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 5 II. METHODOLOGY This section provides a general overview of the methodologies for estimating multimodal level of service and for estimating additional modal performance measures such as travel time, speed, delay, stops, and queuing for an urban street. MULTIMODAL LEVEL OF SERVICE ANALYSIS FRAMEWORK Level of service (LOS) is used to translate complex numerical performance results into a simple letter grade system representative of the travelers’ perception of the resulting quality of service provided by the facility. The letter grade level of service hides much of the complexity of facility performance in order to simplify decision-making regarding whether or not facility performance is generally acceptable and whether or not a change in this performance is likely to be perceived as significant by the general public. Level of service is a quantitative stratification of quality of service into six letter grades with letter grade “A” representing the “best” quality of service, and letter grade “F” representing the “worst” quality of service. “Best” and “Worst” are left undefined, allowing the traveling public to self-identify the “best” and “worst” conditions based on their own experience and perceptions for each individual situation. Research indicates that the traveling public perceives less than six levels of service, however the six letter-grade (A-F) system has been retained to provide agencies with additional thresholds of performance with which to analyze performance. In addition, while there is a wide range in the levels of service reported by the traveling public for any given condition, this LOS methodology reports only a single, weighted average letter grade for the given condition. This is designed to simplify the task of agency decision-making. Instead of reporting a probability distribution of LOS grades for a facility, the LOS methodology reports a single representative letter grade LOS for the facility. The methodology provides for the estimation of a separate mean level of service for each of four modes of travel on the urban street: auto driver, bus passenger, bicyclist, and pedestrian. Other modes of travel; commercial vehicle, truck driver, auto passenger, recreational travel, and messenger/delivery service are not covered by this methodology. The methodology does not provide for the computation of an overall weighted average of the LOS results across the four modes of travel. It enables the analyst to see the changes in LOS from one mode to the other as changes are made to the design and operation of the urban street. Weighing the trade- offs of improving the LOS for one mode versus worsening it for another mode are left to the analyst and the public agency operating the urban street. The Highway Capacity Manual (2000) has historically relied upon a single performance measure to predict level of service. Research indicates however that the traveling public takes into account several factors in evaluating the quality of service provided by an urban street. Consequently, this users guide presents models of level of service that combine these factors into a predicted level of service. The four modal LOS models presented below output numerical ratings, which must be converted into the traditional A-F letter grade system. Exhibit 1: LOS Letter Grade Numerical Equivalents, is used to convert the numerical outputs into letter grades. Level of service is a quantitative stratification of quality of service into 6 levels of service.

Multimodal Level of Service for Urban Streets Page 6 Exhibit 1: LOS Letter Grade Numerical Equivalents LOS Model Outputs LOS Letter Grade Model <=2.00 A 2.00 < Model <= 2.75 B 2.75 < Model <= 3.50 C 3.50 < Model <= 4.25 D 4.25 < Model <= 5.00 E Model > 5.00 F Notes: 1) If any directional segment hourly volume/capacity ratio (v/c) exceeds 1.00 for any mode, that direction of street is considered to be operating at LOS F for that mode of travel for its entire length (regardless of the computed level of service). 2) If the movement of any mode is legally prohibited for a given direction of travel on the street, then the level of service for that mode is LOS “F” for that direction. Division of Street Into Analysis Segments The portion of the street to be evaluated is defined as the study section of the street. Each direction of travel on the street is evaluated separately. The LOS estimation methods require that the demand, control, and geometry of the study section of the street be relatively uniform within the analysis segment. Since demand, control and geometric conditions are rarely uniform over the length of a street, it is usually necessary to divide the study section of the street into segments. Each segment consists of a piece of the street (usually between two intersections) where the demand, control, and geometric conditions are uniform. Demand includes vehicular traffic, transit ridership, and pedestrian flow rates. Segments should be selected so as to ensure that all three of these modal demands are relatively constant within a segment. Bicycle flows do not currently enter into the LOS computations and can be allowed to vary within a segment if only an LOS analysis is to be performed. However, if bicycle performance measures are to be computed and bicycle flow rates are high enough to influence bicycle operations, then the segments should be selected so that bicycle flows are also relatively constant within the segment. Control includes posted speed limits, traffic signals, stop signs, traffic calming devices, and other devices intended to influence vehicle speeds. Segments should be selected so that control devices that slow or stop traffic (such as signals, all-way stops, and roundabouts) are located at the end points of an analysis segment. Geometry includes number of lanes, shoulders, parking lanes, sidewalks, bicycle lanes, planter strips, medians, etcetera. The road cross-section (lane widths, etc.) should be relatively constant within the segment. Auto Level Of Service Auto level of service is a function of the average travel speed over the length of the street and the average number of stops per mile. The auto level of service rating for an urban street is the weighted average of the sum of the probabilities of people reporting each LOS rating multiplied by a system of weights that gives greater weight to the proportion of people who perceive poorer level of service. Auto level of service is a function of stops and left turn lanes. The more stops per mile, the poorer the level of service. The more intersections with exclusive left turn lanes, the better the level of service.

Multimodal Level of Service for Urban Streets Page 7 ∑ ∗== 6 )Pr( WJLOSAutoLOS =1J J )1Pr()Pr()Pr( Equation 1 Where: Pr(LOS<=J) = Probability that an individual will respond with level of service grade of “J”. J = A, B, C, D, E, or F WJ = 1 for LOS A, 2 for B, 3 for C, 4 for D, 5 for E, 6 for F The probability of obtaining an LOS rating equal to “J” is the difference between the probability of rating the facility at LOS = J or lower and the probability of rating the facility at LOS J-1 or lower. ≤−≤== JLOSJLOSJLOS − Equation 2 The probability that a person will rate a given facility as LOS “J” or worse is given by the ordered cumulative logit model as shown below: )exp(1 1)Pr( ∑−−+=≤ xJLOS βα )( k kkJ Equation 3 Where: Pr(LOS<=J) = Probability that an individual will respond with level of service grade of “J” or worse. J = A, B, C, D, E, or F exp = Exponential function. αJ = Alpha, Maximum numerical threshold for LOS grade “J” (see Exhibit 2). βK = Beta, Calibration parameters for attributes (see Exhibit 2). XK = Attributes (k) of the facility (see Exhibit 2). Exhibit 2: Alpha and Beta Parameters for Recommended Auto LOS Model Parameter Value Alpha Values Intercept LOS E = -3.8044 Intercept LOS D = -2.7047 Intercept LOS C = -1.7389 Intercept LOS B = -0.6234 Intercept LOS A = 1.1614 Beta Values X(1) = Stops/Mile= 0.253 X(2) = Proportion of Intersections with Left Turn Lanes = -0.3434 The attribute, “stops per mile” is the number of times a vehicle decelerates to a full stop (zero mph), divided by the length of street (or segment) being evaluated. Where the measurement technology or forecasting method is not precise enough to accurately identify zero mph, then a threshold of 5 mph should be used to determine the number of full stops. When applied to the entire study length of the facility, the attribute, “proportion of intersections with left turn lanes” is the ratio of intersections with one or more exclusive left turn lanes in the direction of travel being evaluated divided by the total number of intersections within the evaluation section of the

Multimodal Level of Service for Urban Streets Page 8 street. All signalized or unsignalized intersections of public roads are counted. Private driveway intersections are not counted, unless they are signal controlled. Special Cases Treatment of Non-Uniform Street Segments The demand, geometry, and control present on any given segment should ideally be relatively uniform within the segment, however; some variation is tolerable (less than 10% of the mean). Left turn bays, right turn bays, short lane additions or drops, and other geometric changes in the vicinity of the downstream intersection of the segment do not trigger the need to divide the segments into subsegments because these geometric aspects are included in the estimation of intersection v/c ratio and stops. If there is a sudden demand increase or decrease in the middle of the segment, such as might occur at a large parking garage, then the segment should be divided into two (or more) subsegments, each with its own appropriate demand level. Generally, a reduction in through lanes in the middle of a segment, such as might occur at a bridge, would trigger the need to subdivide the segment into subsegments. If in doubt as to the seriousness of the geometric or demand variation within the segment, the segment should be divided into additional subsegments and the LOS evaluated to see if the change significantly impacts the computed LOS and facility performance. Treatment of One-Way Street Facilities If one direction of auto travel is prohibited, e.g. a one-way street, then the computations for the allowed direction of travel should proceed normally. For the prohibited direction of travel the LOS is set at “F”. Treatment of Facilities with Unsignalized Intersections For an existing conditions analysis, the number of stops per mile can be measured in the field over the study length of the facility and used to compute the LOS. If forecasting future LOS, the Highway Capacity Manual (2000) does not currently provide a method for estimating stops per vehicle. Until such a methodology becomes available, the analyst can estimate the number of stops per vehicle for each street segment approaching a stop sign at 1.00 stops per vehicle. For forecasted v/c ratios greater than 1.00, set the auto LOS at “F”. The number of stops per vehicle for stop signs is added to the estimated number of stops per vehicle for each of the traffic signals and divided by the study length to obtain the average number of stops per mile for the study length of street. For street segments approaching all other unsignalized intersections (where the approaching traffic is not required to stop), the number of stops per vehicle should be set to zero as long as the intersection approach volume/capacity ratio is below 1.00. For volume/capacity ratio values equal to or greater than 1.00, the segment of street approaching the intersection should be set at LOS “F”.

Multimodal Level of Service for Urban Streets Page 9 Treatment of Bus Lanes and Bus Streets In the case of bus streets, the auto LOS is, by definition, LOS “F” (since autos cannot access this street). The transit and pedestrian LOS are computed normally, with transit vehicles being the only motorized vehicles on the street. If bicycles are allowed in the bus street then bicycle LOS is computed normally, otherwise, it is set to LOS “F” for bicycles. In the case of bus lanes, the auto, transit, bicycle, and pedestrian LOS analyses proceed normally. The only difference is that only transit vehicles (and carpools or taxis, if allowed) are assigned to the bus lane. Treatment of Railroad Crossings The LOS methodology is not designed to account for the impacts of railroad crossings with frequent train traffic. If train frequencies are less than 1 per hour, their impacts of perceived auto level of service can be neglected. Transit Level of Service The transit level of service (LOS) is based on a combination of the access experience, the waiting experience, and the ride experience. The access experience is represented by the pedestrian level of service score for pedestrian access to bus stops in the direction of travel along the street. The waiting and riding experiences are combined into a transit wait/ride score. The formula below is used to combine the various experiences into a single LOS score. Transit LOS Score = 6.0 – 1.50 * TransitWaitRideScore + 0.15 * PedLOS Equation 4 where: PedLOS =The pedestrian LOS numerical value for the direction of the facility being analyzed (A=1, F=6). TransitWaitRideScore =The transit ride and waiting time score, a function of the average headway between buses and the perceived travel time rate via bus. The computed transit level of service score is converted to a letter level of service grade using the equivalencies given in above Exhibit 1 Estimation of the Pedestrian LOS The pedestrian LOS for the urban street is estimated using the pedestrian LOS model described in a later section of this users guide. Estimation of the Transit Wait/Ride Score The transit wait/ride score is a function of the headway between buses and the perceived travel time rate via bus for the urban street. TransitWaitRideScore = fh * fptt Equation 5 Where: fh = headway factor = the multiplicative change in ridership expected on a route at a headway h, relative to the ridership at 60-minute headways; fptt = perceived travel time factor = the multiplicative change in ridership expected at a perceived travel time rate PTTR, relative to the ridership expected at a baseline travel time rate. The baseline travel time rate is 4 minutes/mile except for central business districts of metropolitan areas with over 5 million population, in which case it is 6 min/mile. Transit level of service is a function of its accessibility by pedestrians, the amenities at the bus stop, the waiting time for the bus, and the mean speed of the bus. Better pedestrian access, better shelters, more frequent bus service and higher speed bus service all improve the perceived level of service for bus transit.

Multimodal Level of Service for Urban Streets Page 10 Headway Factor The headway factor (fh) is the ratio of the estimated patronage at the prevailing average bus headway to the estimated patronage at a base headway of 60 minutes. The patronage values for the two headways (the actual or predicted headway and the base headway of 60 minutes) are computed based upon an assumed set of patronage elasticities which relate the percentage change in ridership to the percentage change in headways. Exhibit 3: Headway Factor (fh) Look-Up Table can be used to obtain Fh for a range of bus headways. It was computed using a set of assumed passenger elasticities taken from TCRP (Transit Cooperative Research Program) research. Exhibit 3: Headway Factor (fh) Look-Up Table Headway (minutes) Frequency (Bus/hr) f(h) 60 1.00 1.00 45 1.33 1.33 30 2.00 2.00 15 4.00 2.80 10 6.00 3.16 5 12.00 3.79 This table was constructed using elasticities given in Exhibit 4. If the analyst has information indicating that a different set of patronage elasticities are appropriate, then this table for fh should be reconstructed based on the recomputed patronage ratios for the actual or predicted headway and the base headway. Exhibit 4: Elasticities Used to Construct Headway Factor (Fh) Lookup Table Bus Headways Assumed Patronage Elasticity 30-60 minutes +1.0 15-30 minutes +0.5 10-15 minutes +0.3 < 10 minutes +0.2 Source: Derived from data reported in TCRP Report 95, Chapter 9 The fh values in Exhibit 3 can be approximated using the following formula: )*0239.0exp(*4 Headwayfh −= Equation 6 Where: fh = headway factor Headway = Average number of minutes between buses Perceived Travel Time Factor The perceived travel time factor is estimated based on the perceived travel time rate and the expected demand elasticity for a change in the perceived travel time rate. ( ) ( )[ ] ( ) ( )[ ]BTTReTTRe TTReBTTRe FPTTR 11 11 +−− +−−= Equation 7 Where: F(PTTR) = Perceived Travel Time Factor Note that only the buses and bus routes that actually stop to pickup or drop off passengers within the study section of the street should be included in the computation of mean bus headways for the street. Express bus service without at least one bus stop on the street would be excluded.

Multimodal Level of Service for Urban Streets Page 11 PTTR = Perceived Travel Time Rate (min/mi) BTTR = Base Travel Time Rate (min/mi) Use 6 minutes per mile for the main central business district of metropolitan areas with population greater than or equal to 5 million. Use 4 minutes per mile for all other areas. e = ridership elasticity with respect to changes in the travel time rate. The suggested default value is –0.40, but local values may be substituted. Exhibit 5 below illustrates the application of this equation for selected perceived travel time rates and a selected elasticity. Exhibit 5: Example Perceived Travel Time Factors (F(PTTR)) F(PTTR) BTTR: 4 min/mi 6 min/mi PTTR (min/mi) 2 1.31 1.50 2.4 1.22 1.41 3 1.12 1.31 4 1.00 1.17 6 0.85 1.00 12 0.67 0.76 30 0.53 0.58 Notes: • F(PTTR) = Perceived Travel Time Factor • PTTR = Perceived Travel Time Rate. • BTTR = Base Travel Time Rate (default is 4 minutes per mile. 6 minutes per mile BTTR is used for the central business districts (CBD) of metropolitan areas with 5 million or greater population). • Based on default value of –0.40 for elasticity. The perceived travel time rate (PTTR) is estimated based on the mean speed of the bus service, the average excess wait time for the bus (due to late arrivals), the average trip length, the average load factor for the bus service, and the amenities at the bus stops. PTTR= a1 * IVTTR + a2 * EWTR – ATR Equation 8 Where: PTTR = Perceived travel time rate. IVTTR = Actual in-vehicle travel time rate, in minutes per mile; (Default = 4.00) EWTR = Excess wait time rate due to late arrivals (minutes/mile) = Excess wait time/ average trip length (Default = 2.00). a1 = Passenger load weighting factor (a function of the average load on buses in the analysis segment during the peak 15 minutes) (Default = 1.00); a2 = 2 (wait time factor converting actual wait times into perceived wait times) ATR = Amenity time rate = perceived travel time rate reduction due to the provision of certain bus stop amenities

                       ! "    !#$% & "'(# %   )   (*      #%+ ( #% * *, (   * *    - *(.(+(2000) (     (*  #% * *  * #  %/0 1*                 232 22 22 4 2 5 2 1 62 3 52 44 72 1 12 6 8 % 9   :*  (  :*)22 *  #  %  /0 10*  *                      "*  9; "9#$%**    0*  *   *  ( (  /(* (  -*)(* ( < *(=   0*   **Peak bus load factors are best measured in the field for the street sections being evaluated. However, peak load factors for the entire bus route (typically collected and reported by the bus operating agency) can be used to approximate the values for the subject street segments.

                       ! "    !#$% & "'(# %   )   (*      #%+ ( #% * *, (   * *    - *(.(+(2000) (     (*  #% * *  * #  %/0 1*                 232 22 22 4 2 5 2 1 62 3 52 44 72 1 12 6 8 % 9   :*  (  :*)22 *  #  %  /0 10*  *                      "*  9; "9#$%**    0*  *   *  ( (  /(* (  -*)(* ( < *(=   0*   **Peak bus load factors are best measured in the field for the street sections being evaluated. However, peak load factors for the entire bus route (typically collected and reported by the bus operating agency) can be used to approximate the values for the subject street segments.

Multimodal Level of Service for Urban Streets Page 13 The excess wait time rate is the excess wait time (in minutes) divided by the mean passenger trip length for the bus route(s) within the study section of the street. For average passenger trip length a default value can be taken from national average data reported by the American Public Transit Association (APTA) (http://www.apta.com/research/stats/ridership/trlength.cfm ). In 2004, the mean trip length for bus passenger-trips nationwide was 3.7 miles. More locally specific values of average trip length can be obtained from the National Transit Database (NTD). Look up the annual passenger miles and annual unlinked trips in the transit agency profiles contained stored under NTD Annual Data Publications at: http://www.ntdprogram.gov/ntdprogram/pubs.htm#profiles. The mean trip length is the annual passenger-miles divided by the annual unlinked trips. If field measurements of excess wait time are not feasible, excess wait time rate can be estimated based on a transit agency’s reported “on-time rating” for the route(s), using the following equation. ( )[ ] ATL OTPLate EWTR 21* −= Equation 11 Where: EWTR = Excess Wait Time Rate (min/mi) Late = Minutes late before the agency counts a bus arrival as late. OTP = On-Time Performance. Agency reported proportion of buses arriving on-time. ATL =Average passenger trip length (miles) Amenity Time Rate The amenity time rate is the time value of various bus stop improvements divided by the mean passenger trip length. The mean passenger trip length is the same distance used to compute the Excess Wait Time Rate (described above). ATL BenchShelter ATR *2.0*3.1 += Equation 12 Where: ATR = Amenity Time Rate (min/mi) Shelter = Proportion of bus stops in study section direction with shelters Bench = Proportion of bus stops in study section direction with benches ATL = Average passenger trip length (miles) Notes: 1) Shelters with benches are counted twice, once as shelters, the second time as benches. Computation of Facility LOS From Segment LOS The level of service is computed separately for each direction of travel on each analysis segment. The segment levels of service (for a given direction of travel) are combined into an overall directional level of service for the study section of street by taking a length weighted average of the segment levels of service for the analysis direction. Note that the analyst must use the same definition of “Late” as was used by the transit operator in defining on-time performance. Consult with Transit Capacity and Quality of Service Manual for additional information on estimating on-time performance.

Multimodal Level of Service for Urban Streets Page 14 ∑ ∑= )( )(*)( )( iL iLiLOS facilityLOS Equation 13 Where: LOS (facility) = LOS for subject direction of facility LOS (segment) = LOS is subject direction for each segment (i) L (segment) = Length of each segment (i) Special Cases Treatment of Gaps in Transit Service The portions of street where there is no transit service should be split into their own segments for the purpose of transit LOS analysis (if not already split for other reasons). The transit LOS should be set at “F” for these segments. The rest of the transit LOS analysis proceeds normally, with the overall transit LOS being a length-weighted average including the segments with no transit service. No Through Transit Service Full Length of Study Section If a passenger must transfer one or more times to continue on the same street, then the transfer waiting time should be added into the total travel time used to compute the mean speed of the bus service on the street. If some routes travel the length of the street while others do not, the analyst may compute a weighted average travel time across all of the bus routes including transfer waiting times. The weighted average travel time is divided into the analysis street length to obtain the weighted average mean speed for bus service on the street. Single Direction Transit Service on Two-Way Street The direction of travel for which there is no transit service can be assigned a transit LOS “F”. The other direction of travel is evaluated normally. For cases where transit service is provided as a one-way loop, the analyst can compute the excess time and distance to go around the loop to reach the upstream destination and enter the information into the perceived travel time rate computation for the direction of travel without direct service. Treatment of Bus Lanes and Bus Streets The methodologies are not specifically designed to handle bus streets and bus lanes, but with some judicious adjustments, they can be adapted to these special situations. In the case of bus streets, the auto LOS is, by definition, LOS “F” (since autos cannot access this street). The transit, bicycle, and pedestrian LOS are computed normally, with transit vehicles being the only motorized vehicles on the street. In the case of bus lanes, the auto, transit, bicycle, and pedestrian LOS analyses proceed normally. The only difference is that only transit vehicles (and carpools, if allowed) are assigned to the bus lane. Treatment of Railroad Crossings The LOS methodology is not designed to directly account for the impacts of railroad crossings on transit LOS. Since transit vehicles usually must stop at all railroad crossings, the crossings will reduce the speed of bus service. If the analyst can estimate the added delay due to stops for railroad crossings and stops for trains, then this information can be used to estimate the effect on bus speeds on the street. The speeds can be used to estimate the transit LOS for the street with railroad crossings. Infrequent delays due to

                                         ! "           #     !"#$! %&'() $  %& '       (" '" ) *   )   # %&+ '%       , ') )  -    * ##*) ) )  -       )     **  ))   ) )*  )        #("    )        #       The segment bicycle LOS is calculated according to the following equation:  !* +,)-./ 00/1 $.+ #  *,-" # 2!3 #  * %&'(! $ . & '&      #  '/)  01 '20) 1  )2 *)3#43 ))  ')    ) 5 '6   -  )# /.5789019 1 '(   '#44+:3;3#<3=  '%  )   -   /.7': 05 '      -  )# /. )) >:33*?05)   @)  )>'3? )     )   )   05  # , '10$%!  )     '("* ' % )=)   " $ '%   )  ) '$3"?99+$>8 '$;$:3"?99  ? '  )    2  $ '   )      $ '(  )   ) '$99+57A3       '$9:3#33"599  $ '$ )  )) )   )  2     /. 2 ) )   3?)  #   %                      )   #+  !      "   *  *   )  )     ))   &  # &   )                 * 2     *  @) #  0   )*      *              #  %   2        # Bicycle Segment LOS is a function of the perceived separation between motor vehicle traffic and the bicyclist, parked vehicle interference, and the quality of the pavement. Higher vehicle volumes, higher percent heavy vehicles, and higher vehicle speeds decrease the perceived separation. A striped bike lane increases the perceived separation. **Lightly used driveways, such as residential driveways, should generally be excluded from the driveway and unsignalized intersection counts used to compute the number of conflicts per mile. Unsignalized merges (where side street traffic does not have to stop before entering the arterial), such as might occur at the foot of a freeway off-ramp, should be given much greater weight in the computation of conflicts per mile. The degree of weighting is at the discretion of the analyst.

Multimodal Level of Service for Urban Streets Page 16 Bicycle Intersection LOS The intersection bicycle LOS is calculated according to the following equation: Bint = -0.2144Wt + 0.0153CD + 0.0066 (V/(4*PHF*L)) + 4.1324 Equation 16 Where: Bint = bicycle intersection score Wt = total width of outside through lane and bike lane (if present) on study direction of street (ft). CD = The curb-to-curb width of the cross-street at the intersection (ft). V = Volume of directional traffic (vph) L = Total number of through lanes on the subject approach to the intersection Computation of Facility LOS From Segment LOS The level of service is computed separately for each direction of travel on each analysis segment. The segment levels of service (for a given direction of travel) are combined into an overall directional level of service for the study section of street by taking a length weighted average of the segment levels of service for the analysis direction. ∑ ∑= )( )(*)( )( iL iLiLOS facilityLOS Equation 17 Where: LOS (facility) = LOS for subject direction of facility LOS (segment) = LOS is subject direction for each segment (i) L (segment) = Length of each segment (i) Special Cases Treatment of Sections With Significant Grades The bicycle level of service equations are designed for essentially flat grades (grades of under 2% of any length). For steeper grades the analyst should consider applying an adjustment to the LOS estimation procedure to account for the negative impact of both up-grades and down-grades on bicycle level of service. This adjustment probably should be sensitive both to the steepness of the grade and its length. However, research available at the time of production of this manual did not provide a basis for computing such an adjustment. It is left to the discretion of the analyst. Treatment of Sections with Parallel Bike/Ped Path Auto and transit LOS are computed as for a standard street segment. The bicycle LOS is computed for both bicycles using the street and for bicycles using the parallel path. If the analyst has information on the split of bicycles using the parallel path and the street, then the resulting path and street LOS’s are combined into a single LOS for bicycles based on the percent using each. BLOS = %Using Street *BLOS(Street) + %Using Path * BLOS(Path) Equation 18 Where: BLOS (Street) = Bicycle LOS for bicycles riding in the traveled way of the street. BLOS (Path) = Bicycle LOS for bicycles using the parallel path.

               ! " # $ (2000)%      ,   , and HIgh Bus Volumes The bicycle LOS methodology is not designed to adequately represent bicyclist perceptions of level of service when they are operating on streets with frequent bus service with frequent stops requiring bicyclists to frequently swerve left to pass stopped buses. The analyst may choose to impose a weighting factor on the busvolume to better reflect the greater impact of the stopping buses on bicyclist level of service. The weighting factor would be at the analyst's discretion.          !  !   %     ! ! ! ! % !    %     !  % !! ! !  #!  # ! %        & '  ( ! !    ! % ) ( ! !#*#*#   +) ( ! !  !     +  , -. /# ,/ 0 #  !#1  %  !   !  ! ! #*%2  ##*   !#*%   3!-#4# #! !  ! %2!!!    ! ! .%            ! " # $ (2000)%  #*#   !! #*#         ,)5 56$2+2+75..5 285 &9:&9;&2) "/% !#     ,<+2+ 566< 5)-.&5)5 629+.922 25/  %  !50 -= !  " # $ (2000)%+  !# 0  %50 #1 !#   !! %

Multimodal Level of Service for Urban Streets Page 18 Exhibit 7: Pedestrian Density LOS (DPLOS) LOS Minimum Sidewalk Space Per Person Equivalent Maximum Flow Rate per Unit Width of Sidewalk A > 60 SF per person <= 300 peds/hr/ft B >40 <= 420 C >24 <= 600 D >15 <= 900 E >8 <= 1380 F <= 8 SF > 1380 Adapted from Exhibit 18-3 This analysis is performed separately for each side of the street that has a sidewalk. Pedestrian Non-Density LOS Model (Ped NDLOS) The pedestrian LOS for the facility that is representative of non-density factors is computed according to the equation below. NDPLOS = (0.318 PSeg + 0.220 PInt + 1.606) * (RCDF) Equation 20 Where NDPLOS = Pedestrian non-density (other factors) LOS PSeg = Pedestrian segment LOS value PInt = Pedestrian intersection LOS value RCDF = Roadway crossing difficulty factor The output of this model is a numerical value, which must be translated to an LOS letter grade. Exhibit 1 above provides the numerical ranges that coincide with each LOS letter grade. A separate pedestrian segment LOS analysis is conducted for each side of the street.

fLV LV PLOS = -1.2276 ln (fLV x Wt + 0.5Wl + fp x %OSP + fb x Wb + fsw x Ws) + ) Pedestrian level of service score for a segment Natural log (AADT) is less than or equal to 4,000, in which case fLV=(2 - 0.00025 * AADT) total width of outside lane (and shoulder) pavement Width of shoulder or bicycle lane, or, if there is un-striped parking and %OSP=25 then Wl=10 ft. to account for lateral displacement of Percent of segment with on-street parking 5.37 for any continuous barrier at least 3 feet high separating trees, bollards, etc.) can be considered a continuous barrier if they are at least 3 feet high and are spaced 20 feet on center or less. feet)*** s=10, otherwise fsw = 3.00) Width of widewalk For widths greater than 10 feet, use 10 feet. Directional volume of motorized vehicles in the direction closest to the pedestrian (vph) Peak hour factor pedestrians. Ped SegLOS = ln = fLV = Wt = Wl = fp = %OSP = fb = = Wb = fsw = Ws = V = PHF = L = SPD = Where Multimodal Level of Service for Urban Streets ** **This pedestrian LOS method has not been designed for nor tested for application to rural highways and other roads where a sidewalk is not present and the traffic volumes are low but the speeds are high. For these situations a satisfactory pedestrian level of service may not accurately reflect pedestrian perceptions. ***In cases where street furniture, planter pots, and tree wells occupy the portion of the sidewalk between the pedestrians and the street (such as often occurs in central business districts with wide sidewalks), this portion of the sidewalk can be counted as a buffer strip, even though it is paved.

Multimodal Level of Service for Urban Streets Page 20 Roadway Crossing Difficulty Factor The pedestrian Roadway Crossing Difficulty Factor (RCDF) measures the difficulty of crossing the street between signalized intersections. The RCDF worsens the pedestrian LOS if the crossing difficulty is worse than the non- crossing LOS for the facility. It improves the pedestrian LOS if the crossing difficulty LOS is better than the non-crossing difficulty LOS. The factor is based on the numerical difference between the crossing LOS and the non-crossing LOS. The pedestrian Roadway Crossing Difficulty Factor is limited to a maximum of 1.20 and a minimum of 0.80. RCDF = Max[0.80, Min{[(XLOS#-NXLOS#)/7.5 + 1.00],1.20}] Equation 23 Where RCDF = Roadway crossing difficulty factor XLOS# = Roadway crossing difficulty LOS Number NXLOS# = Non-crossing Pedestrian LOS number = (0.318 PSeg + 0.220 PInt + 1.606) Pseg = Ped. Segment LOS number (computed per equation #20) Pint = Ped. Intersection LOS number (computed per equation #21) The crossing difficulty LOS number is computed based on the minimum of the waiting-for-a-gap LOS number and diverting-to-a-signal LOS number. XLOS = Min [WaitForGap, DivertToSignal] Equation 24 Where: XLOS = Crossing LOS score (based on Exhibit 8) WaitForGap = Delay waiting for safe gap to cross. DivertToSignal = Delay diverting to nearest signalized intersection to cross. The delay is converted into an LOS numerical score based on the minimum of the mean delay waiting for a gap or diverting to a signal, according to the values given in Exhibit 8. Exhibit 8. Pedestrian Crossing LOS Score Minimum of Wait or Divert Delay (Seconds) XLOS Score 10 1 20 2 30 3 40 4 60 5 > 60 6 Wait-For-Gap LOS Calculation The Wait-For-Gap LOS is computed based on the expected waiting time required to find an acceptable gap in the traffic to cross the street. The acceptable gap is computed as a function of the number of lanes, their width, and the average pedestrian walking speed, with 2 seconds added. Acceptable Gap = Crossing Distance / Pedestrian Walk Speed + 2 seconds Equation 25 If there is adequate median refuge for pedestrians (median 6 feet wide or greater), then, at the analyst’s discretion, the shorter crossing distance to the median may be used. Otherwise the crossing distance is to the opposite curb. If it is illegal to cross the street between signalized intersections then WaitForGap is not computed. It is treated as “infinity” in equation 24.

Multimodal Level of Service for Urban Streets Page 21 The expected waiting time until an acceptable gap becomes available is computed as follows: ( )[ ] ttMeanWait −−= 1exp1 λ λ Equation 26 Where: Mean Wait = seconds waiting, must be greater than or equal to zero. t = The acceptable gap plus the time it takes for a vehicle to pass by the pedestrian. = Crossing distance/ped walk speed + vehicle pass-by time The average pass-by time = Average Vehicle Length/Average Speed, converted to seconds. λ = The average vehicle flow rate in vehicles per second. (If vehicle arrival rate is zero, mean wait is zero.) Exp = The exponential function If there is adequate median refuge for pedestrians (median 6 feet wide or greater), then, at the analyst’s discretion, the volume of traffic for only one direction of travel may be used. If vehicle arrival rate is zero, mean wait is zero. Divert To Signal LOS The LOS rating for diverting to the nearest traffic signal to cross the street is computed as a function of the extra delay involved in walking to and from the mid-block crossing point to the nearest signal and the delay waiting to cross at the signal. The geometric delay associated with diverting is the amount of time it takes the pedestrian to walk to a controlled crossing and back. To calculate this delay one must first determine the distance to nearest crossing. This distance is estimated as one-third the block length between signalized intersections. This distance is then divided by the pedestrian’s walking speed (assumed to be 3.5 feet/second) to obtain the geometric delay: Ped Geometric Delay = 2/3 * (Block Length)/Ped Walking Speed Equation 27 If there are no signalized intersections within the study section of the street, assume the Ped Geometric Delay is infinite. (In other words do not compute Divert to Signal Delay. Compute only wait time delay.) The control delay at the intersection is calculated as shown below. Ped Control Delay = (Cycle Length – Green Time)2/(2*Cycle Length) Equation 28 The total delay is the sum of the two: Total Ped Deviation Delay = Ped Geometric Delay + Ped Cycle Delay Equation 29 The total delay is then converted into a numerical LOS score by linearly interpolating numerical scores on the scale provided in Exhibit 8. Computation of Facility LOS From Segment LOS The level of service is computed separately for each direction of travel on each analysis segment. The segment levels of service (for a given direction of travel) are combined into an overall directional level of service for the study section of street by taking

Multimodal Level of Service for Urban Streets Page 22 a length weighted average of the segment levels of service for the analysis direction. ∑ ∑= )( )(*)( )( iL iLiLOS facilityLOS Equation 30 Where: LOS (facility) = LOS for subject direction of facility LOS (segment) = LOS is subject direction for each segment (i) L (segment) = Length of each segment (i) Special Cases Treatment of Sections With Significant Grades The pedestrian level of service equations are designed for essentially flat grades (grades of under 2% of any length). For steeper grades the analyst should consider applying an adjustment to the LOS estimation procedure to account for the negative impact of both up-grades and down-grades on pedestrian level of service. This adjustment probably should be sensitive both to the steepness of the grade and its length. However, research available at the time of production of this manual did not provide a basis for computing such an adjustment. The precise adjustment is left to the discretion of the analyst. Pedestrian LOS and ADA Compliance The Americans with Disabilities Act (ADA) sets various accessibility requirements for public facilities, including sidewalks on public streets. The United States Access Board (www.access-board.gov) has developed specific ADA Accessibility Guidelines For Buildings and Facilities (ADAAG) that apply to sidewalks and pedestrian paths/trails. Since pedestrian LOS is defined to reflect the average perceptions of the public, it is not designed to specifically reflect the perspectives of any particular subgroup of the public. Thus, the analyst should use caution if applying the pedestrian LOS methodology to facilities that are not ADA compliant. Pedestrian LOS is not designed to reflect ADA compliance or non- compliance, and therefore should not be considered a substitute for an ADA compliance assessment of a pedestrian facility. Treatment of Sections with Parallel Bike/Ped Path The pedestrian LOS is estimated using the path procedures contained in Chapter 18, Pedestrians, of the Highway Capacity Manual (2000). If the analyst has information indicating that pedestrians will also walk along the street itself, then the pedestrian LOS should be computed for both the street and the path and the two results combined in the same manner as described above for bicycles using the street and using the path. The weighted average pedestrian LOS for the street is computed based on the estimated share of pedestrians using the street and the path. Treatment of Streets with Sidewalk on Only One-Side The pedestrian LOS analysis for both sides of the street proceeds normally. On one side the sidewalk is evaluated. On the other side, the pedestrian LOS is evaluated for walking in the street. Treatment of Gaps in Sidewalks Segments with relatively long gaps (over 100 feet) in the sidewalk should be split into subsegments and the LOS for each evaluated separately.

Multimodal Level of Service for Urban Streets Page 23 The pedestrian LOS methodology is not designed to take into account the impact of short gaps in sidewalk (under 100 feet). Until such a methodology becomes available short gaps may be neglected in the pedestrian LOS calculation. However, the analyst should report the fact that there are gaps in the sidewalk in addition to reporting the LOS grade. Treatment of One-Way Traffic Streets The pedestrian LOS analysis proceeds normally for both sides of the street, even when it is one-way. Note however that the lane and shoulder width for the left-hand lane are used for the sidewalk on the left-hand side of the street. Treatment of Streets With Pedestrian Prohibitions If pedestrians are prohibited from walking along the street by local ordinance, then the pedestrian level of service is “F”. No pedestrian LOS computations are performed. Treatment of Sidewalk Closures If pedestrians are prohibited from walking along the street by a permanent sidewalk closure, then the pedestrian level of service is “F”. No pedestrian LOS computations are performed. Treatment of Crosswalk Closures If pedestrians are prohibited from crossing the street by a local ordinance (such as a jay-walking ordinance against mid-block crossings between signals), then the pedestrian RCDF (roadway crossing difficulty factor) is set to 1.00 and the rest of the pedestrian LOS analysis performed as normal. If pedestrians are prohibited from crossing the subject urban street at an intersection, then the pedestrian RCDF (roadway crossing difficulty factor) is computed taking into account the extra delay to reach an alternative intersections where crossing the street is allowed. If the crosswalk at an intersection in the direction of travel along the subject urban street is closed, then the pedestrian intersection LOS is computed adding in the extra delay necessary to cross the other 3 legs of the intersection to continue walking along the arterial. Treatment of Streets With Frontage Roads In some cases a jurisdiction will provide frontage roads to an urban street. There will usually be no sidewalks along the urban street, but there will be sidewalks along the outside edge of each frontage road. If the analyst has information indicating that pedestrians walk along the urban street without the sidewalks, then the pedestrian LOS analysis should be performed on the urban street. If the analyst has information indicating that pedestrians walk exclusively along the frontage roads then the pedestrian LOS analysis should be performed for the frontage roads. The analyst may perform some tests to see if inclusion of the urban street traffic in the frontage road pedestrian LOS analyses has a significant effect on pedestrian LOS. It is possible that the urban street traffic is located so far away from the frontage road sidewalks as to have negligible effect on the pedestrian LOS. Treatment of Pedestrian Overcrossings The pedestrian LOS methodology is not designed to account for pedestrian bridges, either across the urban street or along the urban street.

Multimodal Level of Service for Urban Streets Page 24 Treatment of Pedestrian Signals Pedestrian signals are treated as intersections for the purpose of computing diversion delay in the RCDF computations. Pedestrian signals are ignored in the pedestrian intersection LOS computations (Pedestrians experience no delay or crossing traffic walking along the arterial at these signals). A segment with a pedestrian signal is treated in the pedestrian non-density LOS computation as a non-intersection. Treatment of Unsignalized Mid-Block Crosswalks The pedestrian LOS methodology is not designed to account for mid-block unsignalized crosswalks (with or without flashing warning lights). Treatment of Railroad Crossings The LOS methodology is not designed to account for the impacts on pedestrian LOS of railroad crossings with frequent train traffic. The analyst should determine if train frequencies and durations of crossing gate closures are low enough that their impact on pedestrian LOS can be neglected. Treatment When Mid-Block Crossings are Illegal The RCDF factor is computed using the divert to signal delay. The average wait time to cross the street is not computed. Treatment of Unsignalized Intersections Including Roundabouts The pedestrian LOS methodology does not provide for computation of pedestrian intersection LOS at unsignalized intersections. Segments with unsignalized intersections should be analyzed as if no intersections were present for the purposes of the pedestrian LOS analysis. Treatment of Unpaved Paths/Sidewalks The pedestrian LOS methodology is not designed to account for unpaved paths/sidewalks in the urban street right-of-way. The analyst should use local knowledge about the climate and the seasonal walkability of unpaved surfaces to determine whether an unpaved sidewalk can be considered as almost as good as a paved sidewalk for the purpose of the pedestrian LOS computation. Otherwise the unpaved sidewalk should be considered the same as “no- sidewalk” for the purpose of pedestrian LOS computation. Treatment of Multi-Lane Free Right Turn Lanes Multilane free-right turn lanes may be, at times and under certain conditions, an impediment to pedestrian travel (depending on vehicle volumes, visibility, pedestrian volumes, vehicle speeds, and local enforcement). The pedestrian LOS method is not specifically designed for such situations. Should the analyst choose to compute the pedestrian LOS using the methods in this users guide, he or she should carefully evaluate the results and apply adjustments if the initial result does not appear to accurately represent the conditions present.

Multimodal Level of Service for Urban Streets Page 25 Treatment of High-Speed Free Right Turn Lanes At some freeway interchanges with urban streets there may be a high-speed free right turn provided for vehicle traffic wishing to turn onto one or more of the freeway on-ramps. In such cases the on-ramp may be a significant impediment to pedestrian travel along the urban street and through the freeway interchange. The pedestrian LOS method is not specifically designed for such situations. Should the analyst choose to compute the pedestrian LOS using the methods in this users guide, he or she should carefully evaluate the results and apply adjustments if the initial result does not appear to accurately represent the conditions present. Treatment of Bus Lanes and Bus Streets In the case of bus streets, the pedestrian LOS is computed normally, with transit vehicles being the only motorized vehicles on the street. In the case of bus lanes, the pedestrian LOS analysis proceeds normally. The only difference is that only transit vehicles (can carpools, if allowed) are assigned to the bus lane. ESTIMATION OF AUTO PERFORMANCE MEASURES (SPEED, DELAY, STOPS, QUEUE) The methodologies outlined in the Highway Capacity Manual (2000) can be used to predict average auto speed, delay, stops, and queues, when these performance measures cannot be measured in the field. Auto Speed The mean speed for auto traffic is computed by dividing the length of the street being evaluated by the average time it takes an automobile to travel the length of the street including all stops and delays along the way. Chapter 15, Urban Streets, of the Highway Capacity Manual (2000) describes a method for estimating the mean speed of through traffic on a street. Equation 31 Where: S = the mean speed (mph) L = Length (miles) T = Average travel time (hours) The average travel time is equal to the travel time between intersections plus the sum of the average delay at each intersection to through traffic. The intersections may be signalized or unsignalized. They may be stop controlled or roundabouts. Equation 32 Where: T = Average travel time (hours) S0 = posted speed limit (mph) L = Length (miles) di = Average delay for through traffic at intersection “i” (secs)

Multimodal Level of Service for Urban Streets Page 26 The average delay is estimated using one of the intersection analysis methods described in Chapters 16 or 17 of the Highway Capacity Manual (2000). Auto Delay Delay is defined as the difference between the actual travel time and the desired travel time to travel the length of the street. The desired travel time is the length of the street divided by the desired speed. The desired speed can be set by the policy of the local agency. If there is no specific local policy about the desired speed for the street, the posted speed limit can be used as the desired speed for the street. equation 33 Where: D = Delay (hours) T = Average travel time (hours) S0 = posted speed limit (mph) L = Length (miles) Auto Stops The number of stops per mile for autos can be estimated using the following equation fitted to more complex procedures developed by NCHRP 3-79 for estimating stops. equation 34 Where: L = length of street X = volume/capacity ratio for through movement A1, A2, A3 = parameters (see table below) Exhibit 9: Parameters for Auto Stops Per Mile Equation Signal Progression Arrival Type A1 A2 A3 Adverse Signal Progression 1,2 0.636 5.133 0.051 No Signal Coordination 3 0.478 6.650 0.028 Good Signal Progression 4,5,6 0.327 9.572 0.013 ESTIMATION OF TRANSIT PERFORMANCE MEASURES (SPEED, DELAY) The average transit speed is best measured in the field, or using published schedules. It can be predicted using the methods described in Chapter 27 of the Highway Capacity Manual (2000).