Auto LOS 1. Compute Eastbound Hourly Demand (v) Weekday Peak Factor Dir. Factr Pk.Hr.Fac Demand ADT k d PHF v Segment (vpd) (#) (#) (#) (vph) 1 10,000 0.080 0.55 0.92 478 2 15,000 0.080 0.55 0.92 717 3 10,000 0.080 0.55 0.92 478 4 15,000 0.080 0.55 0.92 717 5 10,000 0.080 0.55 0.92 478 2. Compute Eastbound Hourly Capacity and V/C Adjusted Thru Lanes Thru Capacity v/c v/c Check Saturation One-Dir. (g/C) Segment (vphgl) (#) (#) (vph) 1 1550 1 0.50 775 0.62 OK 2 1550 1 0.50 775 0.93 OK 3 1700 1 0.45 765 0.63 OK 4 1700 1 0.45 765 0.94 OK 5 1700 1 0.44 748 0.64 OK Adjusted Sat. Flows computed per Chapter 16, Signalized Intersections. 3. Compute Mean Through Speed Free Segment Cycle Progress. Intersect Average Speed Length Length Quality Delay Speed Segment (mph) (ft) (sec) (#) (sec) (mph) 1 35 600 60 5 6.4 22.6 2 35 600 60 5 11.8 17.4 3 35 1200 90 3 21.8 18.1 4 35 1200 120 3 39.2 13.1 5 35 1680 120 3 29.3 18.5 Total/Ave. 5280 17.0 4. Compute Stops & % Left Lane Progress. Stops Segment Stops Left Trn Ln Quality Per Veh. Length Per Mile (Yes=1) Segment (#) (stps/veh) (ft) (stps/mi) (0,1) 1 5 0.48 600 4.23 1 2 5 0.90 600 7.94 1 3 3 0.72 1200 3.16 1 4 3 1.26 1200 5.53 1 5 3 0.73 1680 2.28 1 Total/Ave 4.08 5280 4.08 1.00 5. Compute Auto LOS LOS A LOS B LOS C LOS D LOS E LOS F Weight. Auto Segment (%) (%) (%) (%) (%) (%) Ave. LOS 1 13.2% 34.3% 25.9% 21.7% 7.7% 4.4% 3.11 C 2 5.6% 20.5% 25.8% 30.6% 15.6% 10.5% 3.87 D 3 16.6% 37.6% 24.1% 18.5% 6.1% 3.4% 2.89 C 4 9.8% 29.5% 27.1% 25.5% 10.1% 6.0% 3.39 C 5 19.9% 39.8% 22.2% 20.9% 5.1% 2.7% 2.91 C Average 13.6% 34.8% 25.7% 14.2% 7.5% 4.3% 2.80 C weights: 1 2 3 4 5 6

Multimodal Level of Service for Urban Streets Page 52 Exhibit 27: Example 2 - Computation of Pedestrian LOS 1. Compute Pedestrian Density LOS Sidewalk Ped. Ped. Ped. Density LOS Lookup Seg. Width Flow Density Density Ped/hr/ft LOS (ft) (pph) LOS # LOS 0 A 1 5 4000 4.12 D 300 B 2 5 1500 1.84 B 420 C 3 5 1000 1.27 A 600 D 4 5 500 0.65 A 900 E 5 5 50 0.07 A 1380 F 2. Compute Pedestrian Segment LOS Outside Bike/Shouldr On-Street Barrier Buffer Dir. Traf Lanes Midblock Ped. Seg. Lane Lane Width Parking Width Volume One-Dir Traf Spd Seg. (ft) (ft) (%) (Y/N) (ft) (vph) (lanes) (mph) LOS 1 12 5 0% Yes 8 478 1 28.8 2.20 2 12 5 0% Yes 8 717 1 26.2 2.77 3 12 5 0% Yes 8 478 1 26.5 2.15 4 12 5 0% Yes 8 717 1 24.0 2.72 5 12 5 0% Yes 8 478 1 26.7 2.16 Note that Midblock traffic speed is average of: auto free-flow speed and mean auto speed with int. delay. 3. Compute Pedestrian Intersection LOS RTOR+ X-Street X-Street X-Street X-Street Ped Right Trn Ped. Seg. Perm LT Volume PHF Speed Lanes Delay Channel Intersect (vph) (vph) (#) (mph) (#) (sec) Islands (#) LOS # 1 50 500 0.92 25 2 7.5 0 2.48 2 75 500 0.92 25 2 7.5 0 2.52 3 50 750 0.92 35 4 13.6 0 3.03 4 50 1200 0.92 40 4 18.2 2 4.12 5 50 1500 0.92 45 6 18.8 2 4.85 Pedestrian Delay computed per Chapter 18 method. 4. Compute Roadway Crossing Difficulty Factor (RCDF) Signal Signal Cross St. Divert Crossing Vehicle Vehicle Ave. Seg. Spacing Cycle g/C Delay Distance Speed Vol 2-Dir Wait (ft) (sec) (#) (sec) (ft) (mph) (vph) (sec) 1 600 60 0.17 135 20 28.8 800 15 2 600 60 0.17 135 20 26.2 1,200 35 3 1200 90 0.11 264 20 26.5 800 15 4 1200 120 0.08 279 20 24.0 1,200 35 5 1680 120 0.08 370 20 26.7 800 15 5. Compute Pedestrian Facility LOS Min. Crossing No Cross RCDF Ped. Ped. Ped. Ped. Seg. Wait,Divrt LOS LOS NDLOS Density Fac. LOS Facility (sec) (#) (#) (#) (#) LOS # (#) LOS 1 15 2.00 2.85 0.89 2.53 4.12 4.12 E 2 35 4.00 3.04 1.13 3.43 1.84 3.43 D 3 15 2.00 2.96 0.87 2.58 1.27 2.58 C 4 35 4.00 3.38 1.08 3.66 0.65 3.66 E 5 15 2.00 3.36 0.82 2.75 0.07 2.75 D Ave 3.15 D

Multimodal Level of Service for Urban Streets Page 53 Exhibit 28: Example 2 - Computation of Transit LOS 1. Input Data Transit On-Time Stops with Stops with Load . Frequency Perform. Shelter Bench Factor CBD Segment (bus/h) (%) (%) (%) (p/seat) (Y/N) 1 18 48% 100% 100% 1.1 Yes 2 18 48% 100% 100% 1.2 Yes 3 9 50% 0% 100% 0.8 No 4 4 65% 0% 0% 0.7 No 5 2 70% 0% 0% 0.5 No 2. Compute Mean Bus Speed Length Auto Spd Bus Stops Delay/Stop Ave Bus Segment (ft) (mph) (#) (sec) (mph) 1 600 22.6 1 20 10.7 2 600 17.4 1 20 9.4 3 1200 18.1 1 15 13.6 4 1200 13.1 1 15 10.6 5 1680 18.5 1 10 15.9 Total/Ave 5280 17.0 12.4 Average Bus Speed computed per Chapter 24 method. 3. Compute Transit Perceived Travel Time and Headway Factors a1 IVTTR EWTTR ATR PTTR Fptt Fh Segment factor min/mi min/mi min/mi 1 1.41 5.59 1.83 0.41 11.12 0.68 3.75 2 1.60 6.38 1.83 0.41 13.43 0.64 3.75 3 1.00 4.42 1.69 0.05 7.74 0.90 3.46 4 1.00 5.69 0.83 0.00 7.34 0.92 2.83 5 1.00 3.77 0.61 0.00 4.99 1.08 1.97 IVTTR = In-Vehicle Travel Time Rate EWTTR = Equivalent Wait Travel Time Rate ATR = Amenity Time Rate PTTR = Perceived Travel Time Rate Fptt = Perceived Travel Time Factor Fh = Headway Factor 4. Compute Transit LOS Wait/Ride Ped LOS Transit Segment Score LOS Score LOS 1 2.56 4.12 2.78 C 2 2.42 3.43 2.89 C 3 3.13 2.58 1.70 A 4 2.61 3.66 2.63 B 5 2.12 2.75 3.23 C Average 2.65 B

Multimodal Level of Service for Urban Streets Page 54 Exhibit 29: Example 2 - Computation of Bicycle LOS 1. Geometric Input Data Outside Bike/Shldr Through Divided/ Sig. Int Unsig.Conf Lane Width Lane Width Lanes Undivided Cross-Dist Per Mile Segment (ft) (ft) (lanes) (D/UD) (ft) (conf/mi) 1 12 5 1 D 40 0.0 2 12 5 1 D 40 10.0 3 12 5 1 D 64 5.0 4 12 5 1 D 64 2.0 5 12 5 1 D 88 1.0 2. Performance and Other Input Data Traffic Heavy Midblock On-Street Pavement Volume Vehicle Traffic Spd Parking Rating Segment (vph) (%) (mph) (%) (#) 1 478 3% 28.8 0% 4.0 2 717 6% 26.2 0% 4.0 3 478 12% 26.5 0% 3.5 4 717 12% 24.0 0% 4.0 5 478 6% 26.7 0% 4.0 Pavement Rating: 1=Poor, 5=Excellent Midblock traffic speed = average of: auto free-flow speed, and mean auto speed with intersection delay. 3. Compute Bicycle LOS Prelim. Effective Speed Segment Intersect Bicycle Bicycle Eff. Width Width Factor LOS LOS Score LOS Segment (Wv) (We) (#) (#) (#) (#) 1 17.0 22.0 3.25 2.38 2.00 3.31 D 2 17.0 22.0 2.86 2.98 2.45 3.80 E 3 17.0 22.0 2.91 4.33 2.37 3.84 E 4 17.0 22.0 2.37 3.86 2.82 3.72 E 5 17.0 22.0 2.95 2.82 2.73 3.51 E Average 3.64 E Exhibit 30. Example Problem 2 Results Summary Direction =Eastbound Auto Transit Bicycle Pedestrian Segment LOS LOS LOS LOS 1 C C D E 2 D C E D 3 C A E C 4 C B E E 5 C C E D Facility C B E D