Predicting Air Quality Effects of Traffic-Flow Improvements: Final Report and User's Guide (2005)

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.

User’s Guide

USER’S G UIDE 1 CHAPTER 1 INTRODUCTION This user’s guide describes the recommended NCHRP 25-21 methodology for predicting the long- and short-term mobile source emission impacts of traffic-flow improvement projects. The application of the methodology is illustrated through example problems consisting of case studies of var- ious traffic-flow improvements. 1.1 OBJECTIVES OF THE NCHRP 25-21 METHODOLOGY The objective of the NCHRP 25-21 research project was to develop and demonstrate, in case study applications, a methodology to predict the short-term and long-term effects of corridor-level, traffic-flow improvement projects on car- bon monoxide (CO), volatile organic compounds (VOCs), oxides of nitrogen (NOX), and particulate emissions (PM). The methodology is designed to evaluate the magnitude, scale (such as regionwide, corridor, or local), and duration of the effects for a variety of representative urbanized areas. It is designed to be implementable in a broad range of existing software used for travel demand modeling. The methodology is not designed to predict pollutant con- centrations or ozone formation resulting from traffic-flow improvement projects. Rather, the methodology uses the best available emission factors and vehicle operations and activ- ity data to estimate net changes in emissions of ozone pre- cursors, CO, and particulates. 1.2 ORGANIZATION OF USER’S GUIDE Chapter 2 provides an overview of the NCHRP 25-21 meth- odology. The next five chapters describe the various modules of the methodology in more detail. Chapters 8 through 19 illustrate the application of the meth- odology to a series of case studies: • Case 1: Addition of a Freeway Lane in a Rural Area, • Case 2: Removal of a Freeway Lane in an Urban Area, • Case 3a: Removal of Freeway HOV lane from an uncon- gested freeway, • Case 3b: Removal of Freeway HOV lane from a con- gested freeway, • Case 4: Narrowing a Street, • Case 5: Access Management, • Case 6: Intersection Channelization Improvement, • Case 7: Signal Coordination, • Case 8: Transit Improvement, • Case 9: Removal of a Freeway Express Bus Park-and- Ride Lot, and • Case 10: Construction of a 30-Year Transportation Improvement Program. The case studies reported here in the user’s guide are iden- tical to the case studies described in the final report. Readers may note, however, that Case Studies 2, 3a, 3b, 4, and 9 here involve the effects of removing lanes or a park-and-ride lot rather than those of adding lanes and a park-and-ride lot, as described in the final report. This is because it is mechani- cally easier to remove an HOV lane or a park-and-ride lot from coded highway and transit networks than it is to add one. When adding facilities (such as an HOV lane or a park- and-ride lot) to a model, the modeler must also code the sup- porting link structure and must be careful to follow the net- work coding conventions used in the network. When deleting a facility, the modeler need not be concerned about the cod- ing conventions. The user’s guide consequently describes the case studies and their results as they were actually performed in the Puget Sound Regional Council (PSRC) travel demand model. The results from Case Studies 2, 3a, 3b, 4, and 9 were then reported in the final report as their mirror image. For example, a lane closure in the user’s guide for Case Study 2 is reported as a lane addition in the final report. The “before” result for Case Study 2 in the user’s guide became the “after” result in the final report. The “after” result in the user’s guide became the “before” result in the final report.

2US ER ’S G UI DE CHAPTER 2 THE METHODOLOGY The NCHRP 25-21 methodology is designed to answer one fundamental question, “Will a specified traffic-flow improve- ment contribute to improved or worsened air quality locally and at the regional level, in the short term and in the long term?” Repeated exercise of the methodology on various case studies will answer the question, “Under what conditions will a specified traffic-flow improvement contribute to improved or worsened air quality?” 2.1 THEORETICAL FOUNDATION The NCHRP 25-21 methodology proceeds from the funda- mental theoretical foundation that “nobody travels for the fun of it.” People travel in order to participate in activities or to obtain goods that are superior to what they could have done or obtained at their original location. Even sightseers use the transportation to experience a vista they could not see at home. They may say they enjoy the drive, but what they really enjoy is what they can see out of the window. The research team will exclude from this blanket statement individuals who test their vehicles or are hired to drive a vehicle. Travel demand is, therefore, not vehicle-miles traveled (VMT). Travel demand is the schedule of activities, by loca- tion, that travelers would like to pursue that day. In modeling parlance, it is the origin-destination (OD) table of person trips for that day by time of day. However, VMT is the most cost-effective measure (from the traveler’s point of view) for measuring that demand. Thus, traffic-flow improvements by reducing average travel times can affect both the total demand for travel and the trav- eler’s choice of the most cost-effective means for satisfying that demand. In addition, some traffic-flow improvements do not change the average travel time but reduce the variance in travel speeds by smoothing out the traffic flow. Thus, it is possible for a traffic-flow improvement to have no effect on demand or on how that demand is satisfied on the street system and yet still have an effect on air quality by smoothing out the “stop-and- go” nature of the trip itself. Finally, a series of traffic-flow improvements can make one portion of the metropolitan area more attractive to growth and new development than older, more established parts of the region. The shifting of growth from centrally located devel- oped areas to undeveloped fringe areas can affect both demand and how it is satisfied on the transportation system. This effect is very long term. (Extensive transportation capacity invest- ments in one metropolitan area can also increase the net in- migration to the region, but this effect will not be considered in this research.) Thus, the NCHRP 25-21 methodology addresses four basic mechanisms by which traffic-flow improvements can influ- ence mobile source emissions: • Operational improvements that smooth out traffic flow and thus reduce acceleration/deceleration events, • Travel time savings and losses on particular routes and modes of travel that influence the traveler’s choice of the most cost-effective means for satisfying their demand to travel, • Travel time savings that increase the total demand for travel, and • Travel time savings that increase the relative attractive- ness and therefore the growth rate of subareas in the region. The first mechanism, operational improvements, will be called the “operations” effect. Traffic flow improvements may increase the average speed on the facility, and/or they may increase the capacity of the facility prior to affecting travel behavior. Operational improvements will also affect vehicle mode of operation activity by reducing acceleration and deceleration events. The operations effect occurs on the first day that an improvement is opened for traffic. Travel- ers have not yet had an opportunity to change their demand schedule in response to the travel time savings provided by the improvement. The second and third effects of traffic-flow improvements will be combined into a single “traveler behavior” effect. This effect comes in the months following opening day. As travel- ers become aware of the improvements, they change route, mode of travel, and departure time to take advantage of them. After the improvement has been in place for sufficient time for travelers to change their demand schedule (e.g., the OD table), they will take advantage of the reduced travel costs brought about by the improvement. Traveler behavior effects include changes in destination choice and trip generation (extra trips or stops along the way of a preexisting trip). The result of the behavior effects will be to partially counteract the

opening day travel time improvements. It is assumed that the traveler behavior effects cannot completely eliminate the “opening day” travel time improvements; otherwise, there is no longer a stimulus to cause the traveler behavior effects. The fourth effect of traffic-flow improvements is a redistri- bution of growth (i.e., new homes and jobs) within the region to areas that benefit from the travel time savings attributable to a traffic-flow improvement. This effect will be called the “growth redistribution” effect. It is possible that the traffic- flow improvement might also enhance the relative competi- tiveness of the entire metropolitan region for new jobs and new homes, thus influencing overall growth of the region. How- ever, this global effect is beyond the scope of this research project and methodology. It would require a full-blown socio- economic forecasting model plus some kind of assumption regarding the pace of traffic-flow improvements in other competing metropolitan areas of the United States, Mexico, and Canada. The research team assumes for the sake of this research project that all competing metropolitan areas have similar policies for implementing traffic-flow improvements. Thus, a specific set of traffic-flow improvements would likely also be implemented in all metro regions. Thus, the relative com- petitiveness of the metro regions would be unaffected. The methodology will focus on redistribution impacts within a region, not total growth impacts for the entire region. The foundation of the NCHRP 25-21 methodology is that traveler behavior response and growth redistribution occur only if the traffic-flow improvement results in a net change in trip travel time. Thus, travel behavior responses and growth redistribution can never drive travel time savings to zero. This research will not account for possible effects of a traffic-flow improvement that do not relate to travel time, such as operating cost improvements. Vehicle operating cost (which can also affect travel demand) is correlated with travel time and is not treated separately here, since the research team is not considering toll changes. The marginal effects of reduced acceleration/deceleration events on vehicle wear and tear (and thus vehicle operating costs) also will be unac- counted for. 2.2 OUTLINE OF THE METHODOLOGY The recommended methodology is a blended macroscopic- microscopic approach composed of five modules: • The HCM Assignment Module predicts the highway travel times based upon traffic operations analysis speed- flow equations contained in the 2000 Highway Capac- ity Manual (HCM). • The Traveler Behavior Response Module uses elastici- ties derived from the Portland Tour-Based Model to pre- dict the impact of travel time changes on trip making by peak period and by mode of travel. 3 • The Growth Redistribution Module predicts the impacts of traffic-flow improvements on growth patterns within the region. Subareas within the region that have better- than-average accessibility improvements will have greater-than-average growth rates in the region. • The Modal Activity Module translates the mean speeds and volumes predicted by the previous modules into a distribution vehicle-hours of travel (VHT) by speed cat- egory and acceleration/deceleration rate category. • The Air Quality Module translates the modal activity data into estimates of vehicle emissions. The methodology employs macroscopic approximations of microscopic behavior throughout each of the modules. The intent is to obtain a practical methodology that can be employed by a wide range of agencies while retaining as much as possible the behavioral accuracy of a microscopic analytical approach. The proposed methodology predicts the change in demand and vehicle emissions caused by traffic-flow improvements at two points in time: short term (5 to 10 years) and long term (25+ years). Figure 1 provides a flow chart overview of the methodology. The methodology requires as input • A set of baseline travel demand tables (OD tables) for AM, PM, and off-peak periods; • A set of baseline highway and transit networks for the AM, PM, and off-peak periods; and • The proposed traffic-flow improvement characterized in terms of its impact on mean free-flow speeds and capac- ities in the baseline networks. The first round of analysis (i.e., the base) assigns the base- line OD tables (by mode of travel and time period) to the baseline (i.e., no improvement) transportation networks for each time period and mode. The methodology then computes the mean speed and flow for each highway link. This link information is fed to the Vehicle Modal Activity Module, which outputs tables of vehicle activity (VHT) by speed and acceleration/deceleration category. The modal activity infor- mation is fed to the Vehicle Emissions Module (VOC, CO, NOX, and PM), which computes the vehicular emissions. The second round of analysis (short term) adds the traffic- flow improvement to the baseline network and computes new vehicle trip travel times for the improved network. The new travel times are compared with the baseline travel times to determine the changes in travel times. The changed travel times are entered into the Traveler Behavior Response Mod- ule, which modifies the baseline OD tables to produce revised OD tables. The revised OD tables are assigned to the high- way network to produce mean speed and flow for each high- way link. The information is then fed to the Vehicle Modal Activity and Vehicle Emissions Modules to obtain emissions for the short term. USER’S G UIDE

The third round of analysis (long term) feeds the short- term results into the Growth Redistribution Module, which computes the impacts of the traffic-flow improvements on the relative growth rates of zones within the region. The revised growth rates are used to redistribute the origins and destinations of the trips in the short-term OD tables. The revised OD tables are then fed back through the Traveler Behavior Response Module one more time to obtain mean speed and flow for each highway link. The information is then fed to the Vehicle Modal Activity and Vehicle Emis- sions Modules to obtain emissions for the long term. The methodology generally follows the recommendations of NCHRP Project 8-33. The methodology is designed to pre- dict the changes due to the traffic-flow improvement projects. It does not predict baseline conditions. Baseline conditions (the baseline OD tables) must be input to the methodology. The methodology does not separately model the demand response of heavy-duty vehicles to traffic-flow improvements. Modeling truck demand response would require a completely 4 separate methodology with separate data requirements. Trucks are presumed to be a fixed percent of current and future traffic demands in this methodology. The methodology does not forecast socioeconomic changes, traffic condition changes, or emission changes that are due to factors other than traffic-flow improvements. The proposed methodology, therefore, must be used in conjunction with some other model for predicting future baseline conditions, usually a conventional travel demand model. 2.3 HCM ASSIGNMENT MODULE On the opening day, drivers will experience the maximum travel time savings provided by an improvement project. The improved road section will have higher operating speeds and fewer and milder acceleration/deceleration events. If the improvement also increases peak capacity, then more vehicles will be able to pass through the improved segment during the US ER ’S G UI DE Growth Redistribution Module Growth Redistribution Module Traveler Behavior Response Module HCM Assignment Module Traffic-Flow Improvement Baseline OD Tables Baseline Network Assign Traffic Using HCM Curves HCM Assignment Module Vehicle Modal Activity Module Vehicle Modal Activity Module Vehicle Emissions Module Vehicle Emissions Module VMT VHT Assign Traffic Using HCM Curves HCM Assignment Module Results Base Short Term Long Term Assign Revised Traffic Using HCM Curves Revised HCM Assignment Module Assign Revised Traffic Using HCM Curves Revised HCM Assignment Module VMT VHT VMT VHT Traveler Behavior Response Module Traveler Behavior Response Module Assign Revised Traffic Using HCM Curves Revised HCM Assignment Module Figure 1. The NCHRP 25-21 methodology.

peak hour, and this increase in peak capacity may impact downstream capacity bottlenecks. The HCM Assignment Module predicts the highway vehi- cle travel time effects of the traffic-flow improvement for a fixed level of demand. Inputting the base demand to the mod- ule is equivalent to predicting travel times for the day that a traffic-flow improvement is first opened to traffic. Travelers have not had time to adjust to the travel time savings, so, at this stage in time, there is no demand response. If future demands are input to the module, then the module will pre- dict future travel times and delays for that level of demand. This module has multiple uses in the methodology, being applied to the base case, short-term, and long-term analyses. The required inputs for the HCM Module are vehicle OD tables (by mode and time period), the baseline geometric char- acteristics of the regional highway network (facility type, free- flow speeds, capacity characteristics, and segment lengths), and similar geometric information for the traffic-flow improve- ment. The module computes the highway link operating char- acteristics: volume/capacity and mean speed. The module uses the 2000 HCM Chapter 30 speed-flow equations (sometimes called the Akcelik equations in the lit- erature) and capacities to estimate the mean speed of traffic on each link of the highway network. A standard static users equilibrium (SUE) assignment of the OD table is performed in this module using the HCM equations for each period of the day (typically AM, PM, and off-peak). It should be noted that the travel time savings on the improved segment may be partially compensated by increased delays at downstream bottlenecks. This “downstream” effect of traffic-flow improvements are neglected by this module. (Tests with the PSRC model show that this effect is not sig- nificant for the conditions of the PSRC model. See Chap- ter 12 of the final report for more details.) The module computes only highway travel times for mixed- flow and high-occupancy vehicle (HOV) lanes. Transit, bicy- cle, and pedestrian travel times (if needed) must be computed using some standard travel demand modeling procedure con- sistent with the procedure used to estimate the baseline OD tables by time period for each of these nonauto modes of travel. 2.4 TRAVELER BEHAVIOR RESPONSE MODULE Travelers will adjust their demand schedule for travel in response to changes in the travel time required to reach their daily activity locations. Demand responses may include changes in trip lengths (i.e., trip distribution), number of trips (i.e., trip generation), time of day (i.e., peaking), and mode of travel (i.e., mode choice). The Traveler Behavior Response Module predicts how travel demand will react to the travel time savings created by traffic-flow improvements. The module computes estimated changes in demand for each entry in the OD table for each mode of travel and each 5 period of the day based on the estimated changes in travel times by mode and by time period. The module employs direct elasticities and cross-elasticities derived from the Port- land Tour-Based Model. An example of a direct elasticity is the percentage change in HOV demand during the AM peak for each percentage change in HOV travel time during the AM peak. An example of a cross-elasticity is the percentage change in HOV demand during the AM peak for each per- centage change in single-occupancy vehicle (SOV) travel time during the AM peak. Cross-elasticities are also used to account for shifting of travel between peak and off peak for each mode of travel. Heavy-duty vehicles (e.g., trucks) are presumed to respond in the same manner as light-duty SOVs to travel changes in this module. They are not modeled separately. If a metropolitan planning organization already has a tour- based model in place that can predict the impacts of travel time and cost changes on out-of-home trip making, time of day, and mode choice, then that model can be used in place of the sim- pler Traveler Behavior Response Module described here. 2.5 GROWTH REDISTRIBUTION MODULE Significant improvements in transportation infrastructure in one part of the urban region will impact the geographic distribution of housing and job growth in the region over the very long term (25+ years). Significant infrastructure invest- ments may also affect the total growth rate for the region by changing the attractiveness of the region to migrants from other regions. This latter effect, however, requires a model at the national level to properly account for migration between regions. Therefore, this overall affect on total regional growth will be excluded from the methodology. The Growth Redistribution Module will predict the very long-term impacts of localized travel time changes (caused by traffic-flow improvements) on the geographic distribution of growth in a metropolitan area. There are already several sophisticated land-use models available (such as UrbanSim) that could be used for the purpose of this module. However, these models require a great deal of specialized economic data and effort to set up for a region (which may be beyond the resources of many metropolitan planning organizations). When a sophisticated land-use model exists in a region, it can be used to predict the long-term effects. When such a model is not available, the simple Growth Redistribution Module is proposed for use to approximate the long-range land-use effects of traffic-flow improvements. The Growth Redistribution Module requires that a base- line 20- to 25-year forecast of land-use growth (households and employment changes) be available for the metropolitan area. This baseline forecast should have been prepared either manually or with a model taking into account accessibility changes as well as all of the other factors that commonly affect the distribution of growth within a region. A simple lin- ear regression model is fitted to the baseline land-use forecast. USER’S G UIDE

The regression model predicts the change in the growth rate in households and employment in each zone of the region as a function of the relative change of accessibility for each zone. Although not sophisticated enough to predict actual growth, the model should be sufficient to predict how small changes in travel time accessibility can affect the predicted baseline growth rate in specific zones of the region. The module presumes that total regional growth will be unaffected by traffic-flow improvements (in other words, the module will not be sensitive to the potential effects of differ- ing levels of regional traffic-flow improvements on the com- petitiveness of regions for attracting new households or jobs). The module predicts only how regional growth might be real- located from marginally less accessible zones to more acces- sible zones within the region. 2.6 VEHICLE MODAL ACTIVITY MODULE The Vehicle Modal Activity Module converts the macro- scopic vehicle activity data produced by the previous modules (VHT and VMT by link, mode, and time period) to micro- scopic modal activity data (VHT by speed and acceleration category). Four tables (Uncongested Freeway, Congested Freeway, Uncongested Arterial, Congested Arterial) contain- ing percentages are used to determine the proportion of total vehicles-hours on each street and freeway segment that are spent in each speed/acceleration category. These tables were derived from microsimulation of vehicle activity on example real-world sections of freeways and arterial streets using the Federal Highway Administration Corridor Simulation (CORSIM) model. Additional tables for other facility types and varying intel- ligent transportation systems (ITS) and traffic management options can be created using the FHWA CORSIM program. The creation of such tables was beyond the resources of this research project and was consequently deferred to future research. 2.7 VEHICLE EMISSION MODULE The Vehicle Emission Module converts the passenger car modal activity data into estimates of vehicular emissions. The 6 potential impacts of traffic-flow improvements on heavy-duty vehicle and transit vehicle emissions are neglected. (The nec- essary information on heavy-duty vehicle emission rates by mode of operation was not available at the time of this research.) Modal emission factors from the University of Cali- fornia, Riverside, NCHRP 25-11 Comprehensive Modal Emission Model (CMEM) and Emission Factor 2000 (EMFAC2000) are used to produce the emission estimates. The primary effects of traffic-flow improvement projects relate to speeds and delay along specific corridors. The direct emission effects include • Running exhaust emissions (due to changes in vehicle speed and acceleration profiles, as well as changes in VMT due to route choice), • Running loss emissions (due to changes in total travel time), and • Refueling and CO2 emissions (due to changes in fuel efficiency). CMEM was used to produce running exhaust emission rates for the specified speed and acceleration frequency dis- tributions (SAFDs) contained in the modal activity data. The impacts of traffic-flow improvements on running loss emissions and refueling/CO2 emissions are currently not included in the NCHRP 25-21 methodology. This is because the NCHRP 25-21 methodology is based on CMEM, which does not include these emissions. There are two secondary effects of traffic-flow improve- ment projects that influence emissions. First, to the extent that traffic-flow improvement projects reduce total travel time, there may be some increase in the number of trips made, resulting in additional start emissions. Second, both reduced travel time and increased numbers of trips alter the number and timing of hot soak, diurnal, and resting loss periods for the vehicle. Neither of these effects is included in CMEM, and, consequently, neither is included in the current version of the NCHRP 25-21 methodology. Virtually all emission rates depend on ambient tempera- ture. The NCHRP 25-21 methodology uses an average sum- mer day temperature profile for VOC and NOX, and an aver- age winter day for CO analyses. US ER ’S G UI DE

7USER’S G UIDE CHAPTER 3 THE HCM ASSIGNMENT MODULE The purpose of the HCM Assignment Module is to improve current methods for estimating the travel delay effects of traf- fic congestion. The approach taken was to replace the con- ventional Bureau of Public Roads (BPR) equation method still used in many travel demand models with more up-to- date traffic operations research results contained in the 2000 HCM. The module substitutes the following HCM-based information into the SUE traffic assignment step of the travel demand model process: • Free-flow speeds by facility type and area type; • Link capacities by facility type, area type, and other characteristics of facility; and • HCM-based Akcelik set of speed-flow equations. 3.1 FREE-FLOW SPEEDS The free-flow speed is the mean speed of traffic when demand is so low that changes in demand do not affect the mean speed of traffic on the segment. For freeways and multi- lane highways, free flow is the mean speed observed when volumes are under 1,300 vehicles per hour per lane. For sig- nalized streets, the free-flow speed is the maximum mean speed of traffic obtained at any point between signalized intersections for low-volume conditions. The mean speed is computed as the sum of the travel times to traverse the length of the segment, divided into the length of the segment times the number of vehicles in the sample. The following linear equations from NCHRP Report 387 can be used to estimate free-flow speed based on the posted speed limit for arterials, freeways, and highways. For posted speed limits of 50 mph or greater, FFS = 0.88 ∗ PSL + 14 Equation 1 For posted speed limits of less than 50 mph, FFS = 0.79 ∗ PSL + 12 Equation 2 Where: FFS = free-flow speed (mph) and PSL = posted speed limit (mph). 3.2 CAPACITIES Highway link capacities are estimated using the proce- dures contained in the 2000 HCM. The following subsec- tions summarize the information contained in Chapter 30 of the HCM. 3.2.1 Freeways, Multilane Highways, and Two-Lane Highways The following equation is used to compute the capacity of a freeway or highway link at its critical point. The critical point is the point on the link with the lowest throughput capacity. c = Q ∗ N ∗ Fhv ∗ Fp ∗ Fg ∗ PHF Equation 3 Where: c = capacity (vph), Q = the passenger car equivalent (p.c.e.) capacity per hour per lane, N = number of through lanes (ignore auxiliary and “exit only” lanes), Fhv = heavy-vehicle adjustment factor, Fp = driver population adjustment factor, Fg = grade adjustment factor, and PHF = peak-hour factor. Table 1 provides the HCM-recommended passenger car equivalent capacities per lane (Q). See the HCM for appro- priate values for the adjustment factors. 3.2.2 Arterials The capacity of an arterial is determined by examining the through movement capacity at each signal-controlled inter- section on the arterial link. The intersection with the lowest through capacity determines the overall capacity of the arte- rial link. The following equation is used to compute the one- direction through capacity at each signal. c = S0 ∗ N ∗ fw ∗ fhv ∗ Fg ∗ fp ∗ fbb ∗ fa ∗ fLU ∗ fLT ∗ fRT ∗ FLpb ∗ fRpb ∗ PHF ∗ g/C Equation 4

8US ER ’S G UI DE Where: c = capacity (vph), s0 = ideal saturation flow rate = 1,900 vehicles per hour of green per lane, N = number of lanes, fw = lane-width adjustment factor, fhv = heavy-vehicle adjustment factor, Fg = grade adjustment factor, fp = on-street parking crossing adjustment factor, fbb = local bus adjustment factor, fa = central business district adjustment factor, fLU = lane use adjustment factor, fLT = left-turn adjustment factor, fRT = right-turn adjustment factor, fLpb = pedestrian/bicycle blockage of left-turn factor, fRpb = pedestrian/bicycle blockage of right-turn factor, PHF = peak-hour factor, and g/C = ratio of effective green time per cycle. See the HCM for appropriate values for the adjustment factors. 3.3 HCM/AKCELIK SPEED-FLOW EQUATION The mean speed for each segment during the peak period is estimated using the following equations taken from the 2000 HCM. The mean vehicle speed for the link is computed by dividing the link length by the link traversal time. The link traversal time (R) is computed according to the following modified Akcelik equation from the HCM: Equation 5 Where: R = segment traversal time (hours), R0 = segment traversal time at free-flow speed (hours), D0 = zero-flow control delay at signals (equals zero if no signals) (hours), R R D D N T x x J L x N T L= + + + ∗ − + −( ) + ∗ ∗   0 0 2 2 2 2 0 25 1 1 16 . ( ) DL = segment delay between signals (equals zero if no signals) (hours), N = number of signals on the segment (equals one if no signals), T = expected duration of the demand (length of analysis period) (hours), x = segment demand/capacity ratio, L = segment length (miles), and J = calibration parameter. The segment traversal time at free-flow conditions is com- puted from the free-flow speed: Equation 6 Where: R0 = free-flow traversal time (hours), L = length (miles), and S0 = the segment free-flow speed (mph). The computation of the signal delay terms (D0, DL) is explained in the following section. The number of signals (N) on the facility segment excludes the signal at the start of the street segment (if present), because this signal should already have been counted in the upstream segment. (Streets are often split into segments (links) starting and ending at signalized intersections. The counting conven- tion suggested here avoids double-counting of the signals located at the start and end points of each segment.) When there are no signals on the facility, N is still set equal to one. This is because N is really the number of “delay- causing elements” on the facility. Each delay-causing ele- ment on the facility adds to the overall segment delay when demand starts to approach and/or exceed capacity at that ele- ment or point. Because demand in excess of capacity must wait its turn to enter the facility segment, there is always at least one “delay-causing element” (the segment itself) on a facility even when there are no signals. The more signals there are on a facility, the more points there are where traffic is delayed along the way. This means that a bottleneck sec- tion of the facility should be coded as a single link and not arbitrarily split into sublinks. The HCM/Akcelik equation R L S0 0= PCE Capacity (passenger cars per hour per lane) Free-Flow Speed Freeways Multilane Hwys Two-Lane Hwys 75 mph (112 km/h) 2400 70 mph (104 km/h) 2350 65 mph (96 km/h) 2300 2200 1700 60 mph (88 km/h) 2250 2100 1700 55 mph (80 km/h) 2000 1700 50 mph (70 km/h) 1900 1700 TABLE 1 Passenger car equivalent (PCE) capacities for freeways and highways

(and the standard BPR equation as well) treats each link as a potential delay-causing bottleneck on the network. Splitting one real-world bottleneck into three hypothetical links, each with the same demand, would triple the estimated delay at the bottleneck. The duration of demand (T) is set equal to the length of the analysis period. The segment demand/capacity ratio (x) is the ratio of the total demand for the analysis period divided by the total capacity for the period. The calibration parameter J is selected so that the traver- sal time equation will predict the mean speed of traffic (aver- aged over the length L of the link) when demand is equal to capacity. It is computed according to the following equation: 9 Equation 7 Where: J = calibration parameter, Rc = link traversal time when demand equals capacity (hours), R0 = free-flow speed traversal time (hours), D0 = zero-flow control delay (hours), and DL = segment delay (hours). The values for J, shown in Tables 2 and 3, reproduce the mean segment speeds at capacity predicted by the analysis J R R D D L c L = − − −( )0 0 2 2 USER’S G UIDE Facility Type Signals Per Km Free-Flow Speed (km/h) Speed at Capacity (km/h) J Freeway n/a 120.0 85.7 1.11E-05 Freeway n/a 110.0 83.9 8.00E-06 Freeway n/a 100.0 82.1 4.75E-06 Freeway n/a 90.0 80.4 1.76E-06 Multilane Hwy n/a 100.0 88.0 1.86E-06 Multilane Hwy n/a 90.0 80.8 1.60E-06 Multilane Hwy n/a 80.0 74.1 9.91E-07 Multilane Hwy n/a 70.0 67.9 1.95E-07 Two-Lane Hwy n/a 110.0 70.0 2.70E-05 Two-Lane Hwy n/a 100.0 60.0 4.44E-05 Two-Lane Hwy n/a 90.0 50.0 7.90E-05 Two-Lane Hwy n/a 80.0 40.0 1.56E-04 Two-Lane Hwy n/a 70.0 30.0 3.63E-04 Customary Units SI Units Facility Type Signals Per Mile Free-Flow Speed (mph) Speed at Capacity (mph) J Freeway n/a 75.0 53.3 2.947E-05 Freeway n/a 70.0 53.3 2.003E-05 Freeway n/a 65.0 52.2 1.423E-05 Freeway n/a 60.0 51.1 8.426E-06 Freeway n/a 55.0 50.0 3.306E-06 Multilane Hwy n/a 60.0 55.0 2.296E-06 Multilane Hwy n/a 55.0 51.2 1.821E-06 Multilane Hwy n/a 50.0 47.5 1.108E-06 Multilane Hwy n/a 45.0 42.2 2.174E-06 Two-Lane Hwy n/a 65.0 40.2 9.043E-05 Two-Lane Hwy n/a 60.0 35.2 0.0001385 Two-Lane Hwy n/a 55.0 30.2 0.0002239 Two-Lane Hwy n/a 50.0 25.2 0.0003893 Two-Lane Hwy n/a 45.0 20.2 0.0007484 TABLE 2 Recommended calibration parameters J for freeways and highways

procedures contained in the 2000 HCM. These two tables use the following HCM definitions of facility types: • Freeway—A multilane, divided highway with a min- imum of two lanes for the exclusive use of traffic in each direction and full control of access without traf- fic interruption. • Multilane highway—A highway with at least two lanes in each direction for the exclusive use of traffic, with no control or partial control of access, but that may have periodic interruptions to flow at signalized intersections no closer than 2 miles apart. • Two-lane highway—A highway with only one lane in each direction (with or without occasional passing lanes) for the exclusive use of traffic, with no control or partial control of access, but that may have periodic interrup- tions to flow at signalized intersections no closer than 2 miles apart. • Arterial—A signalized street that primarily serves through traffic and that secondarily provides access to abutting properties, with signals spaced 2 miles or less apart. Arterials are divided into classes according to the 10 posted speed limit and signal density criteria shown in Table 4. 3.4 SIGNAL DATA REQUIRED BY HCM/AKCELIK The zero-flow control delay and the between-signal delay are required to estimate speeds for signalized arterial streets. The zero-flow control delay (Do) is computed as follows: Equation 8 Where: D0 = the zero-flow control delay at the signal (hours); N = maximum of one, or the number of signals on the segment; 3,600 = conversion from seconds to hours; g/C = average effective green time per cycle for signals on segment; D N C gC0 2 3 600 2 1= ∗ ∗ −( ), DF US ER ’S G UI DE SI Units Facility Type Signals Per Km Free-Flow Speed (km/h) Speed at Capacity (km/h) J Arterial Class I 0.333 80 53 2.21E-05 Arterial Class I 1.000 80 31 2.04E-04 Arterial Class I 2.500 80 15 1.25E-03 Arterial Class II 0.500 64 40 4.99E-05 Arterial Class II 1.000 64 28 2.00E-04 Arterial Class II 2.000 64 18 7.91E-04 Arterial Class III 2.000 56 17 8.02E-04 Arterial Class III 3.000 56 13 1.78E-03 Arterial Class III 4.000 56 10 3.18E-03 Arterial Class IV 4.000 48 10 3.17E-03 Arterial Class IV 5.000 48 8 4.99E-03 Arterial Class IV 6.000 48 7 7.11E-03 Customary Units Facility Type Signals Per Mile Free-Flow Speed (mph) Speed at Capacity (mph) J Arterial Class I 1 50 33.1 2.21E-05 Arterial Class I 2 50 19.3 2.04E-04 Arterial Class I 4 50 9.6 1.25E-03 Arterial Class II 1 40 24.8 4.99E-05 Arterial Class II 2 40 17.8 2.00E-04 Arterial Class II 3 40 11.2 7.91E-04 Arterial Class III 3 35 10.9 8.02E-04 Arterial Class III 5 35 7.9 1.78E-03 Arterial Class III 6 35 6.3 3.18E-03 Arterial Class IV 6 30 6.1 3.17E-03 Arterial Class IV 8 30 5.0 4.99E-03 Arterial Class IV 10 30 4.3 7.11E-03 TABLE 3 Recommended calibration parameters J for signalized streets

C = average cycle length for all signals on the segment (seconds); and DF = delay factor, = 0.9 for uncoordinated traffic-actuated signals, = 1.0 for uncoordinated fixed-time signals, = 1.2 for coordinated signals with unfavorable progression, = 0.9 for coordinated signals with favorable pro- gression, and = 0.6 for coordinated signals with highly favorable progression. If the ratio of green time per cycle for the arterial through movement is not known, a default value of 0.44 can be used. Similarly, if the signal cycle length is not known, then a 11 default value of 120 seconds can be used. A survey of local average signal cycle lengths by area type (e.g., downtown, suburban, and rural) may be desirable to establish appropri- ate local default values. The segment delay between signals (DL) is estimated as follows: Equation 9 Where: L = the length of the segment and dL = the delay per mile, given in Table 5. D L dL L= ∗ 60 USER’S G UIDE SI Units Customary Units Arterial Class Posted Speed Limit Signal Density Posted Speed Limit Signal Density Class I 70-90 km/h 0.3-2.5 signals/km 45-55 mph 0.5-4 signals/mi. Class II 55-70 0.3-3.1 35-45 0.5-5 Class III 50-55 2.5-6.3 30-35 4-10 Class IV 40-50 2.5-12.5 25-35 4-20 Source: Chapter 15, Urban Streets, HCM. Note: There may be instances of overlaps in arterial class definitions. The analyst should consult Chapter 15 of the HCM for additional information on the identification of a specific arterial class. TABLE 4 HCM arterial class criteria Source: 2000 HCM, Exhibit 15-3, Segment Running Time Per Mile. Table computed by subtracting running time if traveling at free-flow speed from running time shown in exhibit. 0.0 0.0 n/a n/a n/a n/a secs/mile Arterial Class: I I I II II II III III IV IV Free-Flow Speed (mph) 55 50 45 45 40 35 35 30 35 30 signal spacing (miles) 0.05 107 0.10 42 35 62 60 0.15 32 21 37 30 0.20 29 25 22 25 14 27 20 0.25 32 28 24 24 20 16 17 7 19 12 0.30 27 23 19 19 12 7 0.40 17 14 14 14 6 2 0.50 8 6 8 8 3 0 1.00 0 0 0 0 0 0 secs/km Arterial Class: I I I II II II III III IV IV Free-Flow Speed (km/h) 88 80 72 72 64 56 56 48 56 48 signal spacing (km) 0.08 n/a n/a n/a n/a n/a n/a n/a n/a n/a 66.9 0.16 n/a n/a n/a n/a n/a n/a 26.3 21.9 38.8 37.5 0.24 n/a n/a n/a n/a n/a n/a 20.1 13.1 23.2 18.8 0.32 n/a n/a n/a 18.1 15.6 13.8 15.7 8.8 17.0 12.5 0.40 19.7 17.5 15.0 15.0 12.5 10.1 10.7 4.4 12.0 7.5 0.48 16.6 14.4 11.9 11.9 7.5 4.5 n/a n/a n/a n/a 0.64 10.3 8.8 8.8 8.8 3.8 1.3 n/a n/a n/a n/a 0.80 4.7 3.8 5.0 5.0 1.9 0.0 n/a n/a n/a n/a 1.60 0.0 0.0 0.0 0.0 TABLE 5 Segment delay between signals

12 US ER ’S G UI DE CHAPTER 4 THE TRAVEL BEHAVIOR RESPONSE MODULE The Portland Tour-Based Model was selected as the basis for the Travel Behavior Response Module because of its abil- ity to predict both modal and temporal shifts in travel behav- ior as well as predict the impact on overall out-of-the-home trip making. The Portland Tour-Based Model is complex, so it is implemented in NCHRP Project 25-21 as a set of elas- ticities rather than as the full model. 4.1 OVERVIEW OF THE PORTLAND TOUR-BASED MODEL The Portland Tour-Based Model was originally developed as part of a project to analyze road pricing policy alternatives in Portland. An overview of the Portland model in a larger context is shown in Figure 2; the tour-based model proper consists of the blocks within the large rectangle. (A full description of the Portland Tour-Based Model is given in Mark Bradley Research and Consulting, A System of Activity- Based Models for Portland, Oregon, Washington, D.C.: Travel Model Improvement Program, U.S. Dept. of Trans- portation, Report No.: DOT-T-99-02, U.S. Environmental Protection Agency, 1998. Consult this reference for details on model structure and coefficients.) A more detailed look at the Portland model is given in Figure 3, which shows information flows between the dif- ferent submodels. The model system is designed to predict the following: • A full-day activity pattern (primary activity and, for tour activities, subtour pattern), • Time of day (outbound, inbound) for home-based tours, • Primary mode and destination, • Work-based subtours, and • Location of intermediate stops. The Portland model is a conceptual descendant of Greig Harvey’s Short-Range Transportation Evaluation Program (STEP) model, with considerable additional detail. A descrip- tion of the STEP model and the theory behind the model is presented in Elizabeth Deakin and Greig Harvey’s Trans- portation Pricing Strategies for California: An Assessment of Congestion, Emissions, Energy and Equity Impacts: Final Report, prepared for the California Air Resources Board, 1996. The model has several features that distinguish it from traditional four-step travel models: • Simultaneous modeling of trip generation, time of day, mode choice, and destination choice. Utilities of lower- level choices (e.g., mode and destination choice) are incorporated in the utilities of higher-level choices (e.g., time of day and primary activity pattern). • Application of the model to individual travelers. This approach, known as sample enumeration when applied to travel survey data, and more generally as microsim- ulation, is considered to be at the forefront of the cur- rent state of the art in travel modeling. Microsimulation allows the incorporation of detailed household and per- son characteristics that can significantly affect travel behavior, such as presence of children in the household and competition for available cars in the household for different trip purposes. • Use of a synthetic sample to develop the base population to which the model is applied. This approach provides the model with a sufficiently large population so that com- plete trip tables can be produced. Sample enumeration approaches based only on travel surveys generally pro- duce results at a much larger scale, such as superdistrict- to-superdistrict trip movements. The synthetic sampling approach has been used for over 25 years. One early application was to the development of a database for research on discrete-choice models. See Gerald Duguay, Woo Jung, and Daniel McFadden, “SYNSAM: A Meth- odology for Synthesizing Household Transportation Sur- vey Data,” Berkeley: Urban Travel Demand Forecasting Project, Working paper no. 7618, September 1976. Syn- thetic sampling is currently used in the TRANSIMS model and in the current version of the STEP model. An additional advantage of the synthetic sampling approach is that it enables disaggregation of benefit and cost esti- mates by socioeconomic category, which is often a sig- nificant issue in transportation policy analysis. 4.2 DERIVATION OF ELASTICITIES The Portland model has several drawbacks in application, chief of which is the length of time required to operate it on

13 USER’S G UIDE even a high-speed computer. Consequently, it was decided to use the Portland model to develop a set of elasticities for pre- dicting small changes in traveler behavior in response to indi- vidual traffic-flow improvement projects. The model was executed several times on a range of travel time saving alter- natives, and the results were used to fit a set of demand/time elasticities. These elasticities were then incorporated into the NCHRP 25-21 methodology. A constant elasticity demand model in the following form was fitted to the Portland model: Equation 10 Where: εmpm′p′ = the elasticity of demand for travel from origin i to destination j by mode m in time period p (denoted by ˜ ˜T T t t ij mp ij mp ij m p ij m p m p m p mp =     ′ ′ ′ ′ ′ ′ ∏ ′ ′ ε Tmpij ) with respect to travel time origin i to destination j by mode m′ in time period p′ (denoted by tijm′p′). For m′ = m and p′ = p, there is an own elasticity; otherwise, the quantity is a (mode or time or mode/time) cross-elasticity. The quantities with tildes represent trips and travel times after some change, and the other quantities represent base case trips and travel times. The equation can be converted to a log-log linear model: Equation 11 Therefore, the elasticities can be estimated by observing the quantities predicted by the Port- land model and running a set of regressions against these results. The approach to generating the necessary data points was straightforward: 1. Define a set of i, j zone pairs to be sampled. These zone pairs were sampled to focus on the areas of interest. For example, given the case study area, the research team focused on movements from within King County to Seattle, from Pierce County to Seattle, and from Sno- homish County to Seattle. Movements to and from Kit- sap County were ignored because the research team believes that the ferry network may not be adequately represented to treat this movement alongside bus tran- sit as a transit mode. 2. Pick a particular zone pair with home zone i and desti- nation zone j. Randomly generate a travel time change in the AM peak period for the auto mode, and run the model only for the population within zone i. Store the relative change in travel time and the relevant changes in travel by mode and time period as a data point. 3. Repeat Step 2 for different values of change to the travel time. 4. Repeat Steps 2 and 3 for different time periods. 5. Repeat Steps 2–4 for different modes. 6. Repeat Steps 2–5 for different i, j zone pairs. 7. Collect the data points and run regressions on the appro- priate variables. The research team believes that the following simplifica- tions were reasonable: • For small travel time changes, the constant elasticity approximation is probably good enough. It can be regarded as a first-order approximation to the demand function. • Capacity improvements are likely to affect the peak peri- ods only. Hence, the main mode shifts are likely to occur T T t tijmp ijmp ijm p ijm p, ˜ , , ˜ and ′ ′ ′ ′ ln ˜ ln ˜T T t t ij mp ij mp m p mp m p ij m p ij m p       =    ′ ′′ ′ ′ ′ ′ ′∑ ε TAZ = traffic analysis zone. LOS = level of service. OD = origin-destination. Input • Employment by sector by TAZ • Sample of households • Modal LOS measures Household-based tour model • Primary activity • Secondary tour choice • Time-of-day choice • Mode/destination choice Work-based subtour model Intermediate stop location model (car driver tours only) Decompose tours to trips Output — OD trip matrices by: • Mode • Time of day • Income group Network Model (trip assignment by mode and time period) Figure 2. Portland Tour-Based Model flow chart.

during the peak periods, and the research team reason- ably ignores off-peak mode shifts. 4.3 ELASTICITIES The final set of elasticities fitted to the Portland Tour- Based Model is shown in Table 6. As shown in the table, a 10-percent decrease in AM peak-period travel time for drive alone would result in the following predicted demand effects: • A 2.25-percent increase in drive alone during the AM peak, • A 0.37-percent decrease in shared ride during the AM peak, • A 0.36-percent decrease in transit riders during the AM peak, • A 1.24-percent increase in drive alone during the PM peak, and • A 1.70-percent increase in drive alone during the off peak. 14 4.4 ALTERNATE METHODS FOR DERIVING ELASTICITIES Most users of the NCHRP 25-21 methodology can prob- ably use the elasticities provided in Table 6 without having to repeat the application of the Portland model to the Seat- tle test bed. However, tour-based models like Portland are a recent development. Little is known about the robustness of their parameters when applied to other areas. Consequently, researchers cannot state with assurance that a particular tour- based model can be applied to similar or dissimilar urban regions. Analysts with greater resources can apply the Portland model or another tour-based model to their own urban region as described in the above sections to see how elasticities derived from application of the tour-based model to their own region vary from those shown in Table 6. Locally derived elasticities would presumably be more reliable than ones bor- rowed from another region, but, again, there is little or no practical experience to back up this conjecture. US ER ’S G UI DE Network supply data by time of day Synthetic population Zonal population and land-use data Full-Day Activity Pattern Home-Based Tour Times of Day Home-Based Tour Mode and Destination Location of Intermediate Stops (car driver tours only) Output: OD trip matrices by mode, purpose, time of day, and income class Work-Based Subtour Models Predicted tours by purpose and chain type Predicted tours by purpose, chain type, and time of day Predicted tours by purpose, chain type, time of day, and mode Accessibility logsum values by tour purpose and tour type Accessibility logsum values by tour purpose, tour type, time of day, mode, and destination (not used in current version of model) Accessibility logsum values by tour purpose, tour type, and time of day Figure 3. Information flows in the Portland Tour-Based Model.

15 USER’S G UIDE Travel Time AM peak PM peak Demand DA SR TR DA SR TR AM peak DA -0.225 0.030 0.010 -0.024 0.000 0.000 SR 0.037 -0.303 0.032 0.000 -0.028 0.000 TR 0.036 0.030 -0.129 0.000 0.000 -0.007 PM peak DA -0.124 0.000 0.000 -0.151 0.015 0.005 SR 0.000 -0.109 0.000 0.019 -0.166 0.016 TR 0.000 0.000 -0.051 0.018 0.015 -0.040 Off peak DA -0.170 0.000 0.000 -0.069 0.000 0.000 SR 0.000 -0.189 0.000 0.000 -0.082 0.000 TR 0.000 0.000 -0.074 0.000 0.000 -0.014 Note: DA = drive alone, SR = shared ride, TR = transit. Source: Portland Tour-Based Model Applied to PSRC data set. Estimates (shown in italics) appear in the table when statistically significant results could not be estimated from the data set. Zero values are shown for cross-elasticities that were deemed (a priori) to be insignificant. TABLE 6 Travel time elasticities

16 US ER ’S G UI DE CHAPTER 5 THE GROWTH REDISTRIBUTION MODULE The Growth Redistribution Module predicts the very long- term impacts of localized travel time changes (caused by traffic-flow improvements) on the geographic distribution of growth in a metropolitan area. There are already several sophisticated land-use models available (such as UrbanSim) that could be used for the purpose of this module. However, these models require a great deal of specialized economic data and effort (which are beyond the resources of many MPOs) to set up for a region. When a sophisticated land-use model exists in a region, it can be used to predict the long-term growth effects. When such a model is not available, the simplified model described here is proposed for use to approximate the long-term land-use effects of traffic-flow improvements. 5.1 MODULE DESCRIPTION The Growth Redistribution Module requires that a baseline 20- to 25-year forecast of land-use growth (i.e., households and employment changes) be available for the metropolitan area. This baseline forecast should have been prepared either manually or with a model, taking into account accessibility changes as well as all of the other factors that commonly affect the distribution of growth within a region. The Growth Redistribution Module consists of a simple linear regression model that is fitted to the baseline forecast. The regression model predicts the change in the growth rate in households and employment in each zone of the region as a function of the relative change of accessibility for each zone. Although not sophisticated enough to predict actual growth, the module should be sufficient to predict how small changes in travel time accessibility can affect the predicted baseline growth rate in specific zones of the region. The mod- ule is as follows: Equation 12 Where: LUinew = predicted sum of the number of households and jobs in zone i after traffic-flow improvement, LUiold = sum of households and jobs in zone i before traffic- flow improvement, LU LU G AA Ri new i old i new i old= ∗ + ∗ −    CP Ainew = predicted AM peak home-based work accessibil- ity of zone i after traffic-flow improvement, Aiold = AM peak home-based work accessibility of zone i before traffic-flow improvement, CP = calibration parameter for model determined from linear regression (CP is the slope of the least- squared error line constrained to go through zero), G = ratio of the total predicted number of households in the region after the traffic-flow improvement divided by the number of households in the region before the improvement, and R = ratio of the total predicted accessibility in the region after the traffic-flow improvement divided by the total accessibility in the region before the improvement. The module presumes that total regional growth will be unaffected by traffic-flow improvements (in other words, the model will not be sensitive to the potential effects of differ- ing levels of regional traffic-flow improvements on the com- petitiveness of regions for attracting new households or jobs). The module predicts only how regional growth might be real- located from marginally less accessible zones to more acces- sible zones within the region. The marginal change in zonal accessibility is obtained by subtracting the average change in regional accessibility from the zone-specific change in acces- sibility (this is accomplished in Equation 12 by subtracting the ratio R from the ratio of new to old accessibility for each zone i). For similar reasons, the amount of household growth that would have normally occurred in a zone (if the zone had grown at the regional average growth rate) is added to the model-predicted growth rate that is due exclusively to mar- ginal changes in the zonal accessibility (this is accomplished in Equation 12 by adding the ratio G). The effect of the above normalization is that if the ratio of the new accessibility to the old accessibility for a zone is less than the average ratio for the entire region, then the zone’s growth will be less than the regional average. If the zonal accessibility ratio is greater than the average regional acces- sibility ratio, then the zone’s growth will be greater than the regional average. The value of G will normally be 1.00, unless there is a significant period of time between the “before” and “after” traffic-flow improvement dates. The ratio G allows the ana-

17 USER’S G UIDE lyst to account for any baseline growth in the region that might have occurred between the “before” condition and the “after” condition that would have occurred with or without the traffic-flow improvement. CP is the calibration parameter that converts a percentage change in zonal accessibility into a percentage change in zonal growth. It is the slope of the regression line fitted to local data on the correlation between the marginal change in zonal acces- sibility and the marginal change in zonal growth expressed as the sum of households and jobs. The measure of zonal accessibility (Ai) is the denominator of the trip distribution gravity model for home-based work trips. The denominator is the sum of the weighted travel time impedances to each destination zone in the region. The AM peak-period accessibility for home-based work trips is used as a proxy for total daily accessibility for all trips, based on the presumption that commute accessibility has the greatest effect on housing and job location decisions. Equation 13 Where: Ai = accessibility of zone i, Tj = total trips generated by zone j, and Fij = AM peak travel time impedance for home-based work travel between zone i and zone j. The impedance is a decreasing function of travel time between zones and takes whatever form was used to calibrate the regional travel demand model. The analyst may experiment with fitting more elaborate linear or nonlinear models to the land-use intensity forecasts. A full-scale land-use forecasting model, like UrbanSim, could be used instead of the simple linear model presented above. Application of a full-scale model like UrbanSim would double or triple the amount of time required to ana- lyze the traffic-flow improvement project. The simple linear model was selected for the sake of efficiency, enabling more rapid computations of the impacts of various traffic-flow improvement projects. A T Fi j ij j = ∗∑ The analyst can also adopt a more elaborate measure of accessibility than the simple gravity model denominator sug- gested above. Ideally, this more elaborate measure should be based upon some kind of trip distribution model for pre- dicting the likelihood that a trip will be made to a particular destination. 5.2 MODULE APPLICATION The Growth Redistribution Module is calibrated for each region in which it is applied. Base and future employment and household forecasts are assembled for the region. A lin- ear regression model of the form shown in Equation 12 is fit- ted to the data to obtain the value of CP. The fitted equation is then used to predict how individual zones will deviate from the regional average growth rate based upon changes in zonal accessibility from the base condition. The following paragraphs illustrate such an application of the module to the Seattle metropolitan area. The PSRC pro- vided household and employment forecasts for the years 1990 and 2020. These forecasts had been produced through a com- bination of inventory (for 1990) and land-use modeling (using Disaggregate Residential Allocation Model/Employment Allo- cation Model [DRAM/EMPAL]) with modifications made in response to local agency input. Accessibility generally improved between the 1990 and 2020 PSRC forecasts; however, some zones experienced sig- nificant changes in accessibility between 1990 and 2020 that varied a great deal from the average (see Figure 4, which plots the percentage change in accessibility for approximately the first 790 of the PSRC zones). The zonal accessibilities for each mode of travel were reported out from the Equilibre Multimodal, Multimodal Equi- librium (EMME2) in which the PSRC model was imple- mented. The reports were then imported into a spreadsheet, which was used to compute the differences between 1990 and 2020 and fit a regression line to the data. A least-squared error regression line was fitted to the 832 zonal data points (see Fig- ure 5). The line was forced through zero. The slope was 0.72, and the resulting correlation coefficient was 67.99 percent.

18 US ER ’S G UI DE -100.0% 0.0% 100.0% 200.0% 300.0% 400.0% 500.0% 1 25 49 73 97 12 1 14 5 16 9 19 3 21 7 24 1 26 5 28 9 31 3 33 7 36 1 38 5 40 9 43 3 45 7 48 1 50 5 52 9 55 3 57 7 60 1 62 5 64 9 67 3 69 7 72 1 74 5 76 9 79 3 Zone % C ha ng e in A cc es si bi lit y y = 0.72x R2 = 0.6799 0% 100% 300% 500% 700% 900% Percent Change Accessibility Beyond Average Change Pe rc en t C ha ng e Jo bs +D w el lin gs B ey on d Av er ag e Ch an ge -200% 200% 400% 600% 800% 1000% -100% Figure 5. Calibration of long-term module to PSRC data. Figure 4. PSRC zonal accessibility changes between 1990 and 2020.

19 USER’S G UIDE CHAPTER 6 THE MODAL ACTIVITY MODULE The purpose of the Modal Activity Module is to calculate the VHT by mode of operation (i.e., cruise, idle, and acceleration/ deceleration), which is defined by speed and acceleration category. The estimates of vehicle activity are then used with modal emission factors (e.g., University of California, Riverside/NCHRP 25-1) to produce the emission estimates. 6.1 METHODOLOGY DEVELOPMENT The methodology for estimating modal activity is largely based on previous research conducted by the investigator under the sponsorship of the California Air Resources Board (CARB). (See Skabardonis. A., “A Modeling Framework for Estimating Emissions in Large Urban Areas,” Transportation Research Record 1587, Transportation Research Board, 1997; and Skabardonis, A., “Feasibility and Demonstration of Net- work Simulation Techniques for Estimation of Emissions in a Large Urban Area,” Final Report, prepared for the California Air Resources Board, DHS Inc., 1994.) This research pro- duced a set of relationships through microscopic simulation that determine the proportion of the time spent Tij on a net- work link i in driving mode j as a function of the link’s type and its volume/capacity ratio. The link classification (type) was based on typical link classifications employed in plan- ning and operational studies (e.g., facility types), and key design/operational characteristics (e.g., number of lanes, free- flow speed, and signal spacing). The relationships were devel- oped through processing of simulated vehicle trajectories using the Integrated Traffic Simulator (INTRAS; the prede- cessor of the Freeway Simulation Model [FRESIM]) and the Traffic Network Simulation (TRAF-NETSIM) microscopic simulation models. Comparisons of simulated and actual field measurements for different facility types, free-flow speeds, and levels of con- gestion showed that a single distribution of time spent versus speed (using the ratio of speed/free-flow speed) could be used to represent many different free-flow speed conditions. Tables 7 through 10 were developed to divide up the total VHT on a link among the various speed and acceleration categories: • Uncongested freeway (volume/capacity [v/c] < 1.00), • Congested freeway (v/c ≥ 1.00), • Uncongested arterial street (v/c < 1.00), and • Congested arterial street (v/c ≥ 1.00). 6.2 METHODOLOGY APPLICATION The methodology requires as input the facility type, the link volume/capacity ratio, and the link VHT. The total VHT on a link is multiplied by the proportions in the appropriate table to obtain the distribution of VHT by speed category and acceleration category.

20 US ER ’S G UI DE TABLE 7 Vehicle modal activity for uncongested freeways ACCELERATION (mph/sec) Spd/FreSpd -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 - - - - - - - - - - - 0.0065 - - - - - - - - - - 0.0167 - - - - - - - - 0.0003 0.0007 0.0008 0.0007 0.0002 - - - - - - - - 0.0333 - - - - - - - - 0.0003 0.0007 0.0008 0.0007 0.0002 - - - - - - - - 0.0500 - - - - - - - - 0.0003 0.0007 0.0008 0.0007 0.0002 - - - - - - - - 0.0667 - - - - - - - - 0.0003 0.0007 0.0008 0.0007 0.0002 - - - - - - - - 0.0833 - - - - - - - - 0.0003 0.0007 0.0008 0.0007 0.0002 - - - - - - - - 0.1000 - - - - - - - 0.0002 0.0004 0.0006 0.0008 0.0008 0.0002 - - - - - - - - 0.1167 - - - - - - - 0.0002 0.0004 0.0006 0.0008 0.0008 0.0002 - - - - - - - - 0.1333 - - - - - - - 0.0002 0.0004 0.0006 0.0008 0.0008 0.0002 - - - - - - - - 0.1500 - - - - - - - 0.0002 0.0004 0.0006 0.0008 0.0008 0.0002 - - - - - - - - 0.1667 - - - - - - - 0.0002 0.0004 0.0006 0.0008 0.0008 0.0002 - - - - - - - - 0.1833 - - - - - - 0.0002 0.0002 0.0004 0.0008 0.0012 0.0010 0.0004 0.0002 - - - - - - - 0.2000 - - - - - - 0.0002 0.0002 0.0004 0.0008 0.0012 0.0010 0.0004 0.0002 - - - - - - - 0.2167 - - - - - - 0.0002 0.0002 0.0004 0.0008 0.0012 0.0010 0.0004 0.0002 - - - - - - - 0.2333 - - - - - - 0.0002 0.0002 0.0004 0.0008 0.0012 0.0010 0.0004 0.0002 - - - - - - - 0.2500 - - - - - - 0.0002 0.0002 0.0004 0.0008 0.0012 0.0010 0.0004 0.0002 - - - - - - - 0.2667 - - - - - - 0.0002 0.0002 0.0004 0.0010 0.0014 0.0010 0.0004 0.0002 - - - - - - - 0.2833 - - - - - - 0.0002 0.0002 0.0004 0.0010 0.0014 0.0010 0.0004 0.0002 - - - - - - - 0.3000 - - - - - - 0.0002 0.0002 0.0004 0.0010 0.0014 0.0010 0.0004 0.0002 - - - - - - - 0.3167 - - - - - - 0.0002 0.0002 0.0004 0.0010 0.0014 0.0010 0.0004 0.0002 - - - - - - - 0.3333 - - - - - - 0.0002 0.0002 0.0004 0.0010 0.0014 0.0010 0.0004 0.0002 - - - - - - - 0.3500 - - - - - - 0.0002 0.0002 0.0004 0.0010 0.0016 0.0012 0.0004 - - - - - - - - 0.3667 - - - - - - 0.0002 0.0002 0.0004 0.0010 0.0016 0.0012 0.0004 - - - - - - - - 0.3833 - - - - - - 0.0002 0.0002 0.0004 0.0010 0.0016 0.0012 0.0004 - - - - - - - - 0.4000 - - - - - - 0.0002 0.0002 0.0004 0.0010 0.0016 0.0012 0.0004 - - - - - - - - 0.4167 - - - - - - 0.0002 0.0002 0.0004 0.0010 0.0016 0.0012 0.0004 - - - - - - - - 0.4333 - - - - - - 0.0002 0.0002 0.0004 0.0012 0.0018 0.0014 0.0004 - - - - - - - - 0.4500 - - - - - - 0.0002 0.0002 0.0004 0.0012 0.0018 0.0014 0.0004 - - - - - - - - 0.4667 - - - - - - 0.0002 0.0002 0.0004 0.0012 0.0018 0.0014 0.0004 - - - - - - - - 0.4833 - - - - - - 0.0002 0.0002 0.0004 0.0012 0.0018 0.0014 0.0004 - - - - - - - - 0.5000 - - - - - - 0.0002 0.0002 0.0004 0.0012 0.0018 0.0014 0.0004 - - - - - - - - 0.5167 - - - - - - 0.0002 0.0002 0.0004 0.0014 0.0022 0.0016 0.0004 - - - - - - - - 0.5333 - - - - - - 0.0002 0.0002 0.0004 0.0014 0.0022 0.0016 0.0004 - - - - - - - - 0.5500 - - - - - - 0.0002 0.0002 0.0004 0.0014 0.0022 0.0016 0.0004 - - - - - - - - 0.5667 - - - - - - 0.0002 0.0002 0.0004 0.0014 0.0022 0.0016 0.0004 - - - - - - - - 0.5833 - - - - - - 0.0002 0.0002 0.0004 0.0014 0.0022 0.0016 0.0004 - - - - - - - - 0.6000 - - - - - - - 0.0002 0.0004 0.0020 0.0022 0.0020 0.0004 - - - - - - - - 0.6167 - - - - - - - 0.0002 0.0004 0.0020 0.0022 0.0020 0.0004 - - - - - - - - 0.6333 - - - - - - - 0.0002 0.0004 0.0020 0.0022 0.0020 0.0004 - - - - - - - - 0.6500 - - - - - - - 0.0002 0.0004 0.0020 0.0022 0.0020 0.0004 - - - - - - - - 0.6667 - - - - - - - 0.0002 0.0004 0.0020 0.0022 0.0020 0.0004 - - - - - - - - 0.6833 - - - - - - 0.0002 0.0002 0.0004 0.0020 0.0024 0.0022 0.0002 - - - - - - - - 0.7000 - - - - - - 0.0002 0.0002 0.0004 0.0020 0.0024 0.0022 0.0002 - - - - - - - - 0.7167 - - - - - - 0.0002 0.0002 0.0004 0.0020 0.0024 0.0022 0.0002 - - - - - - - - 0.7333 - - - - - - 0.0002 0.0002 0.0004 0.0020 0.0024 0.0022 0.0002 - - - - - - - - 0.7500 - - - - - - 0.0003 0.0003 0.0006 0.0030 0.0035 0.0033 0.0003 - - - - - - - - 0.7667 - - - - - - 0.0001 0.0001 0.0002 0.0021 0.0038 0.0021 0.0002 - - - - - - - - 0.7833 - - - - - - 0.0001 0.0001 0.0003 0.0033 0.0059 0.0033 0.0003 - - - - - - - - 0.8000 - - - - - - 0.0001 0.0001 0.0003 0.0033 0.0059 0.0033 0.0003 - - - - - - - - 0.8167 - - - - - - 0.0001 0.0001 0.0002 0.0026 0.0046 0.0026 0.0002 - - - - - - - - 0.8333 - - - - - - 0.0001 0.0001 0.0002 0.0027 0.0048 0.0027 0.0002 - - - - - - - - 0.8500 - - - - - 0.0001 0.0001 0.0001 0.0003 0.0055 0.0156 0.0055 0.0003 - - - - - - - - 0.8667 - - - - - 0.0001 0.0001 0.0001 0.0002 0.0040 0.0113 0.0040 0.0002 - - - - - - - - 0.8833 - - - - - 0.0001 0.0001 0.0001 0.0003 0.0052 0.0146 0.0052 0.0003 - - - - - - - - 0.9000 - - - - - 0.0001 0.0001 0.0001 0.0003 0.0065 0.0184 0.0065 0.0003 - - - - - - - - 0.9167 - - - - - 0.0001 0.0001 0.0001 0.0004 0.0073 0.0207 0.0073 0.0004 - - - - - - - - 0.9333 - - - - - - - - 0.0003 0.0070 0.0225 0.0080 0.0004 0.0001 - - - - - - - 0.9500 - - - - - - - - 0.0004 0.0092 0.0294 0.0105 0.0006 0.0002 - - - - - - - 0.9667 - - - - - - - - 0.0004 0.0104 0.0333 0.0119 0.0006 0.0002 - - - - - - - 0.9833 - - - - - - - - 0.0005 0.0127 0.0406 0.0145 0.0008 0.0003 - - - - - - - 1.0000 - - - - - - - - 0.0017 0.0417 0.1339 0.0476 0.0025 0.0008 - - - - - - - 1.0167 - - - - - - - - 0.0006 0.0116 0.0385 0.0153 0.0006 0.0006 - - - - - - - 1.0333 - - - - - - - - 0.0003 0.0055 0.0184 0.0073 0.0003 0.0003 - - - - - - - 1.0500 - - - - - - - - 0.0001 0.0019 0.0064 0.0026 0.0001 0.0001 - - - - - - - 1.0667 - - - - - - - - - 0.0005 0.0018 0.0007 - - - - - - - - - 1.0833 - - - - - - - - - 0.0002 0.0008 0.0003 - - - - - - - - - 1.1000 - - - - - - - - - - - - - - - - - - - - - Note: entries are the proportion of total vehicle-hours on the link that fall in each speed/acceleration category. Columns are the acceleration rate category in units of miles per hour per second. Rows are the speed category expressed as a ratio of the link free-flow speed. Spd/FreSpd = ratio of speed over free-flow speed.

21 USER’S G UIDE TABLE 8 Vehicle modal activity for congested freeway sections ACCELERATION (mph/sec) Spd/FreSpd -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 0.0000 - - - - - 0.0002 0.0004 0.0008 0.0017 0.0025 0.0443 - - - - - - - - - 0.0167 - - - - - 0.0001 0.0001 0.0005 0.0010 0.0015 0.0013 0.0019 - - - - - - - - 0.0333 - - - - - - 0.0002 0.0006 0.0008 0.0023 0.0025 0.0010 0.0017 - - - - - - - 0.0500 - - - - - 0.0001 0.0002 0.0003 0.0011 0.0030 0.0049 0.0016 0.0010 0.0010 - - - - - - 0.0667 - - - - - 0.0001 0.0006 0.0006 0.0023 0.0054 0.0122 0.0033 0.0013 0.0005 0.0007 - - - - - 0.0833 - - - - - 0.0002 0.0004 0.0005 0.0022 0.0057 0.0163 0.0056 0.0020 0.0006 0.0003 0.0002 - - - - 0.1000 - - - - - 0.0001 0.0002 0.0010 0.0018 0.0044 0.0126 0.0053 0.0013 0.0006 0.0003 0.0001 - - - - 0.1167 - - - - - - 0.0003 0.0007 0.0015 0.0039 0.0095 0.0052 0.0017 0.0005 0.0002 0.0001 - - - - 0.1333 - - - - - 0.0001 0.0003 0.0006 0.0017 0.0038 0.0091 0.0049 0.0015 0.0006 0.0002 - - - - - 0.1500 - - - - - 0.0001 0.0004 0.0007 0.0018 0.0039 0.0080 0.0042 0.0018 0.0006 0.0003 - - - - - 0.1667 - - - - - 0.0001 0.0003 0.0009 0.0022 0.0051 0.0100 0.0061 0.0023 0.0007 0.0002 0.0001 - - - - 0.1833 - - - - - - 0.0003 0.0007 0.0014 0.0044 0.0090 0.0046 0.0017 0.0006 0.0002 - - - - - 0.2000 - - - - - 0.0001 0.0003 0.0007 0.0016 0.0040 0.0094 0.0047 0.0021 0.0006 0.0001 - - - - - 0.2167 - - - - - 0.0002 0.0006 0.0009 0.0026 0.0071 0.0126 0.0068 0.0031 0.0008 0.0002 - - - - - 0.2333 - - - - - - 0.0004 0.0005 0.0016 0.0047 0.0085 0.0056 0.0023 0.0006 0.0001 - - - - - 0.2500 - - - - - 0.0001 0.0003 0.0009 0.0024 0.0054 0.0122 0.0054 0.0027 0.0006 0.0002 - - - - - 0.2667 - - - - - 0.0001 0.0002 0.0005 0.0016 0.0043 0.0089 0.0053 0.0022 0.0007 0.0001 0.0001 - - - - 0.2833 - - - - - 0.0001 0.0003 0.0008 0.0018 0.0050 0.0113 0.0052 0.0022 0.0008 0.0001 - - - - - 0.3000 - - - - 0.0001 - 0.0002 0.0007 0.0015 0.0048 0.0106 0.0057 0.0021 0.0004 0.0001 - - - - - 0.3167 - - - - 0.0001 - 0.0003 0.0008 0.0017 0.0044 0.0107 0.0056 0.0020 0.0006 0.0001 - - - - - 0.3333 - - - - - - 0.0003 0.0007 0.0021 0.0053 0.0125 0.0066 0.0026 0.0008 0.0001 - - - - - 0.3500 - - - - - 0.0001 0.0004 0.0005 0.0013 0.0045 0.0125 0.0059 0.0024 0.0004 - - - - - - 0.3667 - - - - 0.0001 0.0001 0.0002 0.0004 0.0025 0.0069 0.0144 0.0075 0.0025 0.0008 - - - - - - 0.3833 - - - - - 0.0001 0.0002 0.0004 0.0012 0.0038 0.0107 0.0056 0.0020 0.0004 0.0001 - - - - - 0.4000 - - - - - - 0.0003 0.0004 0.0015 0.0047 0.0093 0.0050 0.0020 0.0003 0.0001 - - - - - 0.4167 - - - - - - 0.0001 0.0002 0.0012 0.0043 0.0115 0.0059 0.0018 0.0003 0.0001 - - - - - 0.4333 - - - - - 0.0002 0.0002 0.0004 0.0014 0.0042 0.0130 0.0066 0.0014 0.0003 0.0001 - - - - - 0.4500 - - - - - - 0.0002 0.0003 0.0012 0.0034 0.0092 0.0045 0.0017 0.0003 - - - - - - 0.4667 - - - - - - 0.0002 0.0004 0.0009 0.0033 0.0091 0.0046 0.0011 0.0002 - - - - - - 0.4833 - - - - - 0.0001 0.0002 0.0004 0.0010 0.0033 0.0076 0.0045 0.0014 0.0003 - - - - - - 0.5000 - - - - 0.0001 0.0001 0.0002 0.0003 0.0012 0.0045 0.0105 0.0055 0.0014 0.0003 0.0001 - - - - - 0.5167 - - - - - 0.0001 0.0002 0.0004 0.0013 0.0034 0.0109 0.0051 0.0013 0.0003 - - - - - - 0.5333 - - - - - - 0.0001 0.0002 0.0005 0.0025 0.0077 0.0038 0.0008 0.0002 0.0001 - - - - - 0.5500 - - - - - - 0.0002 0.0003 0.0007 0.0028 0.0067 0.0039 0.0012 0.0002 - - - - - - 0.5667 - - - - - 0.0001 0.0001 0.0004 0.0006 0.0019 0.0063 0.0032 0.0007 0.0001 - - - - - - 0.5833 - - - - - 0.0001 - 0.0003 0.0006 0.0015 0.0047 0.0029 0.0009 0.0001 - - - - - - 0.6000 - - - - - 0.0001 0.0002 0.0002 0.0006 0.0023 0.0080 0.0037 0.0009 0.0001 - - - - - - 0.6167 - - - - - - 0.0002 0.0003 0.0006 0.0020 0.0064 0.0033 0.0005 0.0001 - - - - - - 0.6333 - - - - - - 0.0001 0.0001 0.0006 0.0015 0.0057 0.0026 0.0008 0.0001 - - - - - - 0.6500 - - - - - - - 0.0002 0.0009 0.0022 0.0054 0.0036 0.0011 0.0002 - - - - - - 0.6667 - - - - - - - 0.0002 0.0001 0.0007 0.0022 0.0012 0.0003 0.0001 - - - - - - 0.6833 - - - - - - 0.0001 0.0001 0.0002 0.0009 0.0035 0.0014 0.0003 - - - - - - - 0.7000 - - - - - - - 0.0001 0.0006 0.0011 0.0033 0.0017 0.0006 0.0001 - - - - - - 0.7167 - - - - - - 0.0001 0.0001 0.0004 0.0013 0.0027 0.0018 0.0004 0.0001 - - - - - - 0.7333 - - - - - - - 0.0001 0.0005 0.0015 0.0035 0.0019 0.0006 - - - - - - - 0.7500 - - - - - - 0.0001 0.0002 0.0003 0.0013 0.0044 0.0023 0.0004 - - - - - - - 0.7667 - - - - - 0.0001 0.0001 0.0002 0.0002 0.0020 0.0050 0.0021 0.0003 0.0001 - - - - - - 0.7833 - - - - - - 0.0001 0.0002 0.0006 0.0016 0.0046 0.0023 0.0007 0.0001 - - - - - - 0.8000 - - - - - - - 0.0002 0.0002 0.0014 0.0048 0.0017 0.0006 - - - - - - - 0.8167 - - - - - - 0.0001 0.0001 0.0002 0.0013 0.0043 0.0017 0.0003 0.0001 - - - - - - 0.8333 - - - - - - - 0.0001 - 0.0004 0.0019 0.0006 0.0001 - - - - - - - 0.8500 - - - - - - - - - - - - - - - - - - - - 0.8667 - - - - - - - - - - - - - - - - - - - - 0.8833 - - - - - - - - - - - - - - - - - - - - 0.9000 - - - - - - - - - - - - - - - - - - - - 0.9167 - - - - - - - - - - - - - - - - - - - - 0.9333 - - - - - - - - - - - - - - - - - - - - 0.9500 - - - - - - - - - - - - - - - - - - - - 0.9667 - - - - - - - - - - - - - - - - - - - - 0.9833 - - - - - - - - - - - - - - - - - - - - 1.0000 - - - - - - - - - - - - - - - - - - - - 1.0167 - - - - - - - - - - - - - - - - - - - - 1.0333 - - - - - - - - - - - - - - - - - - - - 1.0500 - - - - - - - - - - - - - - - - - - - - 1.0667 - - - - - - - - - - - - - - - - - - - - 1.0833 - - - - - - - - - - - - - - - - - - - - 1.1000 - - - - - - - - - - - - - - - - - - - - Note: entries are the proportion of total vehicle-hours on the link that fall in each speed/acceleration category. Columns are the acceleration rate category in units of miles per hour per second. Rows are the speed category expressed as a ratio of the link free-flow speed. Spd/FreSpd = ratio of speed over free-flow speed.

22 US ER ’S G UI DE TABLE 9 Vehicle modal activity for uncongested arterials ACCELERATION (mph/sec) Spd/FreSpd -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 0.000 - - - - - - - - - - 0.2006 - - - - - - - - - 0.0286 - - - - - 0.0001 - - 0.0008 0.0005 0.0009 0.0005 0.0001 0.0002 0.0001 0.0005 - - - - 0.0571 - - - - - 0.0002 - - 0.0015 0.0010 0.0018 0.0009 0.0002 0.0004 0.0001 0.0011 - - - - 0.0857 - - - - - 0.0002 - - 0.0015 0.0010 0.0018 0.0009 0.0002 0.0004 0.0001 0.0011 - - - - 0.1143 - - - - - 0.0002 - - 0.0015 0.0010 0.0018 0.0009 0.0002 0.0004 0.0001 0.0011 - - - - 0.1429 - - - - - 0.0002 - - 0.0015 0.0010 0.0018 0.0009 0.0002 0.0004 0.0001 0.0011 - - - - 0.1714 - - - - - 0.0002 0.0001 0.0003 0.0012 0.0009 0.0020 0.0012 0.0005 0.0006 0.0006 0.0005 0.0002 - - - 0.2000 - - - - - 0.0002 0.0002 0.0003 0.0014 0.0011 0.0023 0.0014 0.0005 0.0007 0.0007 0.0006 0.0002 - - - 0.2286 - - - - - 0.0002 0.0002 0.0003 0.0014 0.0011 0.0023 0.0014 0.0005 0.0007 0.0007 0.0006 0.0002 - - - 0.2571 - - - - - 0.0002 0.0002 0.0003 0.0014 0.0011 0.0023 0.0014 0.0005 0.0007 0.0007 0.0006 0.0002 - - - 0.2857 - - - - - 0.0002 0.0002 0.0003 0.0014 0.0011 0.0023 0.0014 0.0005 0.0007 0.0007 0.0006 0.0002 - - - 0.3143 - - - - - 0.0002 0.0004 0.0004 0.0013 0.0010 0.0016 0.0010 0.0004 0.0011 0.0011 0.0004 0.0003 - - - 0.3429 - - - - - 0.0002 0.0004 0.0004 0.0012 0.0009 0.0015 0.0010 0.0004 0.0010 0.0011 0.0004 0.0003 - - - 0.3714 - - - - - 0.0002 0.0004 0.0004 0.0012 0.0009 0.0015 0.0010 0.0004 0.0010 0.0011 0.0004 0.0003 - - - 0.4000 - - - - - 0.0002 0.0004 0.0004 0.0012 0.0009 0.0015 0.0010 0.0004 0.0010 0.0011 0.0004 0.0003 - - - 0.4286 - - - - - 0.0002 0.0004 0.0004 0.0012 0.0009 0.0015 0.0010 0.0004 0.0010 0.0011 0.0004 0.0003 - - - 0.4571 - - - - - 0.0002 0.0005 0.0003 0.0012 0.0009 0.0018 0.0011 0.0005 0.0028 0.0002 - 0.0003 - - - 0.4857 - - - - - 0.0002 0.0006 0.0003 0.0013 0.0010 0.0020 0.0012 0.0006 0.0031 0.0002 - 0.0003 - - - 0.5143 - - - - - 0.0002 0.0006 0.0003 0.0013 0.0010 0.0020 0.0012 0.0006 0.0031 0.0002 - 0.0003 - - - 0.5429 - - - - - 0.0002 0.0006 0.0003 0.0013 0.0010 0.0020 0.0012 0.0006 0.0031 0.0002 - 0.0003 - - - 0.5714 - - - - - 0.0002 0.0006 0.0003 0.0013 0.0010 0.0020 0.0012 0.0006 0.0031 0.0002 - 0.0003 - - - 0.6000 - - - - - 0.0001 0.0004 0.0004 0.0011 0.0009 0.0020 0.0022 0.0019 0.0025 - - - - - - 0.6286 - - - - - 0.0002 0.0005 0.0004 0.0012 0.0009 0.0022 0.0024 0.0020 0.0027 - - - - - - 0.6571 - - - - - 0.0002 0.0005 0.0004 0.0012 0.0009 0.0022 0.0024 0.0020 0.0027 - - - - - - 0.6857 - - - - - 0.0002 0.0005 0.0004 0.0012 0.0009 0.0022 0.0024 0.0020 0.0027 - - - - - - 0.7143 - - - - - 0.0002 0.0005 0.0004 0.0012 0.0009 0.0022 0.0024 0.0020 0.0027 - - - - - - 0.7429 - - - - - 0.0001 0.0005 0.0003 0.0010 0.0009 0.0043 0.0065 0.0025 0.0006 - - - - - - 0.7714 - - - - - 0.0002 0.0006 0.0003 0.0013 0.0012 0.0054 0.0081 0.0031 0.0008 - - - - - - 0.8000 - - - - - 0.0002 0.0006 0.0003 0.0013 0.0012 0.0054 0.0081 0.0031 0.0008 - - - - - - 0.8286 - - - - - 0.0002 0.0006 0.0003 0.0013 0.0012 0.0054 0.0081 0.0031 0.0008 - - - - - - 0.8571 - - - - - 0.0002 0.0006 0.0003 0.0013 0.0012 0.0054 0.0081 0.0031 0.0008 - - - - - - 0.8857 - - - - - 0.0001 0.0003 0.0002 0.0010 0.0020 0.0209 0.0175 0.0010 - - - - - - - 0.9143 - - - - - 0.0001 0.0003 0.0002 0.0012 0.0023 0.0253 0.0198 0.0011 - - - - - - - 0.9429 - - - - - 0.0002 0.0005 0.0004 0.0019 0.0039 0.0395 0.0330 0.0018 0.0001 - - - - - - 0.9714 - - - - - 0.0001 0.0004 0.0003 0.0015 0.0031 0.0316 0.0264 0.0015 0.0001 - - - - - - 1.0000 - - - - - 0.0001 0.0004 0.0003 0.0015 0.0031 0.0316 0.0264 0.0015 0.0001 - - - - - - 1.0286 - - - - - 0.0001 0.0003 0.0003 0.0011 0.0016 0.0214 0.0172 0.0002 - - - - - - - 1.0571 - - - - - - 0.0001 0.0001 0.0005 0.0007 0.0098 0.0079 0.0001 - - - - - - - 1.0857 - - - - - - 0.0001 0.0001 0.0005 0.0007 0.0098 0.0079 0.0001 - - - - - - - 1.1143 - - - - - - 0.0001 0.0001 0.0005 0.0007 0.0098 0.0079 0.0001 - - - - - - - 1.1429 - - - - - - 0.0001 0.0001 0.0005 0.0007 0.0098 0.0079 0.0001 - - - - - - - 1.1714 - - - - - 0.0001 0.0001 0.0001 0.0003 0.0002 0.0073 0.0038 - - - - - - - - 1.2000 - - - - - - - - 0.0001 0.0001 0.0027 0.0014 - - - - - - - - 1.2286 - - - - - - - - 0.0001 0.0001 0.0027 0.0014 - - - - - - - - 1.2571 - - - - - - - - 0.0001 0.0001 0.0027 0.0014 - - - - - - - - 1.2857 - - - - - - - - 0.0001 0.0001 0.0027 0.0014 - - - - - - - - 1.3143 - - - - - - - - - 0.0001 0.0015 0.0011 - - - - - - - - 1.3429 - - - - - - - - - 0.0001 0.0008 0.0006 - - - - - - - - 1.3714 - - - - - - - - - 0.0001 0.0008 0.0006 - - - - - - - - 1.4000 - - - - - - - - - - 0.0006 0.0005 - - - - - - - - 1.4286 - - - - - - - - - - 0.0006 0.0005 - - - - - - - - 1.4571 - - - - - - - - - - - - - - - - - - - - 1.4857 - - - - - - - - - - - - - - - - - - - - 1.5143 - - - - - - - - - - - - - - - - - - - - 1.5429 - - - - - - - - - - - - - - - - - - - - 1.5714 - - - - - - - - - - - - - - - - - - - - 1.6000 - - - - - - - - - - - - - - - - - - - - 1.6286 - - - - - - - - - - - - - - - - - - - - 1.6571 - - - - - - - - - - - - - - - - - - - - 1.6857 - - - - - - - - - - - - - - - - - - - - 1.7143 - - - - - - - - - - - - - - - - - - - - 1.7429 - - - - - - - - - - - - - - - - - - - - 1.7714 - - - - - - - - - - - - - - - - - - - - 1.8000 - - - - - - - - - - - - - - - - - - - - 1.8286 - - - - - - - - - - - - - - - - - - - - 1.8571 - - - - - - - - - - - - - - - - - - - - 1.8857 - - - - - - - - - - - - - - - - - - - - Note: entries are the proportion of total vehicle-hours on the link that fall in each speed/acceleration category. Columns are the acceleration rate category in units of miles per hour per second. Rows are the speed category expressed as a ratio of the link free-flow speed. Spd/FreSpd = ratio of speed over free-flow speed.

23 USER’S G UIDE TABLE 10 Vehicle modal activity for congested arterials Spd/FreSpd -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 0.0000 - - - - - - - - - - 0.5317 - - - - - - - 0.0286 - - - - - 0.0001 - - 0.0013 0.0006 0.0007 0.0002 0.0001 0.0003 0.0001 0.0009 - - 0.0571 - - - - - 0.0003 - - 0.0025 0.0012 0.0013 0.0003 0.0003 0.0006 0.0002 0.0017 - - 0.0857 - - - - - 0.0003 - - 0.0025 0.0012 0.0013 0.0003 0.0003 0.0006 0.0002 0.0017 - - 0.1143 - - - - - 0.0003 - - 0.0025 0.0012 0.0013 0.0003 0.0003 0.0006 0.0002 0.0017 - - 0.1429 - - - - - 0.0003 - - 0.0025 0.0012 0.0013 0.0003 0.0003 0.0006 0.0002 0.0017 - - 0.1714 - - - - - 0.0001 0.0001 0.0007 0.0017 0.0008 0.0018 0.0012 0.0005 0.0009 0.0010 0.0004 0.0002 - 0.2000 - - - - - 0.0001 0.0001 0.0008 0.0019 0.0009 0.0019 0.0013 0.0006 0.0010 0.0011 0.0005 0.0003 - 0.2286 - - - - - 0.0001 0.0001 0.0008 0.0019 0.0009 0.0019 0.0013 0.0006 0.0010 0.0011 0.0005 0.0003 - 0.2571 - - - - - 0.0001 0.0001 0.0008 0.0019 0.0009 0.0019 0.0013 0.0006 0.0010 0.0011 0.0005 0.0003 - 0.2857 - - - - - 0.0001 0.0001 0.0008 0.0019 0.0009 0.0019 0.0013 0.0006 0.0010 0.0011 0.0005 0.0003 - 0.3143 - - - - - 0.0002 0.0009 0.0007 0.0013 0.0007 0.0012 0.0008 0.0002 0.0017 0.0014 0.0003 0.0004 - 0.3429 - - - - - 0.0002 0.0009 0.0007 0.0012 0.0006 0.0012 0.0007 0.0002 0.0016 0.0013 0.0003 0.0004 - 0.3714 - - - - - 0.0002 0.0009 0.0007 0.0012 0.0006 0.0012 0.0007 0.0002 0.0016 0.0013 0.0003 0.0004 - 0.4000 - - - - - 0.0002 0.0009 0.0007 0.0012 0.0006 0.0012 0.0007 0.0002 0.0016 0.0013 0.0003 0.0004 - 0.4286 - - - - - 0.0002 0.0009 0.0007 0.0012 0.0006 0.0012 0.0007 0.0002 0.0016 0.0013 0.0003 0.0004 - 0.4571 - - - - - 0.0002 0.0010 0.0005 0.0012 0.0006 0.0007 0.0008 0.0008 0.0040 0.0002 0.0001 0.0003 - 0.4857 - - - - - 0.0002 0.0011 0.0006 0.0014 0.0007 0.0008 0.0008 0.0009 0.0045 0.0002 0.0001 0.0003 - 0.5143 - - - - - 0.0002 0.0011 0.0006 0.0014 0.0007 0.0008 0.0008 0.0009 0.0045 0.0002 0.0001 0.0003 - 0.5429 - - - - - 0.0002 0.0011 0.0006 0.0014 0.0007 0.0008 0.0008 0.0009 0.0045 0.0002 0.0001 0.0003 - 0.5714 - - - - - 0.0002 0.0011 0.0006 0.0014 0.0007 0.0008 0.0008 0.0009 0.0045 0.0002 0.0001 0.0003 - 0.6000 - - - - - 0.0002 0.0009 0.0003 0.0012 0.0005 0.0010 0.0035 0.0027 0.0019 - - 0.0001 - 0.6286 - - - - - 0.0002 0.0010 0.0003 0.0012 0.0006 0.0011 0.0037 0.0029 0.0020 - - 0.0001 - 0.6571 - - - - - 0.0002 0.0010 0.0003 0.0012 0.0006 0.0011 0.0037 0.0029 0.0020 - - 0.0001 - 0.6857 - - - - - 0.0002 0.0010 0.0003 0.0012 0.0006 0.0011 0.0037 0.0029 0.0020 - - 0.0001 - 0.7143 - - - - - 0.0002 0.0010 0.0003 0.0012 0.0006 0.0011 0.0037 0.0029 0.0020 - - 0.0001 - 0.7429 - - - - - 0.0001 0.0006 0.0003 0.0005 0.0007 0.0082 0.0075 0.0019 0.0001 - - - - 0.7714 - - - - - 0.0001 0.0008 0.0004 0.0007 0.0009 0.0110 0.0101 0.0025 0.0002 - - - - 0.8000 - - - - - 0.0001 0.0008 0.0004 0.0007 0.0009 0.0110 0.0101 0.0025 0.0002 - - - - 0.8286 - - - - - 0.0001 0.0008 0.0004 0.0007 0.0009 0.0110 0.0101 0.0025 0.0002 - - - - 0.8571 - - - - - 0.0001 0.0008 0.0004 0.0007 0.0009 0.0110 0.0101 0.0025 0.0002 - - - - 0.8857 - - - - - 0.0001 0.0004 0.0002 0.0009 0.0005 0.0069 0.0094 0.0005 - - - - - 0.9143 - - - - - - 0.0002 0.0001 0.0004 0.0002 0.0031 0.0042 0.0002 - - - - - 0.9429 - - - - - 0.0001 0.0003 0.0002 0.0007 0.0004 0.0051 0.0070 0.0004 - - - - - 0.9714 - - - - - 0.0001 0.0002 0.0001 0.0006 0.0003 0.0041 0.0056 0.0003 - - - - - 1.0000 - - - - - 0.0001 0.0002 0.0001 0.0006 0.0003 0.0041 0.0056 0.0003 - - - - - 1.0286 - - - - - 0.0001 0.0001 0.0001 0.0003 0.0003 0.0029 0.0033 - - - - - - 1.0571 - - - - - - - - 0.0001 0.0001 0.0011 0.0012 - - - - - - 1.0857 - - - - - - - - 0.0001 0.0001 0.0011 0.0012 - - - - - - 1.1143 - - - - - - - - 0.0001 0.0001 0.0011 0.0012 - - - - - - 1.1429 - - - - - - - - 0.0001 0.0001 0.0011 0.0012 - - - - - - 1.1714 - - - - - - - - - 0.0001 0.0004 0.0009 - - - - - - 1.2000 - - - - - - - - - - 0.0001 0.0002 - - - - - - 1.2286 - - - - - - - - - - 0.0001 0.0002 - - - - - - 1.2571 - - - - - - - - - - 0.0001 0.0002 - - - - - - 1.2857 - - - - - - - - - - 0.0001 0.0002 - - - - - - 1.3143 - - - - - - - - - - - - - - - - - - 1.3429 - - - - - - - - - - - - - - - - - - 1.3714 - - - - - - - - - - - - - - - - - - 1.4000 - - - - - - - - - - - - - - - - - - 1.4286 - - - - - - - - - - - - - - - - - - 1.4571 - - - - - - - - - - - - - - - - - - 1.4857 - - - - - - - - - - - - - - - - - - 1.5143 - - - - - - - - - - - - - - - - - - 1.5429 - - - - - - - - - - - - - - - - - - 1.5714 - - - - - - - - - - - - - - - - - - 1.6000 - - - - - - - - - - - - - - - - - - 1.6286 - - - - - - - - - - - - - - - - - - 1.6571 - - - - - - - - - - - - - - - - - - 1.6857 - - - - - - - - - - - - - - - - - - 1.7143 - - - - - - - - - - - - - - - - - - 1.7429 - - - - - - - - - - - - - - - - - - 1.7714 - - - - - - - - - - - - - - - - - - 1.8000 - - - - - - - - - - - - - - - - - - 1.8286 - - - - - - - - - - - - - - - - - - 1.8571 - - - - - - - - - - - - - - - - - - 1.8857 - - - - - - - - - - - - - - - - - - 1.9143 - - - - - - - - - - - - - - - - - - 1.9429 - - - - - - - - - - - - - - - - - - Note: entries are the proportion of total vehicle-hours on the link that fall in each speed/acceleration category. Columns are the acceleration rate category in units of miles per hour per second. Rows are the speed category expressed as a ratio of the link free-flow speed. Spd/FreSpd = ratio of speed over free-flow speed.

24 US ER ’S G UI DE CHAPTER 7 THE VEHICLE EMISSION MODULE This chapter describes the method for estimating vehicle emissions based on VHT by speed and acceleration category. 7.1 METHODOLOGY DEVELOPMENT The underlying concept for traditional on-road emission inventory development using composite emission factors expressed in grams per mile can be thought of as “traffic on roads.” That is, the fundamental processes affecting emis- sions can be decomposed to roadway segments and charac- terized by the nature of traffic occurring on them. Currently, no single model addresses the range of specific emission processes in sufficient detail to capture the effects of traffic-flow improvement projects. CMEM provides the most detailed and best tested estimates of hot-stabilized vehi- cle exhaust emissions at different speeds and accelerations. Similarly, EMFAC2000 provides the most detailed estimates of process-specific evaporative emissions and excess start emissions. The methodology described here relies on emis- sion rate estimates from these two models. (As described pre- viously, no currently available models address either heavy- duty vehicle emissions or PM emissions at the same level of detail as CMEM.) The rates depend on ambient temperature, which fluctuates by time of day and season of the year. A typical afternoon peak-hour temperature for a summer day is selected for the total hydrocarbons (THC) and nitric oxides (NOX) emission rates. A typical afternoon peak hour for an average winter day is selected for CO analyses. The exhaust emissions for THC, NOX, and CO are esti- mated using the following equation: Equation 14 Where: ER = emissions for pollutant R in terms of grams, qR(i, j) = CMEM emission rate for pollutant R in terms of grams per hour for movement at speed i and accel- eration j, and E q i j v i jR R ij = ∗∑ ( , ) ( , ) v(i, j) = VHT at speed i and at acceleration j. CMEM calculates emission rates for feasible values of vehicle speeds and accelerations based on vehicle weight and engine power output. The development of speed-acceleration vehicle activity in the traffic module must be constrained to these feasible values. Otherwise, emissions will be under- estimated, as vehicles will be assumed to travel at higher- than-achievable speeds (and for shorter time periods) than would actually be the case. Because of a lack of the necessary data, the emission estimates do not take into account the following emission effects that might be potentially impacted by traffic-flow improvements: • Starts and stops (e.g., cold starts and hot soaks), • Heavy-duty vehicle emissions, and • PM emissions. 7.2 METHODOLOGY APPLICATION Three emission rate tables (hydrocarbons [HC], CO, and NOX; see Tables 11 through 13, respectively) are used to convert estimates of vehicle activity by speed and acceler- ation into estimates of emissions. One simply looks up the appropriate rate for the speed and acceleration category and multiplies that rate by the VHT in the speed and accel- eration category to obtain the vehicle emissions for that pollutant. 7.3 NONTECHNOLOGY UPDATES TO VEHICLE EMISSION MODULE Emission rate models are frequently being updated. To the extent that new CMEM rates become available, the analyst will need to exercise CMEM to develop new tables of aver- age rates for each acceleration and speed category in Tables 11 through 13.

25 USER’S G UIDE 7.4 TECHNOLOGY UPDATES TO VEHICLE EMISSION MODULE The impacts of new emission control technologies, includ- ing new fuel standards, can be incorporated into the NCHRP 25-21 methodology by developing new tables of modal emis- sion rates to replace Tables 11 through 13. The analyst would need to exercise CMEM with the new technology and fuel standards to develop new tables of running exhaust emission rates for each acceleration and speed category. 7.5 ADDITIONS TO VEHICLE EMISSION MODULE The current NCHRP 25-21 methodology does not treat the impacts of traffic-flow improvements on running evap- orative emissions, refueling emissions, cold starts, and heavy- duty vehicles. If the analyst can create modifications to the basic CMEM rate tables to account for these effects, then the modified tables can be substituted into the NCHRP 25-21 methodology. Speed Acceleration (mph/sec) (mph) -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 0 0.002658 0.002696 0.002734 0.00277 0.002815 0.00287 0.002936 0.002973 0.003021 0.003079 0.003148 0.003227 0.00319 0.003152 0.003114 0.003076 1 0.002677 0.002718 0.002758 0.002797 0.002846 0.002905 0.002976 0.003017 0.00307 0.002965 0.003611 0.004525 0.005438 0.006488 0.007662 0.008949 2 0.002696 0.002739 0.002782 0.002824 0.002876 0.00294 0.003016 0.003061 0.003118 0.002851 0.004073 0.005822 0.007686 0.009823 0.01221 0.014822 3 0.002715 0.002761 0.002807 0.002851 0.002907 0.002975 0.003057 0.003106 0.003168 0.002912 0.004172 0.005968 0.007879 0.010071 0.01252 0.015199 4 0.002734 0.002782 0.002831 0.002877 0.002937 0.00301 0.003097 0.00315 0.003217 0.002974 0.00427 0.006113 0.008071 0.010319 0.012829 0.015577 5 0.002753 0.002804 0.002855 0.002904 0.002968 0.003046 0.003138 0.003195 0.003267 0.003036 0.004372 0.006267 0.008271 0.010584 0.013162 0.015984 6 0.002772 0.002826 0.00288 0.002931 0.002998 0.003081 0.003179 0.00324 0.003318 0.003099 0.004473 0.006421 0.00847 0.010849 0.013495 0.01639 7 0.002791 0.002847 0.002904 0.002958 0.003029 0.003116 0.003221 0.003286 0.003369 0.003199 0.004639 0.006647 0.008797 0.011255 0.013995 0.016985 8 0.002809 0.002869 0.002928 0.002985 0.00306 0.003152 0.003262 0.003332 0.003421 0.003299 0.004805 0.006874 0.009123 0.011661 0.014495 0.017579 9 0.002828 0.00289 0.002952 0.003012 0.003091 0.003188 0.003304 0.003379 0.003473 0.003454 0.005077 0.007247 0.009612 0.012262 0.016072 0.019497 10 0.002847 0.002912 0.002977 0.003039 0.003121 0.003224 0.003346 0.003425 0.003526 0.003609 0.005349 0.00762 0.0101 0.012864 0.017649 0.021414 11 0.002866 0.002933 0.003001 0.003066 0.003152 0.00326 0.003389 0.003473 0.00358 0.003831 0.005681 0.008066 0.010651 0.013562 0.018357 0.022567 12 0.002885 0.002955 0.003025 0.003093 0.003183 0.003296 0.003432 0.003521 0.003634 0.004053 0.006013 0.008512 0.011202 0.01426 0.019065 0.02372 13 0.002885 0.002955 0.003025 0.003094 0.003185 0.0033 0.00344 0.003534 0.003654 0.004294 0.006375 0.009001 0.011812 0.015005 0.020031 0.026457 14 0.002885 0.002955 0.003025 0.003094 0.003186 0.003304 0.003448 0.003548 0.003675 0.004536 0.006738 0.009489 0.012423 0.01575 0.020998 0.029195 15 0.002885 0.002955 0.003025 0.003094 0.003188 0.003307 0.003454 0.003557 0.003687 0.004175 0.006286 0.009011 0.012397 0.016558 0.021999 0.032044 16 0.002885 0.002955 0.003025 0.003094 0.003189 0.003311 0.003459 0.003565 0.003699 0.003815 0.005835 0.008534 0.012372 0.017365 0.023001 0.034894 17 0.002885 0.002955 0.003025 0.003095 0.00319 0.003314 0.003465 0.003575 0.003714 0.003918 0.006039 0.008855 0.012567 0.018326 0.024525 0.037048 18 0.002885 0.002955 0.003025 0.003095 0.003192 0.003318 0.003472 0.003585 0.003728 0.00402 0.006243 0.009176 0.012761 0.019287 0.026049 0.039203 19 0.002885 0.002955 0.003025 0.003096 0.003194 0.003322 0.00348 0.003599 0.003747 0.00415 0.006484 0.009548 0.013303 0.02048 0.028191 0.041716 20 0.002885 0.002955 0.003025 0.003096 0.003195 0.003326 0.003488 0.003613 0.003766 0.00428 0.006725 0.009921 0.013844 0.021672 0.030333 0.04423 21 0.002885 0.002955 0.003025 0.003097 0.003198 0.003331 0.003498 0.003629 0.003787 0.00444 0.00701 0.010353 0.014466 0.023047 0.032765 0.047201 22 0.002885 0.002955 0.003025 0.003097 0.0032 0.003336 0.003508 0.003645 0.003808 0.0046 0.007294 0.010786 0.015089 0.024421 0.035196 0.050173 23 0.002885 0.002955 0.003025 0.003098 0.003202 0.003341 0.00352 0.003663 0.003831 0.004781 0.007615 0.011274 0.015786 0.02489 0.038191 0.053409 24 0.002885 0.002955 0.003025 0.003098 0.003205 0.003347 0.003532 0.00368 0.003853 0.004962 0.007936 0.011762 0.016483 0.025359 0.041187 0.056644 25 0.002885 0.002955 0.003025 0.003099 0.003208 0.003353 0.003545 0.003699 0.003877 0.004673 0.007614 0.011577 0.01727 0.027457 0.044686 0.060157 26 0.002885 0.002955 0.003025 0.0031 0.003211 0.003359 0.003558 0.003718 0.0039 0.004384 0.007292 0.011392 0.018056 0.029554 0.048185 0.063669 27 0.002885 0.002955 0.003025 0.003101 0.003214 0.003366 0.003571 0.003737 0.003924 0.004506 0.007568 0.011896 0.018792 0.031874 0.052215 0.067293 28 0.002885 0.002955 0.003025 0.003101 0.003217 0.003373 0.003585 0.003756 0.003948 0.004628 0.007845 0.012399 0.019527 0.034194 0.056244 0.070917 29 0.002885 0.002955 0.003025 0.003102 0.00322 0.003381 0.003599 0.003776 0.003972 0.004768 0.008165 0.012974 0.020779 0.037172 0.059341 0.073925 30 0.002885 0.002955 0.003025 0.003103 0.003224 0.003389 0.003613 0.003796 0.003996 0.004908 0.008485 0.013549 0.022031 0.040151 0.062437 0.076934 31 0.002885 0.002955 0.003025 0.003104 0.003228 0.003397 0.003628 0.003816 0.004021 0.005069 0.008846 0.014246 0.026044 0.055502 0.083352 0.09644 32 0.002885 0.002955 0.003025 0.003105 0.003231 0.003406 0.003643 0.003836 0.004045 0.005231 0.009208 0.014942 0.030058 0.070853 0.104266 0.115947 33 0.002885 0.002955 0.003025 0.003106 0.003235 0.003415 0.003659 0.003857 0.00407 0.005415 0.0096 0.015646 0.034014 0.075459 0.10678 0.116582 34 0.002885 0.002955 0.003025 0.003107 0.00324 0.003425 0.003675 0.003878 0.004095 0.005598 0.009993 0.016351 0.03797 0.080066 0.109294 0.117217 35 0.002885 0.002955 0.003025 0.003108 0.003244 0.003435 0.003691 0.0039 0.00412 0.0058 0.010414 0.017107 0.042012 0.085436 0.111487 0.117473 36 0.002885 0.002955 0.003025 0.003109 0.003248 0.003446 0.003708 0.003921 0.004146 0.006003 0.010836 0.017863 0.046054 0.090806 0.113679 0.117729 37 0.002885 0.002955 0.003025 0.003111 0.003252 0.003456 0.003724 0.003942 0.00417 0.006221 0.011284 0.019046 0.050309 0.094876 0.115585 0.117879 38 0.002885 0.002955 0.003025 0.003112 0.003256 0.003467 0.003741 0.003963 0.004195 0.00644 0.011732 0.020228 0.054563 0.098945 0.117491 0.11803 39 0.002885 0.002955 0.003025 0.003113 0.003262 0.003479 0.003759 0.003986 0.00422 0.006678 0.012211 0.021496 0.05862 0.102344 0.117605 0.118131 40 0.002885 0.002955 0.003025 0.003115 0.003267 0.003491 0.003777 0.004008 0.004245 0.006916 0.01269 0.022763 0.062677 0.105743 0.117719 0.118233 41 0.002885 0.002955 0.003025 0.003116 0.003272 0.003503 0.003794 0.00403 0.00427 0.007158 0.013193 0.024449 0.066554 0.107877 0.117829 0.118409 42 0.002885 0.002955 0.003025 0.003118 0.003278 0.003516 0.003812 0.004051 0.004294 0.0074 0.013696 0.026134 0.07043 0.11001 0.117938 0.118585 43 0.002885 0.002955 0.003025 0.003119 0.003283 0.003528 0.00383 0.004072 0.004317 0.007675 0.014233 0.028355 0.074045 0.111712 0.118044 0.118629 44 0.002885 0.002955 0.003025 0.003121 0.003289 0.00354 0.003847 0.004094 0.00434 0.007951 0.01477 0.030575 0.07766 0.113413 0.11815 0.118674 45 0.002885 0.002955 0.003025 0.003123 0.003296 0.003553 0.003864 0.004111 0.004358 0.007571 0.014581 0.033722 0.08189 0.115578 0.118223 0.118712 46 0.002885 0.002955 0.003025 0.003125 0.003302 0.003566 0.00388 0.004128 0.004377 0.00719 0.014392 0.036869 0.086119 0.117744 0.118296 0.118751 47 0.002885 0.002955 0.003025 0.003127 0.003309 0.003582 0.003902 0.004156 0.004381 0.007407 0.014936 0.040112 0.091233 0.117894 0.118592 0.118837 48 0.002885 0.002955 0.003025 0.003129 0.003315 0.003598 0.003925 0.004183 0.004386 0.007624 0.01548 0.043356 0.096347 0.118044 0.118887 0.118924 49 0.002885 0.002955 0.003025 0.003131 0.003322 0.003614 0.003946 0.004209 0.00436 0.007874 0.016025 0.046748 0.098776 0.118163 0.118963 0.118996 50 0.002885 0.002955 0.003025 0.003134 0.003329 0.00363 0.003967 0.004235 0.004335 0.008124 0.016569 0.05014 0.101206 0.118282 0.119039 0.119068 51 0.002885 0.002955 0.003025 0.003136 0.003335 0.003646 0.003988 0.004225 0.004315 0.008402 0.017218 0.053512 0.104209 0.118369 0.119097 0.119122 52 0.002885 0.002955 0.003024 0.003138 0.003342 0.003661 0.004008 0.004216 0.004294 0.00868 0.017866 0.056883 0.107212 0.118456 0.119154 0.119177 53 0.002885 0.002955 0.003024 0.003141 0.003349 0.003676 0.004027 0.0042 0.004278 0.008991 0.018885 0.061015 0.109287 0.118535 0.119205 0.119225 54 0.002885 0.002955 0.003024 0.003143 0.003356 0.003692 0.004047 0.004185 0.004262 0.009303 0.019905 0.065148 0.111362 0.118613 0.119256 0.119274 55 0.002885 0.002955 0.003024 0.003146 0.003364 0.003707 0.004067 0.004173 0.00425 0.009647 0.02097 0.068089 0.112713 0.118971 0.119301 0.119318 56 0.002885 0.002955 0.003024 0.003149 0.003372 0.003722 0.004087 0.00416 0.004237 0.009991 0.022035 0.07103 0.114065 0.119329 0.119347 0.119361 57 0.002885 0.002955 0.003024 0.003152 0.00338 0.003738 0.004091 0.00415 0.004227 0.010388 0.023019 0.073712 0.116439 0.119372 0.119388 0.119401 58 0.002885 0.002955 0.003024 0.003155 0.003388 0.003753 0.004096 0.00414 0.004216 0.010785 0.024003 0.076395 0.118812 0.119415 0.11943 0.119441 59 0.002885 0.002955 0.003025 0.003159 0.003397 0.003768 0.004087 0.004132 0.004208 0.011203 0.025289 0.080088 0.119213 0.119789 0.119825 0.119854 60 0.002885 0.002955 0.003025 0.003163 0.003405 0.003783 0.004079 0.004124 0.0042 0.011621 0.026575 0.083781 0.119614 0.120164 0.12022 0.120267 61 0.002885 0.002955 0.003026 0.003167 0.003415 0.003798 0.004072 0.004117 0.004193 0.012076 0.028303 0.08672 0.11971 0.120203 0.120256 0.120301 62 0.002885 0.002955 0.003026 0.003171 0.003424 0.003813 0.004065 0.00411 0.004186 0.012531 0.03003 0.089659 0.119807 0.120242 0.120293 0.120335 63 0.002885 0.002955 0.003027 0.003175 0.003435 0.003828 0.004059 0.004105 0.004393 0.013026 0.032123 0.092225 0.119899 0.120276 0.120325 0.120365 64 0.002885 0.002955 0.003028 0.00318 0.003445 0.003844 0.004054 0.0041 0.0046 0.013521 0.034216 0.094791 0.119991 0.12031 0.120357 0.120396 65 0.002885 0.002955 0.003029 0.003185 0.003459 0.003862 0.004047 0.004094 0.004742 0.014057 0.037802 0.09815 0.120064 0.120344 0.120388 0.120425 TABLE 11 CMEM light-duty vehicle emission rates—HC (grams/hour)

26 US ER ’S G UI DE Speed Acceleration (mph/sec) (mph) -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 0 0.028047 0.028482 0.028916 0.029297 0.029973 0.030957 0.032259 0.033447 0.034957 0.03679 0.038944 0.041419 0.040984 0.04055 0.040116 0.039681 1 0.028264 0.02873 0.029195 0.029604 0.030337 0.031408 0.032827 0.034132 0.03579 0.034819 0.046809 0.069906 0.08806 0.106167 0.125502 0.145875 2 0.028482 0.028978 0.029474 0.029912 0.030701 0.031859 0.033395 0.034817 0.036623 0.032847 0.054673 0.098394 0.135136 0.171784 0.210888 0.252068 3 0.028699 0.029226 0.029753 0.03022 0.031068 0.032316 0.033975 0.03552 0.037481 0.03399 0.056689 0.101211 0.138787 0.176257 0.216237 0.258334 4 0.028916 0.029474 0.030033 0.030529 0.031436 0.032774 0.034554 0.036223 0.038339 0.035134 0.058704 0.104028 0.142439 0.18073 0.221585 0.264599 5 0.029133 0.029722 0.030312 0.030838 0.031806 0.033238 0.035146 0.036945 0.039226 0.036318 0.060684 0.10697 0.143501 0.185463 0.227265 0.271271 6 0.02935 0.029971 0.030591 0.031147 0.032176 0.033702 0.035738 0.037667 0.040113 0.037501 0.062664 0.109912 0.144563 0.190196 0.232945 0.277942 7 0.029567 0.030219 0.03087 0.031457 0.03255 0.034175 0.036344 0.038412 0.041031 0.039142 0.064794 0.110489 0.15009 0.19462 0.24136 0.287548 8 0.029784 0.030467 0.031149 0.031767 0.032924 0.034648 0.036951 0.039157 0.041949 0.040783 0.066923 0.111066 0.155617 0.199045 0.249774 0.297153 9 0.030002 0.030715 0.031428 0.032078 0.033303 0.035129 0.037573 0.039926 0.042903 0.043185 0.070557 0.11468 0.162242 0.206266 0.348428 0.41193 10 0.030219 0.030963 0.031707 0.03239 0.033681 0.035611 0.038196 0.040696 0.043857 0.045588 0.07419 0.118295 0.168867 0.213487 0.447081 0.526707 11 0.030436 0.031211 0.031987 0.032702 0.034064 0.036104 0.038838 0.041495 0.044853 0.048782 0.078812 0.123638 0.173964 0.224166 0.440598 0.564039 12 0.030653 0.031459 0.032266 0.033015 0.034446 0.036597 0.03948 0.042295 0.045849 0.051976 0.083433 0.128981 0.179062 0.234845 0.434115 0.601371 13 0.030653 0.031459 0.032266 0.033019 0.034493 0.03673 0.039742 0.042727 0.046493 0.055636 0.088675 0.135375 0.185885 0.243104 0.449627 0.730662 14 0.030653 0.031459 0.032266 0.033023 0.03454 0.036864 0.040005 0.04316 0.047138 0.059297 0.093917 0.141769 0.192708 0.251363 0.46514 0.859953 15 0.030653 0.031459 0.032266 0.033034 0.034582 0.036962 0.040179 0.043431 0.047524 0.055207 0.090169 0.142488 0.201027 0.25945 0.476636 1.057539 16 0.030653 0.031459 0.032266 0.033045 0.034624 0.03706 0.040353 0.043701 0.047911 0.051117 0.086422 0.143206 0.209345 0.267537 0.488132 1.255126 17 0.030653 0.031459 0.032266 0.033057 0.034672 0.037172 0.040551 0.044008 0.048348 0.052658 0.08933 0.147124 0.213806 0.281838 0.521418 1.396491 18 0.030653 0.031459 0.032266 0.03307 0.034719 0.037283 0.04075 0.044315 0.048785 0.054199 0.092238 0.151042 0.218266 0.296139 0.554704 1.537856 19 0.030653 0.031459 0.032266 0.033085 0.034773 0.037412 0.041009 0.044756 0.049381 0.056202 0.095788 0.155977 0.226882 0.323353 0.617387 1.704427 20 0.030653 0.031459 0.032266 0.0331 0.034827 0.037541 0.041269 0.045197 0.049977 0.058206 0.099339 0.160912 0.235498 0.350567 0.680069 1.870999 21 0.030653 0.031459 0.032266 0.033117 0.034898 0.037698 0.041582 0.045689 0.050621 0.060715 0.103593 0.166882 0.244733 0.381279 0.763732 2.111163 22 0.030653 0.031459 0.032266 0.033134 0.03497 0.037854 0.041895 0.046182 0.051265 0.063224 0.107846 0.172851 0.253969 0.411991 0.847395 2.351327 23 0.030653 0.031459 0.032266 0.033153 0.03505 0.038031 0.042259 0.046722 0.051953 0.066021 0.112615 0.17967 0.26321 0.47856 1.006917 2.601943 24 0.030653 0.031459 0.032266 0.033172 0.035131 0.038208 0.042623 0.047261 0.052641 0.068819 0.117384 0.18649 0.272451 0.545129 1.166439 2.852558 25 0.030653 0.031459 0.032266 0.033194 0.03522 0.038405 0.043023 0.047835 0.053358 0.065894 0.117043 0.193401 0.283011 0.617672 1.388613 3.105985 26 0.030653 0.031459 0.032266 0.033215 0.03531 0.038603 0.043423 0.048408 0.054075 0.062969 0.116703 0.200311 0.29357 0.690215 1.610787 3.359413 27 0.030653 0.031459 0.032266 0.03324 0.03541 0.038822 0.043843 0.048997 0.054803 0.064945 0.121222 0.20854 0.316367 0.772146 1.927233 3.558146 28 0.030653 0.031459 0.032266 0.033264 0.035509 0.039042 0.044263 0.049587 0.055531 0.066921 0.125741 0.216768 0.339165 0.854078 2.243678 3.756879 29 0.030653 0.031459 0.032266 0.033291 0.035618 0.039285 0.044703 0.050192 0.05627 0.069154 0.130954 0.225943 0.377428 1.04598 2.4849 3.824236 30 0.030653 0.031459 0.032266 0.033318 0.035728 0.039528 0.045143 0.050798 0.057008 0.071388 0.136167 0.235117 0.415691 1.237882 2.726122 3.891593 31 0.030653 0.031459 0.032266 0.033348 0.035847 0.039795 0.045605 0.05142 0.057758 0.073924 0.14199 0.245964 0.55502 2.441528 4.409657 5.124772 32 0.030653 0.031459 0.032266 0.033378 0.035966 0.040063 0.046067 0.052043 0.058507 0.076459 0.147813 0.25681 0.69435 3.645173 6.093192 6.357952 33 0.030653 0.031459 0.032266 0.033411 0.036096 0.040356 0.046551 0.052681 0.059265 0.079425 0.154055 0.267552 0.917787 3.920339 6.170853 6.394732 34 0.030653 0.031459 0.032266 0.033444 0.036226 0.040649 0.047034 0.05332 0.060023 0.08239 0.160298 0.278295 1.141225 4.195504 6.248513 6.431511 35 0.030653 0.031459 0.032266 0.033481 0.036366 0.04097 0.047541 0.053976 0.06079 0.085646 0.16688 0.289615 1.491987 4.572735 6.285364 6.445984 36 0.030653 0.031459 0.032266 0.033517 0.036506 0.041291 0.048048 0.054631 0.061557 0.088901 0.173462 0.300935 1.842749 4.949966 6.322216 6.460458 37 0.030653 0.031459 0.032266 0.033557 0.036631 0.041615 0.048552 0.055276 0.062304 0.092351 0.180323 0.350712 2.329718 5.301288 6.382241 6.468102 38 0.030653 0.031459 0.032266 0.033597 0.036756 0.041939 0.049057 0.055921 0.06305 0.0958 0.187183 0.400489 2.816687 5.652609 6.442266 6.475747 39 0.030653 0.031459 0.032266 0.03364 0.036917 0.042314 0.049604 0.056598 0.063816 0.099501 0.194386 0.448877 3.001184 5.854157 6.446879 6.479942 40 0.030653 0.031459 0.032266 0.033684 0.037078 0.042689 0.050151 0.057276 0.064582 0.103202 0.201589 0.497264 3.185683 6.055706 6.451492 6.484138 41 0.030653 0.031459 0.032266 0.033731 0.03725 0.043068 0.050692 0.057936 0.065317 0.106758 0.209033 0.531855 3.277467 6.168096 6.456062 6.5064 42 0.030653 0.031459 0.032266 0.033778 0.037422 0.043447 0.051234 0.058596 0.066052 0.110314 0.216476 0.566446 3.369252 6.280485 6.460631 6.528662 43 0.030653 0.031459 0.032266 0.033829 0.037604 0.043829 0.051767 0.059235 0.066748 0.114437 0.224304 0.639643 3.484976 6.303052 6.465085 6.530388 44 0.030653 0.031459 0.032266 0.033879 0.037787 0.04421 0.052301 0.059873 0.067445 0.118559 0.232131 0.71284 3.6007 6.32562 6.469538 6.532115 45 0.030653 0.031459 0.032266 0.033943 0.037988 0.044601 0.0528 0.060402 0.068005 0.115489 0.240296 0.876628 3.844832 6.384479 6.470284 6.533744 46 0.030653 0.031459 0.032266 0.034006 0.03819 0.044992 0.053298 0.060931 0.068565 0.112419 0.24846 1.040417 4.088964 6.443338 6.47103 6.535373 47 0.030653 0.031459 0.032266 0.034075 0.038392 0.046587 0.056188 0.065067 0.073945 0.122485 0.263922 1.289181 4.703476 6.455601 6.513739 6.545506 48 0.030653 0.031459 0.032266 0.034144 0.038594 0.048182 0.059078 0.069202 0.079326 0.132551 0.279383 1.537945 5.317988 6.467863 6.556447 6.55564 49 0.030653 0.031459 0.032266 0.034218 0.038804 0.049771 0.061928 0.073295 0.084662 0.141913 0.292048 1.799182 5.468068 6.479035 6.567725 6.567054 50 0.030653 0.031459 0.032266 0.034292 0.039013 0.051361 0.064778 0.077388 0.089998 0.151274 0.304713 2.060419 5.618147 6.490208 6.579003 6.578468 51 0.030653 0.031459 0.032256 0.034361 0.039222 0.052934 0.067592 0.081443 0.089294 0.152942 0.315395 2.416013 5.763675 6.494328 6.585236 6.584744 52 0.030653 0.031459 0.032246 0.034431 0.03943 0.054507 0.070406 0.085498 0.088589 0.15461 0.326077 2.771608 5.909202 6.498448 6.591469 6.59102 53 0.030653 0.031459 0.03224 0.034511 0.039651 0.056077 0.073215 0.083953 0.086315 0.158401 0.371649 2.840733 6.106389 6.499936 6.595277 6.594852 54 0.030653 0.031459 0.032234 0.03459 0.039872 0.057648 0.076024 0.082408 0.084041 0.162192 0.417221 2.909858 6.303576 6.501425 6.599084 6.598685 55 0.030653 0.031459 0.032234 0.034681 0.040108 0.059216 0.077493 0.081054 0.082756 0.167163 0.461675 3.125623 6.320415 6.553251 6.601884 6.601498 56 0.030653 0.031459 0.032233 0.034771 0.040343 0.060784 0.078962 0.079699 0.08147 0.172135 0.506128 3.341387 6.337254 6.605077 6.604683 6.604311 57 0.030653 0.031459 0.032237 0.034873 0.040594 0.062349 0.078017 0.078853 0.080713 0.178873 0.536202 3.48152 6.447301 6.607379 6.606992 6.606624 58 0.030653 0.031459 0.032242 0.034975 0.040845 0.063914 0.077072 0.078006 0.079955 0.185611 0.566276 3.621652 6.557348 6.609681 6.609301 6.608937 59 0.030653 0.031459 0.032253 0.03509 0.041113 0.065475 0.076429 0.077463 0.079516 0.192368 0.5896 4.071856 6.706609 6.756441 6.756442 6.756446 60 0.030653 0.031459 0.032264 0.035204 0.04138 0.067037 0.075786 0.076921 0.079077 0.199126 0.612923 4.522059 6.85587 6.903201 6.903583 6.903955 61 0.030653 0.031459 0.032281 0.035331 0.041665 0.068594 0.075337 0.076575 0.078848 0.206551 0.67175 4.718489 6.865068 6.909377 6.909732 6.910086 62 0.030653 0.031459 0.032298 0.035459 0.041951 0.070151 0.074888 0.07623 0.078618 0.213977 0.730576 4.914918 6.874267 6.915552 6.915882 6.916216 63 0.030653 0.031459 0.032323 0.0356 0.042277 0.071724 0.074587 0.076034 0.079409 0.222057 0.810598 4.965501 6.882252 6.920411 6.920721 6.921045 64 0.030653 0.031459 0.032347 0.035741 0.042604 0.073298 0.074285 0.075838 0.0802 0.230138 0.890619 5.016084 6.890237 6.925269 6.925561 6.925874 65 0.030653 0.031459 0.03238 0.035896 0.044123 0.073073 0.073884 0.07565 0.08228 0.238931 1.066209 5.396034 6.896957 6.929721 6.929997 6.930302 TABLE 12 CMEM light-duty vehicle emission rates—CO (grams/hour)

27 USER’S G UIDE Speed Acceleration (mph/sec) (mph) -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 0 0.00071 0.000712 0.000715 0.000716 0.000734 0.000782 0.000868 0.000982 0.001126 0.001299 0.001501 0.001732 0.001729 0.001726 0.001724 0.001721 1 0.000711 0.000714 0.000717 0.000718 0.000738 0.000791 0.000886 0.001011 0.001168 0.001117 0.002465 0.004046 0.005485 0.007039 0.008712 0.010503 2 0.000712 0.000716 0.000719 0.000719 0.000742 0.0008 0.000903 0.00104 0.001211 0.000935 0.00343 0.00636 0.00924 0.012351 0.015699 0.019285 3 0.000714 0.000717 0.000721 0.000721 0.000746 0.000809 0.000922 0.00107 0.001256 0.001004 0.003575 0.006584 0.009528 0.012711 0.016141 0.019816 4 0.000715 0.000719 0.000722 0.000723 0.00075 0.000819 0.00094 0.0011 0.001301 0.001073 0.00372 0.006808 0.009816 0.013072 0.016582 0.020346 5 0.000717 0.00072 0.000724 0.000725 0.000754 0.000829 0.00096 0.001132 0.001348 0.001147 0.003875 0.007046 0.010133 0.013458 0.017057 0.020919 6 0.000718 0.000722 0.000726 0.000727 0.000759 0.00084 0.00098 0.001164 0.001396 0.001222 0.004029 0.007285 0.01045 0.013845 0.017532 0.021493 7 0.000719 0.000724 0.000728 0.000729 0.000763 0.000851 0.001001 0.001199 0.001446 0.001345 0.004261 0.007614 0.010857 0.014367 0.018152 0.022229 8 0.000721 0.000725 0.000729 0.000731 0.000768 0.000862 0.001022 0.001233 0.001497 0.001468 0.004493 0.007942 0.011264 0.014889 0.018771 0.022966 9 0.000722 0.000727 0.000731 0.000733 0.000773 0.000874 0.001045 0.00127 0.001551 0.001753 0.004925 0.0085 0.011939 0.015685 0.020202 0.024612 10 0.000724 0.000728 0.000733 0.000735 0.000778 0.000886 0.001068 0.001306 0.001605 0.002038 0.005357 0.009058 0.012613 0.016481 0.021632 0.026258 11 0.000725 0.00073 0.000735 0.000737 0.000783 0.0009 0.001093 0.001346 0.001663 0.002408 0.005864 0.009707 0.01339 0.017407 0.02251 0.027521 12 0.000726 0.000731 0.000737 0.000739 0.000788 0.000913 0.001117 0.001385 0.001721 0.002778 0.006371 0.010355 0.014167 0.018334 0.023388 0.028784 13 0.000726 0.000731 0.000737 0.000739 0.000792 0.000925 0.001142 0.001426 0.001781 0.003201 0.006941 0.01108 0.015037 0.019366 0.024615 0.03061 14 0.000726 0.000731 0.000737 0.00074 0.000796 0.000937 0.001166 0.001466 0.001841 0.003623 0.007511 0.011805 0.015907 0.020397 0.025841 0.032436 15 0.000726 0.000731 0.000737 0.00074 0.000799 0.000946 0.001183 0.001492 0.001877 0.003016 0.00689 0.0112 0.01647 0.021469 0.027123 0.036146 16 0.000726 0.000731 0.000737 0.000741 0.000803 0.000956 0.001199 0.001517 0.001913 0.002408 0.006269 0.010596 0.017033 0.022542 0.028404 0.039856 17 0.000726 0.000731 0.000737 0.000741 0.000807 0.000967 0.001219 0.001546 0.001954 0.002631 0.006621 0.0111 0.017074 0.023872 0.029932 0.043863 18 0.000726 0.000731 0.000737 0.000742 0.000811 0.000978 0.001238 0.001575 0.001995 0.002853 0.006973 0.011604 0.017114 0.025202 0.03146 0.04787 19 0.000726 0.000731 0.000737 0.000743 0.000815 0.00099 0.001262 0.001616 0.00205 0.003115 0.007374 0.012176 0.017578 0.02657 0.033189 0.050153 20 0.000726 0.000731 0.000737 0.000744 0.00082 0.001003 0.001287 0.001657 0.002106 0.003377 0.007776 0.012748 0.018042 0.027939 0.034917 0.052437 21 0.000726 0.000731 0.000737 0.000745 0.000827 0.001018 0.001317 0.001704 0.002166 0.003674 0.008225 0.013388 0.018837 0.029514 0.037043 0.0548 22 0.000726 0.000731 0.000737 0.000745 0.000834 0.001034 0.001346 0.00175 0.002226 0.00397 0.008675 0.014027 0.019633 0.031089 0.039168 0.057164 23 0.000726 0.000731 0.000737 0.000747 0.000842 0.001051 0.001381 0.0018 0.00229 0.004291 0.009167 0.014732 0.020586 0.032628 0.046087 0.059711 24 0.000726 0.000731 0.000737 0.000748 0.000849 0.001069 0.001415 0.001851 0.002354 0.004612 0.009658 0.015437 0.02154 0.034167 0.053007 0.062259 25 0.000726 0.000731 0.000737 0.000749 0.000858 0.001088 0.001453 0.001905 0.002421 0.00417 0.009259 0.015249 0.022603 0.034959 0.054569 0.065133 26 0.000726 0.000731 0.000737 0.000751 0.000867 0.001107 0.001491 0.001958 0.002488 0.003729 0.00886 0.01506 0.023666 0.03575 0.056131 0.068007 27 0.000726 0.000731 0.000737 0.000752 0.000876 0.001128 0.00153 0.002013 0.002555 0.003986 0.009312 0.015788 0.024719 0.037164 0.060508 0.071253 28 0.000726 0.000731 0.000737 0.000754 0.000886 0.001149 0.00157 0.002068 0.002623 0.004244 0.009763 0.016515 0.025771 0.038578 0.064885 0.0745 29 0.000726 0.000731 0.000737 0.000756 0.000896 0.001172 0.001611 0.002124 0.002691 0.004528 0.010271 0.017333 0.027154 0.04665 0.068097 0.079059 30 0.000726 0.000731 0.000737 0.000757 0.000906 0.001195 0.001652 0.002181 0.00276 0.004812 0.010779 0.018151 0.028537 0.054721 0.07131 0.083618 31 0.000726 0.000731 0.000737 0.000759 0.000918 0.001221 0.001695 0.002238 0.002829 0.005121 0.011334 0.019389 0.032247 0.069823 0.090544 0.09832 32 0.000726 0.000731 0.000737 0.000762 0.000929 0.001246 0.001738 0.002296 0.002899 0.005431 0.011888 0.020628 0.035958 0.084926 0.109778 0.113022 33 0.000726 0.000731 0.000737 0.000764 0.000941 0.001273 0.001783 0.002355 0.002969 0.005795 0.012474 0.021645 0.043996 0.089076 0.111044 0.113725 34 0.000726 0.000731 0.000737 0.000766 0.000954 0.001301 0.001828 0.002414 0.003039 0.006159 0.01306 0.022661 0.052035 0.093227 0.112309 0.114429 35 0.000726 0.000731 0.000737 0.000769 0.000967 0.001331 0.001875 0.002475 0.00311 0.006513 0.013675 0.02375 0.053835 0.099019 0.113086 0.114694 36 0.000726 0.000731 0.000737 0.000772 0.00098 0.001361 0.001922 0.002535 0.00318 0.006867 0.01429 0.024839 0.055635 0.104812 0.113862 0.114959 37 0.000726 0.000731 0.000737 0.000775 0.000992 0.001391 0.001968 0.002595 0.003249 0.007237 0.014933 0.0262 0.059043 0.106868 0.114481 0.115094 38 0.000726 0.000731 0.000737 0.000779 0.001004 0.001421 0.002015 0.002654 0.003318 0.007608 0.015575 0.02756 0.062452 0.108925 0.1151 0.115228 39 0.000726 0.000731 0.000737 0.000782 0.00102 0.001456 0.002066 0.002717 0.003389 0.008 0.016252 0.028914 0.068435 0.110739 0.115208 0.115334 40 0.000726 0.000731 0.000737 0.000786 0.001035 0.001491 0.002116 0.00278 0.00346 0.008393 0.016928 0.030267 0.074417 0.112553 0.115316 0.11544 41 0.000726 0.000731 0.000737 0.00079 0.001051 0.001526 0.002167 0.002841 0.003528 0.00874 0.017624 0.032984 0.079901 0.112976 0.11542 0.115564 42 0.000726 0.000731 0.000737 0.000794 0.001068 0.001562 0.002217 0.002902 0.003596 0.009087 0.01832 0.035701 0.085384 0.113399 0.115524 0.115688 43 0.000726 0.000731 0.000737 0.000799 0.001085 0.001597 0.002267 0.002962 0.003661 0.009512 0.019061 0.0371 0.090015 0.113918 0.115623 0.115771 44 0.000726 0.000731 0.000737 0.000804 0.001103 0.001633 0.002316 0.003021 0.003726 0.009937 0.019802 0.038499 0.094646 0.114437 0.115723 0.115853 45 0.000726 0.000731 0.000737 0.00081 0.001123 0.00167 0.002363 0.003071 0.003778 0.009502 0.019581 0.045323 0.099275 0.115095 0.115804 0.115925 46 0.000726 0.000731 0.000737 0.000817 0.001143 0.001708 0.002411 0.003121 0.003831 0.009067 0.01936 0.052146 0.103905 0.115754 0.115885 0.115996 47 0.000726 0.000731 0.000737 0.000824 0.001162 0.001783 0.002533 0.003285 0.004038 0.009658 0.020316 0.055965 0.106478 0.116131 0.11631 0.116366 48 0.000726 0.000731 0.000737 0.000831 0.001181 0.001858 0.002655 0.00345 0.004245 0.010249 0.021272 0.059784 0.109052 0.116508 0.116734 0.116736 49 0.000726 0.000731 0.000737 0.000839 0.001201 0.001933 0.002773 0.003611 0.004261 0.010497 0.022051 0.062572 0.11115 0.117239 0.117453 0.117454 50 0.000726 0.000731 0.000737 0.000847 0.001221 0.002007 0.002892 0.003772 0.004276 0.010744 0.022829 0.065361 0.113248 0.11797 0.118173 0.118173 51 0.000726 0.000731 0.000736 0.000855 0.001242 0.002081 0.003008 0.003733 0.004076 0.011035 0.023839 0.068341 0.115 0.118388 0.118583 0.118582 52 0.000726 0.000731 0.000736 0.000863 0.001262 0.002156 0.003124 0.003695 0.003876 0.011326 0.024849 0.071321 0.116751 0.118806 0.118993 0.118992 53 0.000726 0.000731 0.000736 0.000871 0.001284 0.002229 0.003239 0.003576 0.00375 0.011728 0.026264 0.07117 0.117085 0.119093 0.119273 0.119272 54 0.000726 0.000731 0.000736 0.00088 0.001305 0.002302 0.003355 0.003457 0.003624 0.01213 0.02768 0.071018 0.11742 0.11938 0.119553 0.119551 55 0.000726 0.000731 0.000736 0.000889 0.001328 0.002375 0.003286 0.003375 0.003539 0.012611 0.028856 0.074127 0.117975 0.119679 0.119764 0.119763 56 0.000726 0.000731 0.000736 0.000899 0.001351 0.002448 0.003217 0.003294 0.003455 0.013093 0.030033 0.077235 0.11853 0.119978 0.119976 0.119975 57 0.000726 0.000731 0.000736 0.000909 0.001374 0.002521 0.003158 0.003235 0.003395 0.01376 0.031211 0.08091 0.119344 0.12015 0.120149 0.120147 58 0.000726 0.000731 0.000736 0.000919 0.001398 0.002594 0.003099 0.003177 0.003336 0.014427 0.032389 0.084586 0.120157 0.120323 0.120321 0.12032 59 0.000726 0.000731 0.000737 0.00093 0.001423 0.002666 0.003055 0.003134 0.003294 0.015032 0.034573 0.096071 0.12467 0.124824 0.124825 0.124826 60 0.000726 0.000731 0.000738 0.000941 0.001449 0.002738 0.003011 0.003092 0.003252 0.015637 0.036756 0.107557 0.129183 0.129325 0.129328 0.129333 61 0.000726 0.000731 0.000739 0.000953 0.001475 0.002809 0.002977 0.00306 0.003222 0.016295 0.037851 0.110544 0.129464 0.12959 0.129594 0.129598 62 0.000726 0.000731 0.00074 0.000966 0.001502 0.002881 0.002944 0.003028 0.003192 0.016953 0.038946 0.11353 0.129745 0.129856 0.129859 0.129864 63 0.000726 0.000731 0.000741 0.000979 0.001532 0.002931 0.002919 0.003005 0.003491 0.017665 0.042293 0.115898 0.129976 0.130071 0.130074 0.130079 64 0.000726 0.000731 0.000743 0.000992 0.001562 0.002981 0.002894 0.002982 0.00379 0.018377 0.045641 0.118266 0.130208 0.130286 0.130289 0.130295 65 0.000726 0.000731 0.000745 0.001007 0.00163 0.002923 0.002858 0.002955 0.003955 0.019146 0.050374 0.121041 0.130421 0.130488 0.130491 0.130497 TABLE 13 CMEM light-duty vehicle emission rates—NOX (grams/hour)

28 US ER ’S G UI DE CHAPTER 8 BASE CASE The PSRC travel model data set was selected for the appli- cation of the NCHRP 25-21 methodology to case studies. The PSRC travel demand model covers four counties of the Seattle/Tacoma metropolitan area with a population of about 3 million people. (See the University of Washington and Cambridge Systematics’s “Land Use and Travel Demand Forecasting Models, Model Documentation,” prepared for the Puget Sound Regional Council, final report, June 30, 2001, www.psrc.org/datapubs/pubs/model_modelrequirements.pdf.) The model represents the PSRC region using 852 internal zones, about 19,000 directional road links, and 317 transit lines. The model splits travel demand between three time peri- ods (3-hour AM peak, 3-hour PM peak, and rest of day) and three modes of travel (drive alone, carpool, and transit). An economic forecasting model and a pair of land-use allocation models (DRAM and EMPAL) are used by PSRC to generate the socioeconomic data required by the travel demand model. 8.1 INPUT The PSRC model for the year 2020 was selected as the base case for demonstrating the application of the NCHRP 25-21 methodology. All of the other case studies using the NCHRP 25-21 were run in comparison to this base case for the year 2020. Three key inputs are required from the PSRC model for application in the NCHRP 25-21 methodology: the highway network (Table 14), the transit network (Table 15), and the base case OD travel demand (Table 16). The highway network contains the following data items for each directional highway link, where ul1, ul2, and ul3 are user-definable fields: • Length (in miles), • Modes (SOV, HOV, bus, rail, ferry, transit walk access, transit auto access), • Number of lanes, • Volume/delay function, • Capacity per lane (vph) (ul1), • Free-flow travel time (minutes) (ul2), and • Facility type (0 = bus/walk link, 1 = freeway, 2 = express- way, 3 = urban arterial, 4 = urban one way, 5 = centroid connector, 6 = rural arterial) (ul3). Freeway HOV lanes are coded as parallel links to the free- way with HOV/bus-only cross connectors. For each transit line, the following data are available: • Mode, • Vehicle type, • Headway (minutes), • Speed (mph), • Length (miles), and • Number of segments. The projected year 2020 population is 4.3 million people, and the projected 2020 employment is 2.3 million jobs. The PSRC model estimated travel demand for 2020 is 12.4 mil- lion daily person trips in nine OD tables by mode and time period (summarized in Table 16). 8.2 APPLICATION OF THE HCM ASSIGNMENT MODULE TO THE PSRC DATA SET The basic PSRC highway must be modified before the HCM Assignment Module can be applied to it. 8.2.1 Step 1: Code Free-Flow Speeds and Capacities Step 1 consists of substituting HCM-based capacities and free-flow speeds for the planning values in the model. In the case of the PSRC model, the capacities and free-flow speeds are customized for individual links. Each facility type in the PSRC model is applied to a wide range of conditions. For example, ramps are sometimes coded as freeway facility types, arterial street types, or one-way arterial street types. The free- flow speeds for freeway-type links consequently range from 20 mph to 70 mph. Similar ranges occur for the other facil- ity types. It is therefore not possible to make a blanket sub- stitution of capacities and free-flow speeds based upon facil- ity type and area type. The substitutions would have to be made on a link-by-link basis. Because this basis is not practi- cal for a demonstration of the methodology, the link-specific capacities and free-flow speeds will be left unchanged. The one change made to the current PSRC method was to replace the current link free-flow travel times (ul2) in the AM

29 USER’S G UIDE and PM scenarios (which, in the PSRC model, are computed from congested speed output by the daily assignment) with the free-flow speeds from the daily assignment. 8.2.2 Step 2: Replace BPR Equations with HCM Equation The BPR speed-flow equations used in the PSRC model are replaced with the HCM 2000 speed-flow equation. The existing PSRC volume delay functions (VDFs) for the daily and off-peak scenarios were not touched. The VDFs involve 24-hour and 18-hour demand assignments and are only moderately capacity constrained (12-hour capacities for the daily assignment and 8-hour capacities for the off-peak assignment). The off-peak assignment currently uses the con- gested travel times from the daily assignment for its free-flow times. This use was unchanged. The AM and PM peak-hour assignments currently use the following VDFs, where fd10, fd30, fd40, fd47, fd49, and fd59 are functions and volau is the auto volume: • fd10 = ul2 ∗ (1 + .15 ∗ (.08 ∗ volau/(lanes ∗ ul1)) ∧ 4) • fd30 = ul2 + (((.34 ∗ (volau/ul1) / lanes) − 1) .max. 0) ∗ (60/lanes) • fd40 = ul2 • fd47 = ul2 ∗ (1 + .15 ∗ (.125 ∗ volau/(lanes ∗ ul1)) ∧ 4) • fd49 = ul2 ∗ (1 + .15 ∗ (.375 ∗ volau/(lanes ∗ ul1)) ∧ 4) • fd59 = ul2 ∗ (1 + .15 ∗ (.455 − .125) ∗ volau/(3 ∗ lanes ∗ ul1)) Fd10 is used primarily in the daily assignment for all roads. Although 179 links appear to use fd10 in the AM peak assignment, the rationale for this use is unclear, so fd10 was replaced with fd59 for these 179 links. Fd10 is not used in the PM assignment. Fd30 is used for 14 auto-ferry links in both the AM and PM assignments. These VDFs were retained unchanged. Fd40 is used in both the AM and PM assignments for 404 nonauto ferry and walk links for the 1990 network. This function is also used for 1,465 links in the AM assignment and 873 links in the PM assignment for the 2020 network. In essence, the travel time for the link is fixed at whatever value was originally coded by the PSRC modeler. This VDF was not changed. Fd47 is used for 10 freeway HOV lane links in the AM and PM peak assignments for the 2020 network (not present in the 1990 network) and was not changed. Fd49 is used for 16 short connector links between the free- way HOV lane links and the mixed-flow lane links of the freeway for the AM and PM peak 2020 network assignments (not used in 1990 network). This VDF is also used for some rural arterial links and really short urban arterial links. This VDF was not changed. Centerline-Miles Lane-Miles No. of Links Capacity-Miles (VMT) Mean Free-Flow Speed (mph) 11,388 17,390 17,711 20,194,252 19.9 Network Transit Vehicles Lines Route-Miles 2020 1,286 542 9,716 Peak Mode Person Trips % Mode AM SOV 1,720,034 79.9% HOV 273,841 12.7% Transit 160,154 7.4% PM SOV 2,766,570 88.9% HOV 345,056 11.1% Transit ?* ?* Off Peak SOV 6,732,642 96.3% HOV 258,595 3.7% Transit ?* ?* Daily SOV 11,219,246 90.6% HOV 877,492 7.1% Transit 287,932 2.3% Total 12,384,670 *The PSRC model does not split transit trips into PM and off peak, but these trips are included in the estimated daily transit trips. TABLE 14 Base case 2020 highway network TABLE 15 Base case 2020 transit network TABLE 16 Base case 2020 person trips

Fd59 is used for the vast majority of the road links in the AM and PM peak assignments. This VDF will be replaced with the HCM speed-flow function. 8.2.3 Step 3: Generate Additional Network Parameters Required by HCM Equation The HCM equation requires several additional parameters not coded in the PSRC network: • The number of signals on a link (N), • The zero-flow signal delay (D0), • The segment delay between signal (DL), and • The calibration parameter (J). 8.2.3.1 Number of Signals The number of signals on a link (N) is computed and stored for each link as follows: • For freeways (ul3 = 1), centroid connectors (ul3 = 5), and rural arterials (ul3 = 6), the number of signals is zero, but because N must be at least 1, N = 1 for these links. • For all other facility types, N is computed as follows: Equation 15 Where: N = the number of signals on the segment, Sd = the signal density for the link (signals/mile), L = the length of the link (miles), max = maximum function (outputs the maximum of two values), and INT = integer divide function (outputs result truncated to integer value). Note that the first signal at the start of a link is excluded from N, so if a link is 1 mile long and signals are spaced 1 mile apart, there will be two signals on the link (one at the start and one at the end), but because the first signal is excluded N INT L Sd= ∗( )[ ]max ,1 30 (to avoid double counting the signal at the end of one link and the beginning of the next link), N is equal to 1. An integer divide is used to obtain the number of signals, since modelers usually terminate a link at a major intersection, which is likely to be signalized. So a 1.5-mile-long link with signals assumed to be spaced an average of 1 mile apart would have one signal at the start, one signal at the end, and no sig- nals in between. Thus, the default signal density assumption is used as a rough guide for determining whether multiple sig- nals might exist within the stretch of a model link; however, if the link length is close to a multiple of the signal density, the coded-link length is assumed to be more accurate than the assumed default signal density. Table 17 shows the signal density (Sd). The table was cre- ated using local knowledge of typical signal densities on expressways and arterials. 8.2.3.2 Zero-Flow Signal Delay The zero-flow delay in hours (D0) is computed and stored for each link. The zero-flow control delay is zero for freeway, centroid, and rural facility types (ul3 = 1, 5, 6). For ul3 = 2, 3, 4, it is computed using the equation in the methodology: Equation 16 Where: D0 = the zero-flow control delay at the signal (hours); N = maximum of 1, or the number of signals on the segment; 3,600 = conversion from seconds to hours; g/C = average effective green time per cycle for signals on segment; C = average cycle length for all signals on the segment (seconds); and DF = delay factor, = 0.9 for uncoordinated traffic-actuated signals, = 1.0 for uncoordinated fixed-time signals, D N C gC0 2 3 600 2 1= ∗ ∗ −( ), DF US ER ’S G UI DE Signals/Mile Free Speed Arterial Class Expressway Ul3 = 2 Urban Arterial Ul3 = 3,4 55+ I 1 2 50 I 1 2 45 I 1 2 40 II 1 2 35 III 3 5 30 IV 6 8 25- IV 8 8 TABLE 17 Facility type, free speed, arterial class, and signal density

= 1.2 for coordinated signals with unfavorable progression, = 0.9 for coordinated signals with favorable pro- gression, and = 0.6 for coordinated signals with highly favorable progression. A default value of 0.44 is used for the g/C ratio. A default signal cycle length of 120 seconds is used. 8.2.3.3 Between-Signal-Segment Delay The segment delay (DL) is computed and stored as follows: Equation 17 Where: L = the length of the segment. D L dL L= ∗ 60 31 The delay per mile (dL) is given in Table 18, which was derived from the assumed signal density and Exhibit 15-3 of the HCM 2000. 8.2.3.4 The Calibration Parameter (J) The calibration parameter J is stored for each link. Table 19 was created from the table provided in the methodology using the facility types and free-flow speeds coded in the PSRC model network. Centroid connectors were given a flat speed-flow equation taken from freeways (for 75+ mph free- flow speed). Table 20 shows the final combined set of parameters for the new HCM speed-flow equations for the PSRC model. The selection criteria are used to select the default values used to compute the additional parameters for the HCM equations for each link. The standard BPR parameters (also used by the HCM equations) are already coded in the PSRC model for each link. USER’S G UIDE Free-Flow Speed Expressway ul3 = 2 Urban Arterial ul3 = 3, 4 55+ 0 secs 8 secs 50 0 8 45 0 8 40 0 8 35 0 20 30 25 45 25- 60 60 Free-Flow Speed Freeway Ul3 = 1 Expressway Ul3 = 2 Urban Arterial Ul3 = 3 One-Way Arterial Ul3 = 4 Rural Arterial Ul3 = 6 Lanes > 1 Rural Arterial Ul3 = 6 Lanes = 1 75+ 29.47E-06 22.1E-06 204E-06 204E-06 2.296E-06 90.43E-06 70 20.03E-06 22.1E-06 204E-06 204E-06 2.296E-06 90.43E-06 65 14.23E-06 22.1E-06 204E-06 204E-06 2.296E-06 90.43E-06 60 8.426E-06 22.1E-06 204E-06 204E-06 2.296E-06 138.5E-06 55 3.306E-06 22.1E-06 204E-06 204E-06 1.821E-06 223.9E-06 50 3.306E-06 22.1E-06 204E-06 204E-06 1.108E-06 389.3E-06 45 3.306E-06 22.1E-06 204E-06 204E-06 2.174E-06 748.4E-06 40 3.306E-06 49.9E-06 200E-06 200E-06 2.174E-06 748.4E-06 35 3.306E-06 802E-06 1780E-06 1780E-06 2.174E-06 748.4E-06 30 3.306E-06 3170E-06 4990E-06 4990E-06 2.174E-06 748.4E-06 ≥ 25 3.306E-06 3170E-06 4990E-06 4990E-06 2.174E-06 748.4E-06 TABLE 18 Segment delay by facility type and free-flow speed TABLE 19 J Parameters by facility type and free-flow speed

32 US ER ’S G UI DE TABLE 20 Final parameters for HCM equation VDF 59 Selection Criteria Standard BPR Parameters Additional Parameters for HCM Equations Facility ul3 @fresp Lanes L Ro X Do DL N T J 1 >70 all len @ul21 volau/(ul1*lanes) 0 0 1 1 2.947E-05 1 65-70 all len @ul21 volau/(ul1*lanes) 0 0 1 1 2.003E-05 1 60-65 all len @ul21 volau/(ul1*lanes) 0 0 1 1 1.423E-05 1 55-60 all len @ul21 volau/(ul1*lanes) 0 0 1 1 8.426E-06 Freeway 1 <=55 all len @ul21 volau/(ul1*lanes) 0 0 1 1 3.306E-06 2 >45 all len @ul21 volau/(ul1*lanes) N*16.93/60 0 max(1,INT(len*1)) 1 2.21E-05 2 40-45 all len @ul21 volau/(ul1*lanes) N*16.93/60 0 max(1,INT(len*1)) 1 4.99E-05 2 35-40 all len @ul21 volau/(ul1*lanes) N*16.93/60 0 max(1,INT(len*1)) 1 8.02E-04 Expressway 2 <35 all len @ul21 volau/(ul1*lanes) N*16.93/60 L*25/60 max(1,INT(len*3)) 1 3.17E-03 3 >45 all len @ul21 volau/(ul1*lanes) N*16.93/60 L*8/60 max(1,INT(len*2)) 1 2.04E-04 3 40-45 all len @ul21 volau/(ul1*lanes) N*16.93/60 L*8/60 max(1,INT(len*2)) 1 2.00E-04 3 35-40 all len @ul21 volau/(ul1*lanes) N*16.93/60 L*20/60 max(1,INT(len*2)) 1 1.78E-03 Urban Arterial 3 <35 all len @ul21 volau/(ul1*lanes) N*16.93/60 L*45/60 max(1,INT(len*5)) 1 4.99E-03 4 >45 all len @ul21 volau/(ul1*lanes) N*6.00/60 L*8/60 max(1,INT(len*2)) 1 2.04E-04 4 40-45 all len @ul21 volau/(ul1*lanes) N*6.00/60 L*8/60 max(1,INT(len*2)) 1 2.00E-04 4 35-40 all len @ul21 volau/(ul1*lanes) N*6.00/60 L*20/60 max(1,INT(len*2)) 1 1.78E-03 One-Way Arterial 4 <35 all len @ul21 volau/(ul1*lanes) N*6.00/60 L*45/60 max(1,INT(len*5)) 1 4.99E-03 Centroid Connector 5 all all len @ul21 volau/(ul1*lanes) 0 0 1 1 2.947E-05 6 >60 >1 len @ul21 volau/(ul1*lanes) 0 0 1 1 2.296E-06 6 55-60 >1 len @ul21 volau/(ul1*lanes) 0 0 1 1 1.821E-06 6 50-55 >1 len @ul21 volau/(ul1*lanes) 0 0 1 1 1.108E-06 6 <50 >1 len @ul21 volau/(ul1*lanes) 0 0 1 1 2.174E-06 6 >65 1 len @ul21 volau/(ul1*lanes) 0 0 1 1 9.043E-05 6 60-65 1 len @ul21 volau/(ul1*lanes) 0 0 1 1 0.0001385 6 55-60 1 len @ul21 volau/(ul1*lanes) 0 0 1 1 0.0002239 6 50-55 1 len @ul21 volau/(ul1*lanes) 0 0 1 1 0.0003893 Rural Arterial 6 <50 1 len @ul21 volau/(ul1*lanes) 0 0 1 1 0.0007484 ul3, ul21 = user-definable fields. @fresp = at free-flow speed (mph). L = segment length. R0 = segment traversal time at free-flow speed. X = volume/capacity ratio. D0 = zero-flow control delay at the signal. DL = delay per mile. N = maximum of 1, or the number of signals on the segment. T = length of analysis period, in hours. J = calibration parameter. volau = auto volume. INT = integer divide function.

33 USER’S G UIDE CHAPTER 9 CASE STUDY 1: ADD FREEWAY LANE—RURAL Case Study 1 adds a single through lane to each direction of a freeway in a rural mountainous area. The freeway is uncongested under the base 2020 condition; thus, adding the lane has no effect on the operating speeds of vehicles on the freeway. The specific location for this case study is a 7.6-mile-long section of the Interstate 90 freeway between the S.E. 68th Street Interchange and the SR 202 Interchange near North Bend, Washington (see Figure 6). 9.1 APPLICATION The 2020 base case has four lanes in the uphill direction and three in the reverse direction. The project adds one through lane in each direction over the entire length of the project. Thus, after the improvement, there are five lanes in the uphill direction, including truck-climbing lanes, and four lanes in the reverse direction. There was no change in the 70-mph free- flow speed and the 1,800-vehicles/hour/lane capacity for this freeway. 9.2 CASE STUDY RESULTS The NCHRP 25-21 methodology was used to compute the impacts of the traffic-flow improvement on the 2020 base case trip tables by time period (AM peak, PM peak, and off peak) and by mode (SOV, HOV, and transit). The revised trip tables were then reassigned to the improved network to deter- mine the impact on VMT, VHT, and emissions. The results are summarized in Tables 21 through 23. This case study is an example of an unnecessary capacity improvement being placed at a location where there is already plenty of excess capacity. The extra capacity on the freeway has no effect on the speed of traffic on the freeway, which is all traveling at the free-flow speed during both the peak and off-peak periods. The result is that the capacity improvement has no effect on travel times, trip making, VMT, or emissions, as expected.

34 US ER ’S G UI DE Project Figure 6. Case Study 1: I-90 North Bend. Scenario EB WB Period V/C Speed (mph) Time (min) V/C Speed (mph) Time (min) Before 0.10 69.3 6.54 0.18 69.7 6.51 After 0.07 69.3 6.54 0.14 69.7 6.51 Difference -0.02 0.0 0.00 -0.04 0.0 0.00 AM Peak % Difference -24.17% 0.00% 0.00% -21.94% 0.00% 0.00% Before 0.21 69.3 6.54 0.11 69.7 6.51 After 0.16 69.3 6.54 0.09 69.7 6.51 Difference -0.05 0.0 0.00 -0.02 0.0 0.00 PM Peak % Difference -23.77% 0.00% 0.00% -21.44% 0.00% 0.00% Before 0.08 69.3 6.54 0.06 69.7 6.51 After 0.12 69.3 6.54 0.10 69.7 6.51 Difference 0.04 0.0 0.00 0.04 0.0 0.00 Off Peak % Difference 51.33% 0.00% 0.00% 56.23% 0.00% 0.00% Period Direction Before After Difference % Difference EB 1,643 1,643 0 0.00% WB 3,545 3,545 0 0.00% AM Peak TOT 5,188 5,188 0 0.00% EB 3,623 3,623 0 0.00% WB 2,299 2,299 0 0.00% PM Peak TOT 5,922 5,922 0 0.00% EB 15,830 15,830 0 0.00% WB 14,904 14,904 0 0.00% Off Peak TOT 30,735 30,735 0 0.00% EB 21,096 21,096 0 0.00% WB 20,749 20,749 0 0.00% Total TOT 41,844 41,844 0 0.00% TABLE 21 Case Study 1: Travel time changes on the facility TABLE 22 Case Study 1: Volume changes on the facility

35 USER’S G UIDE Scenario Period VMT VHT Speed THC CO NOX (mi) (hrs) (mph) (gm) (gm) (gm) AM Peak 12,152,900 381,540 31.9 PM Peak 15,261,700 518,222 29.5 Off Peak 37,208,800 1,179,200 31.6 Before Total 64,623,400 2,078,962 31.1 45,000,186 714,064,881 46,356,879 AM Peak 12,152,900 381,540 31.9 PM Peak 15,261,700 518,222 29.5 Off Peak 37,208,800 1,179,200 31.6 After Total 64,623,400 2,078,962 31.1 45,000,186 714,064,881 46,356,879 Difference 0 0 0.0 0 0 0 % Difference 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% TABLE 23 Case Study 1: Regional results

36 US ER ’S G UI DE CHAPTER 10 CASE STUDY 2: CLOSE FREEWAY LANE—URBAN Case Study 2 removes a single through lane in each direc- tion of a freeway in an urban area. The freeway is uncon- gested under the base 2020 condition, but deleting the lane makes the freeway congested. The specific location for this case study is a 6.6-mile-long section of the State Route 520 freeway (the Evergreen Point Bridge) between the I-5 and I-405 interchanges in Seattle, Washington (see Figure 7). 10.1 APPLICATION Before the improvement, this section had three lanes in each direction. Note that this section also has a barrier-separated HOV lane in each direction. Because the links were not very congested with three lanes (peak-period volume/capacity ratio was less than 1.00), a lane was removed in each direc- tion so as to provide a case study where the impacts of an improvement were to make conditions more congested. Thus, after the “improvement,” the section has two lanes in each direction with a barrier-separated HOV lane. No change was made in the 50- to 60-mph free-flow speed and the 1,800- to 1,850-vehicles/hour/lane capacity for this freeway. 10.2 CASE STUDY RESULTS The NCHRP 25-21 methodology was used to compute the impacts of the traffic-flow improvement on the 2020 base case trip tables by time period (AM peak, PM peak, and off peak) and by mode (SOV, HOV, and transit). The revised trip tables were then reassigned to the improved network to deter- mine the impact on VMT, VHT, and emissions. The results are summarized in Tables 24 through 26. This case study illustrates how removing or closing a free- way lane causes an initially uncongested freeway to become congested during the peak periods. This action reduces peak- period travel on the facility significantly, but has a very small effect on regional travel (VMT). The mean speed on the facility is reduced significantly, and the mean travel time is increased significantly (on the order of 30 to 50 percent). Reverse commute directions are less affected. Peak-period volumes are reduced by about 15 per- cent, while off-peak travel is relatively unaffected. Total daily VMT is changed by less than 0.01 percent. Total emissions of THC, CO, and NOX are reduced by 0.2 to 0.4 percent. The net effect of closing the freeway lane is to reduce daily vehicle emissions by 0.2 to 0.4 percent.

37 USER’S G UIDE Project Figure 7. Case Study 2: Delete lane from SR 520 Evergreen Point Bridge. Period Scenario EB WB V/C Speed (mph) Time (min) V/C Speed (mph) Time (min) Before 0.45 48.5 8.55 0.78 47.9 8.62 After 0.61 44.7 9.28 0.94 43.6 11.19 Difference 0.16 -3.7 0.73 0.17 -4.3 2.57 AM Peak % Difference 36.05% -7.72% 8.54% 21.48% -8.98% 29.81% Before 0.82 48.5 8.55 0.66 47.9 8.62 After 0.92 45.1 13.22 0.93 47.3 8.72 Difference 0.10 -3.4 4.67 0.27 -0.6 0.10 PM Peak % Difference 12.22% -6.93% 54.62% 41.69% -1.25% 1.16% Before 0.32 48.5 8.55 0.32 47.9 8.62 Off Peak After 0.47 48.5 8.55 0.48 47.9 8.62 Difference 0.16 0.0 0.00 0.16 0.0 0.00 % Difference 49.73% 0.00% 0.00% 49.71% 0.00% 0.00% Period Direction Before After Difference %Difference EB 7,445 6,755 -691 -9.28% WB 12,808 10,372 -2,435 -19.02% AM Peak TOT 20,253 17,127 -3,126 -15.43% EB 13,495 10,097 -3,398 -25.18% WB 10,835 10,236 -600 -5.53% PM Peak TOT 24,330 20,333 -3,998 -16.43% EB 31,369 31,314 -56 -0.18% WB 31,526 31,465 -62 -0.20% Off Peak TOT 62,896 62,778 -117 -0.19% EB 52,310 48,165 -4,144 -7.92% WB 55,170 52,073 -3,097 -5.61% Total TOT 107,479 100,238 -7,241 -6.74% TABLE 24 Case Study 2: Travel time changes on the facility TABLE 25 Case Study 2: Volume changes on the facility

38 US ER ’S G UI DE Scenario Period AM Peak 12,152,900 381,540 31.9 PM Peak 15,261,700 518,222 29.5 Off Peak 37,208,800 1,179,200 31.6 Before Total 64,623,400 2,078,962 31.1 45,000,186 714,064,881 46,356,879 AM Peak 12,156,500 381,945 31.8 PM Peak 15,259,800 519,266 29.4 Off Peak 37,207,400 1,179,100 31.6 After Total 64,623,700 2,080,311 31.1 44,914,702 712,081,618 46,184,899 Difference 300 1,349 0.0 -85,484 -1,983,263 -171,980 % Difference 0.00% 0.06% -0.06% -0.19% -0.28% -0.37% VMT VHT Speed THC CO NOX (mi) (hrs) (mph) (gm) (gm) (gm) TABLE 26 Case Study 2: Regional results

39 USER’S G UIDE CHAPTER 11 CASE STUDY 3A: REMOVE FREEWAY HOV LANE Case Study 3a removes the freeway HOV lanes (one HOV lane in each direction) from a freeway in an urban area. The freeway is uncongested before and after the removal of the lanes. The specific location for this case study is a 2.1-mile- long section of the I-405 freeway between the S.E. 181st and SR 169 interchanges in Renton, Washington (see Fig- ure 8). 11.1 APPLICATION The I-405 freeway mainline has • Four lanes in each direction with a 45-mph free-flow speed and a 1,800-vph/lane capacity and • An HOV lane in each direction with a 60-mph free-flow speed and a 1,500-vph/lane capacity. In this project, the HOV lanes in both directions are removed. Thus, the section now has four lanes in each direc- tion and no HOV lane. 11.2 CASE STUDY RESULTS The NCHRP 25-21 methodology was used to compute the impacts of the traffic-flow improvement on the 2020 base case trip tables by time period (AM peak, PM peak, and off peak) and by mode (SOV, HOV, and transit). The revised trip tables were then reassigned to the improved network to deter- mine the impact on VMT, VHT, and emissions. The results are summarized in Tables 27 through 29. This case study illustrates the impacts of closing the HOV lanes on an existing, uncongested freeway. The action has no significant effect on speeds and travel times for the mixed- flow lanes, but because it closes an HOV facility, there is a significant effect on facility volumes. Peak-period volumes on the freeway increase by 12 to 20 percent. Daily volumes are increased by slightly less than 10 percent. Total regional VMT is increased by 0.02 percent, but mobile source emissions are decreased by 0.02 to 0.04 percent. The slight speed increase appears to have overcome the slight VMT increase. This may be the result of a model coding prac- tice that gives the HOV lane lower free-flow speeds and capac- ities than mixed-flow lanes on a freeway. Fewer HOVs in the HOV lanes may have slightly boosted the regional mean speed of all traffic.

40 US ER ’S G UI DE Project Figure 8. Case Study 3a: I-405 Renton. TABLE 27 Case Study 3a: Travel time changes on the facility Period Scenario EB WB V/C Speed (mph) Time (min) V/C Speed (mph) Time (min) Before 0.46 43.0 4.18 0.64 43.2 4.14 After 0.58 43.0 4.18 0.74 43.2 4.14 Difference 0.12 0.0 0.00 0.10 0.0 0.00 AM Peak % Difference 25.58% 0.00% 0.00% 16.43% 0.00% 0.00% Before 0.81 43.0 4.18 0.75 43.2 4.14 After 0.89 43.0 4.18 0.86 43.2 4.14 Difference 0.08 0.0 0.00 0.10 0.0 0.00 PM Peak % Difference 10.11% 0.00% 0.00% 13.76% 0.00% 0.00% Before 0.32 43.0 4.18 0.31 43.2 4.14 Off Peak After 0.34 43.0 4.18 0.32 43.2 4.14 Difference 0.02 0.0 0.00 0.02 0.0 0.00 % Difference 5.72% 0.00% 0.00% 5.98% 0.00% 0.00% TABLE 28 Case Study 3a: Volume changes on the facility Period Direction Before After Difference %Difference EB 9,991 12,547 2,556 25.58% WB 13,718 15,972 2,254 16.43% AM Peak TOT 23,709 28,519 4,810 20.29% EB 17,410 19,171 1,761 10.11% WB 16,305 18,547 2,243 13.76% PM Peak TOT 33,714 37,718 4,004 11.87% EB 41,681 44,063 2,383 5.72% WB 39,612 41,980 2,369 5.98% Off Peak TOT 81,292 86,043 4,751 5.84% EB 69,081 75,780 6,699 9.70% WB 69,635 76,500 6,865 9.86% Total TOT 138,716 152,280 13,564 9.78%

41 USER’S G UIDE Scenario Period AM Peak 12,152,900 381,540 31.9 PM Peak 15,261,700 518,222 29.5 Off Peak 37,208,800 1,179,200 31.6 Before Total 64,623,400 2,078,962 31.1 45,000,186 714,064,881 46,356,879 AM Peak 12,157,300 381,813 31.8 PM Peak 15,264,100 517,755 29.5 Off Peak 37,212,500 1,179,300 31.6 After Total 64,633,900 2,078,868 31.1 44,988,995 713,780,208 46,344,037 Difference 10,500 -94 0.0 -11,191 -284,673 -12,842 % Difference 0.02% 0.00% 0.02% -0.02% -0.04% -0.03% VMT VHT Speed THC CO NOX (mi) (hrs) (mph) (gm) (gm) (gm) TABLE 29 Case Study 3a: Regional results

42 US ER ’S G UI DE CHAPTER 12 CASE STUDY 3B: REMOVE FREEWAY HOV LANE Case Study 3b removes the freeway HOV lanes (one HOV lane in each direction) from a freeway in an urban area. The freeway is uncongested before removal and becomes con- gested after the removal of the lanes. The specific location for this case study is a 2-mile-long section of the Interstate 5 freeway feeding downtown Seat- tle, Washington (see Figure 9). 12.1 APPLICATION The I-5 freeway mainline has four lanes in each direction with a 45-mph free-flow speed and 1,800-vph/lane capacity and an HOV lane in each direction with a 60-mph free-flow speed and a 1,500-vph/lane capacity. In this project, the HOV lanes in both directions are removed. Thus, the section now has four lanes in each direction and no HOV lane. 12.2 CASE STUDY RESULTS The NCHRP 25-21 methodology was used to compute the impacts of the traffic-flow improvement on the 2020 base case trip tables by time period (AM peak, PM peak, and off peak) and by mode (SOV, HOV, and transit). The revised trip tables were then reassigned to the improved network to deter- mine the impact on VMT, VHT, and emissions. The results are summarized in Tables 30 through 32. This case study illustrates the impacts of closing an HOV facility on a freeway that is near capacity. The effect is to increase congestion on the freeway in the mixed-flow lanes in the peak direction of travel (reverse commute is generally unaffected). The travel time in the peak direction is increased from 6 to 50 percent during the peak periods. Peak-period traffic (HOV plus SOV) on the facility is reduced by almost 10 percent. The result of the HOV lane closure is a 0.02-percent reduc- tion in regional daily VMT and a 0.06-percent reduction in mobile source emissions.

43 USER’S G UIDE Project Figure 9. Case Study 3b: Remove HOV lanes from I-5. Period Scenario NB SB V/C Speed (mph) Time (min) V/C Speed (mph) Time (min) AM Peak Before 0.90 35.7 3.84 0.57 36.6 3.70 After 0.94 35.0 4.06 0.67 36.6 3.70 Difference 0.04 -0.7 0.22 0.10 0.0 0.00 % Difference 4.66% -1.84% 5.73% 18.01% 0.00% 0.00% PM Peak Before 0.71 37.1 3.61 0.91 36.6 3.70 After 0.82 37.1 3.61 0.96 31.4 5.44 Difference 0.11 0.0 0.00 0.05 -5.2 1.74 % Difference 15.69% 0.00% 0.00% 5.32% -14.21% 47.03% Off Peak Before 0.39 37.1 3.61 0.36 36.6 3.70 After 0.46 37.1 3.61 0.42 36.6 3.70 Difference 0.07 0.0 0.00 0.06 0.0 0.00 % Difference 19.30% 0.00% 0.00% 17.71% 0.00% 0.00% TABLE 30 Case Study 3b: Travel time changes on the facility Period Direction Before After Difference %Difference AM Peak NB 22,667 19,439 -3,229 -14.24% SB 14,256 13,809 -447 -3.14% TOT 36,923 33,248 -3,676 -9.96% PM Peak NB 17,771 16,865 -906 -5.10% SB 22,799 19,706 -3,093 -13.57% TOT 40,570 36,571 -3,999 -9.86% Off Peak NB 57,936 56,577 -1,359 -2.35% SB 54,086 52,309 -1,777 -3.29% TOT 112,022 108,886 -3,136 -2.80% Total NB 98,374 92,881 -5,493 -5.58% SB 91,141 85,824 -5,317 -5.83% TOT 189,515 178,705 -10,810 -5.70% TABLE 31 Case Study 3b: Volume changes on the facility

44 US ER ’S G UI DE Scenario Period Before AM Peak 12,152,900 381,540 31.9 PM Peak 15,261,700 518,222 29.5 Off Peak 37,208,800 1,179,200 31.6 Total 64,623,400 2,078,962 31.1 45,000,186 714,064,881 46,356,879 After AM Period 12,151,000 381,565 31.8 PM Period 15,254,600 517,364 29.5 Off-Peak 37,205,400 1,179,400 31.5 Total 64,611,000 2,078,329 31.1 44,973,564 713,524,692 46,320,617 Difference -12,400 -633 0.0 -26,622 -540,189 -36,262 % Difference -0.02% -0.03% 0.01% -0.06% -0.08% -0.08% VMT VHT Speed THC CO NOX (mi) (hrs) (mph) (gm) (gm) (gm) TABLE 32 Case Study 3b: Regional results

45 USER’S G UIDE CHAPTER 13 CASE STUDY 4: NARROW STREET Case Study 4 removes a single through lane from each direction of an uncongested suburban highway. The specific location for this case study is a 10.1-mile-long section of State Route 169 between I-405 and SR 18 near Renton, Washington (see Figure 10). 13.1 APPLICATION The project removes one mixed-flow lane in each direction to SR169, the Renton-Maple Valley Highway. Before the case study changes, SR 169 has a single-lane HOV and two lanes in each direction. Thus, the project results in the sec- tion having one mixed-flow lane and a single HOV lane in each direction. 13.2 CASE STUDY RESULTS The NCHRP 25-21 methodology was used to compute the impacts of the traffic-flow improvement on the 2020 base case trip tables by time period (AM peak, PM peak, and off peak) and by mode (SOV, HOV, and transit). The revised trip tables were then reassigned to the improved network to deter- mine the impact on VMT, VHT, and emissions. The results are summarized in Tables 33 through 35. This case study illustrates the impacts of removing a mixed-flow lane in each direction from a rural highway that is near capacity. The effect is to cause modest congestion in the peak direction of travel during each peak period. Peak- period travel times increase 6 to 10 percent in the peak direc- tions only. Total daily traffic volumes drop about 7 percent. The net effect of the lane removal on regional VMT and emissions is less than 0.01 percent.

46 US ER ’S G UI DE Project Figure 10. Case Study 4: Remove through lane from SR 169. Period Scenario EB WB V/C Speed (mph) Time (min) V/C Speed (mph) Time (min) AM Peak Before 0.07 35.8 17.19 0.30 35.7 17.22 After 0.13 35.8 17.19 0.47 32.8 18.25 Difference 0.07 0.0 0.00 0.18 -2.9 1.03 % Difference 100.22% 0.00% 0.00% 59.59% -8.22% 5.98% PM Peak Before 0.29 35.8 17.19 0.13 35.7 17.22 After 0.46 31.7 18.82 0.25 35.7 17.22 Difference 0.18 -4.1 1.63 0.12 0.0 0.00 % Difference 60.70% -11.55% 9.48% 98.87% 0.00% 0.00% Off Peak Before 0.08 35.8 17.19 0.07 35.7 17.22 After 0.16 35.8 17.19 0.14 35.7 17.22 Difference 0.08 0.0 0.00 0.07 0.0 0.00 % Difference 98.03% 0.00% 0.00% 98.00% 0.00% 0.00% Period Direction Before After Difference %Difference AM Peak EB 568 568 0 -0.06% WB 2,517 2,002 -515 -20.47% TOT 3,085 2,569 -516 -16.71% PM Peak EB 2,464 1,976 -488 -19.81% WB 1,066 1,058 -8 -0.73% TOT 3,530 3,034 -496 -14.05% Off Peak EB 4,217 4,175 -42 -0.99% WB 3,728 3,690 -37 -1.00% TOT 7,944 7,865 -79 -0.99% Total EB 7,249 6,719 -530 -7.31% WB 7,311 6,750 -560 -7.66% TOT 14,559 13,469 -1,090 -7.49% TABLE 33 Case Study 4: Travel time changes on the facility TABLE 34 Case Study 4: Volume changes on the facility

47 USER’S G UIDE Scenario Period Before AM Peak 12,152,900 381,540 31.9 PM Peak 15,261,700 518,222 29.5 Off Peak 37,208,800 1,179,200 31.6 Total 64,623,400 2,078,962 31.1 45,000,186 714,064,881 46,356,879 After AM Peak 12,154,000 381,502 31.9 PM Peak 15,260,900 518,126 29.5 Off Peak 37,207,400 1,179,100 31.6 Total 64,622,300 2,078,728 31.1 44,998,481 714,063,588 46,356,164 Difference -1,100 -234 0.0 -1,705 -1,293 -715 % Difference 0.00% -0.01% 0.01% 0.00% 0.00% 0.00% VMT VHT Speed THC CO NOX (mi) (hrs) (mph) (gm) (gm) (gm) TABLE 35 Case Study 4: Regional results

48 US ER ’S G UI DE CHAPTER 14 CASE STUDY 5: ACCESS MANAGEMENT Case Study 5 tests the impacts of converting a suburban highway into a limited-access expressway. The specific loca- tion for this case study is a 10.1-mile-long section of State Route 169 between I-405 and SR 18 near Renton, Washing- ton (see Figure 11). 14.1 APPLICATION The stretch of SR 169 is a suburban highway with few sig- nalized intersections and frequent driveways to access fronting land uses. The access management project includes median barriers the length of the expressway (thus enabling increased speeds) with signalized intersections at median breaks to serve fronting land uses and allow U-turns. The more frequent sig- nalized intersections and the concentration of access at these intersections result in less green time available for through traf- fic on the expressway, thus reducing capacity. However, this reduced capacity is counterbalanced by increased speeds between intersections with the elimination of fronting access to the facility except at the signalized intersections. SR 169 has two lanes in each direction. The section between I-405 and 140th Avenue SE is an urban arterial with added HOV lanes. The access management project was estimated to yield the following improvements: • The capacity of the rural portion of the route was reduced from 1,500 vph/lane to 1,200 vph/lane to account for the effects of adding traffic signals to the rural unsignalized intersections. • The free-flow speed for the entire length of the route was increased from 34 mph to 40 mph to account for the greater speeds possible with expressway operations. 14.2 CASE STUDY RESULTS The NCHRP 25-21 methodology was used to compute the impacts of the traffic-flow improvement on the 2020 base case trip tables by time period (AM peak, PM peak, and off peak) and by mode (SOV, HOV, and transit). The revised trip tables were then reassigned to the improved network to deter- mine the impact on VMT, VHT, and emissions. The results are summarized in Tables 36 through 38. This case study illustrates the impacts of improving facility speed and capacity through access control of fronting devel- opment driveways and side streets. Unlike the earlier case studies, both peak-period and daily travel times are improved by over 10 percent. Daily traffic on the facility is increased 67 percent. The net effect of the access control improvements is to increase regional VMT by less than 0.01 percent. However, the traffic-flow smoothing effects of the project cause a reduc- tion in regional emissions of 0.02 to 0.04 percent.

49 USER’S G UIDE Project Figure 11. Case Study 5: SR-169 access management. Period Scenario EB WB V/C Speed (mph) Time (min) V/C Speed (mph) Time (min) AM Peak Before 0.07 35.8 17.19 0.30 35.7 17.22 After 0.12 39.8 15.15 0.62 39.7 15.18 Difference 0.06 4.0 -2.04 0.33 4.0 -2.04 % Difference 82.61% 11.18% -11.87% 110.86% 11.20% -11.85% PM Peak Before 0.29 35.8 17.19 0.13 35.7 17.22 After 0.56 39.8 15.15 0.24 39.7 15.18 Difference 0.28 4.0 -2.04 0.12 4.0 -2.04 % Difference 95.12% 11.18% -11.87% 94.76% 11.20% -11.85% Off Peak Before 0.08 35.8 17.19 0.07 35.7 17.22 After 0.16 39.8 15.15 0.15 39.7 15.18 Difference 0.08 4.0 -2.04 0.07 4.0 -2.04 % Difference 92.60% 11.18% -11.87% 103.51% 11.20% -11.85% Period Direction Before After Difference %Difference AM Peak EB 568 878 310 54.64% WB 2,517 4,492 1,975 78.48% TOT 3,085 5,370 2,286 74.09% PM Peak EB 2,464 4,063 1,599 64.91% WB 1,066 1,753 687 64.45% TOT 3,530 5,817 2,287 64.77% Off Peak EB 4,217 6,794 2,578 61.13% WB 3,728 6,354 2,626 70.45% TOT 7,944 13,148 5,204 65.50% Total EB 7,249 11,736 4,487 61.90% WB 7,311 12,599 5,288 72.34% TOT 14,559 24,335 9,776 67.14% TABLE 36 Case Study 5: Travel time changes on the facility TABLE 37 Case Study 5: Volume changes on the facility

50 US ER ’S G UI DE Scenario Period Before AM Peak 12,152,900 381,540 31.9 PM Peak 15,261,700 518,222 29.5 Off Peak 37,208,800 1,179,200 31.6 Total 64,623,400 2,078,962 31.1 45,000,186 714,064,881 46,356,879 After AM Peak 12,152,800 381,263 31.9 PM Peak 15,260,400 517,202 29.5 Off Peak 37,211,700 1,179,000 31.6 Total 64,624,900 2,077,465 31.1 44,983,926 713,840,873 46,349,753 Difference 1,500 -1,497 0.0 -16,260 -224,008 -7,126 % Difference 0.00% -0.07% 0.07% -0.04% -0.03% -0.02% VMT VHT Speed THC CO NOX (mi) (hrs) (mph) (gm) (gm) (gm) TABLE 38 Case Study 5: Regional results

51 USER’S G UIDE CHAPTER 15 CASE STUDY 6: INTERSECTION CHANNELIZATION Case Study 6 consists of the addition of left-turn and right- turn lanes to all four approaches of an urban intersection at Martin Luther King Jr. Way and Rainer Avenue in Seattle (see Figure 12). 15.1 APPLICATION The addition of left-turn and right-turn lanes in the case study were coded as a 20-percent increase in the capacity on each approach. Before the improvement, all the approaches to the intersection had two lanes and a capacity of 1,300 vph other than the northbound approach, which had a capacity of 1,400 vph. After a 20-percent increase in capacity, the north- bound approach has a capacity of 1,650 vph, and all the other approaches have a capacity of 1,550 vph, while the speed and number of lanes is unchanged. 15.2 CASE STUDY RESULTS The NCHRP 25-21 methodology was used to compute the impacts of the traffic-flow improvement on the 2020 base case trip tables by time period (AM peak, PM peak, and off peak) and by mode (SOV, HOV, and transit). The revised trip tables were then reassigned to the improved network to deter- mine the impact on VMT, VHT, and emissions. The results are summarized in Tables 39 through 43. This case study shows the impacts of intersection chan- nelization improvements at an uncongested intersection. The channelization improvements result in travel time improve- ments of below the threshold of detectability for the method- ology. The result is no change in predicted traffic volumes for the intersection, no predicted changes in regional VMT, and no predicted changes in regional emissions.

52 US ER ’S G UI DE Project Figure 12. Case Study 6: Intersection of MLK Jr. Way and Rainer Avenue in Seattle. Period Scenario NB SB V/C Speed (mph) Time (min) V/C Speed (mph) Time (min) AM Peak Before 0.04 31.4 2.55 0.04 31.4 2.55 After 0.03 31.4 2.55 0.03 31.4 2.55 Difference 0.00 0.0 0.00 0.00 0.0 0.00 % Difference -9.68% 0.00% 0.00% -8.92% 0.00% 0.00% PM Peak Before 0.05 31.4 2.55 0.05 31.4 2.55 After 0.05 31.4 2.55 0.04 31.4 2.55 Difference 0.00 0.0 0.00 0.00 0.0 0.00 % Difference -8.59% 0.00% 0.00% -7.07% 0.00% 0.00% Off Peak Before 0.02 31.4 2.55 0.02 31.4 2.55 After 0.02 31.4 2.55 0.02 31.4 2.55 Difference 0.00 0.0 0.00 0.00 0.0 0.00 % Difference -9.01% 0.00% 0.00% -7.59% 0.00% 0.00% Period Scenario EB WB V/C Speed (mph) Time (min) V/C Speed (mph) Time (min) AM Peak Before 0.10 31.9 2.40 0.28 32.0 2.40 After 0.09 31.9 2.40 0.26 32.0 2.40 Difference -0.01 0.0 0.00 -0.02 0.0 0.00 % Difference -8.04% 0.00% 0.00% -8.21% 0.00% 0.00% PM Peak Before 0.32 31.9 2.40 0.17 32.0 2.40 After 0.30 31.9 2.40 0.15 32.0 2.40 Difference -0.02 0.0 0.00 -0.01 0.0 0.00 % Difference -7.66% 0.00% 0.00% -8.31% 0.00% 0.00% Off Peak Before 0.10 31.9 2.40 0.09 32.0 2.40 After 0.10 31.9 2.40 0.09 32.0 2.40 Difference -0.01 0.0 0.00 -0.01 0.0 0.00 % Difference -7.87% 0.00% 0.00% -8.24% 0.00% 0.00% TABLE 39 Case Study 6: North/south travel time changes on the facility TABLE 40 Case Study 6: East/west travel time changes on the facility

53 USER’S G UIDE Period Direction Before After Difference %Difference AM Peak NB 302 302 0 0.00% SB 286 286 0 0.00% TOT 588 588 0 0.00% PM Peak NB 446 446 0 0.00% SB 373 373 0 0.00% TOT 819 819 0 0.00% Off Peak NB 1,153 1,153 0 0.00% SB 943 943 0 0.00% TOT 2,096 2,096 0 0.00% Total NB 1,901 1,901 0 0.00% SB 1,601 1,601 0 0.00% TOT 3,502 3,502 0 0.00% Period Direction Before After Difference %Difference AM Peak EB 799 799 0 0.00% WB 2,192 2,192 0 0.00% TOT 2,991 2,991 0 0.00% PM Peak EB 2,522 2,522 0 0.00% WB 1,298 1,298 0 0.00% TOT 3,820 3,820 0 0.00% Off Peak EB 4,880 4,880 0 0.00% WB 4,446 4,446 0 0.00% TOT 9,326 9,326 0 0.00% Total EB 8,201 8,201 0 0.00% WB 7,935 7,935 0 0.00% TOT 16,136 16,136 0 0.00% Scenario Period Before AM Peak 12,152,900 381,540 31.9 PM Peak 15,261,700 518,222 29.5 Off Peak 37,208,800 1,179,200 31.6 Total 64,623,400 2,078,962 31.1 45,000,186 714,064,881 46,356,879 After AM Peak 12,152,900 381,540 31.9 PM Peak 15,261,700 518,222 29.5 Off Peak 37,208,800 1,179,200 31.6 Total 64,623,400 2,078,962 31.1 45,000,186 714,064,881 46,356,879 Difference 0 0 0.0 0 0 0 % Difference 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% VMT VHT Speed THC CO NOX (mi) (hrs) (mph) (gm) (gm) (gm) TABLE 41 Case Study 6: North/south volume changes on the facility TABLE 42 Case Study 6: East/west volume changes on the facility TABLE 43 Case Study 6: Regional results

54 US ER ’S G UI DE CHAPTER 16 CASE STUDY 7: SIGNAL COORDINATION Case Study 7 tests the impacts of a 0.5-mile-long arterial signal coordination project that increases mean speeds by 10 percent. The specific location for this case study is Montblake Boulevard, at the University of Washington in Seattle (see Fig- ure 13). The project consists of six signals over 0.54 miles. 16.1 APPLICATION The signal coordination project is coded as a 10-percent improvement in the free-flow speed. No capacity changes were made. 16.2 CASE STUDY RESULTS The NCHRP 25-21 methodology was used to compute the impacts of the traffic-flow improvement on the 2020 base case trip tables by time period (AM peak, PM peak, and off peak) and by mode (SOV, HOV, and transit). The revised trip tables were then reassigned to the improved network to deter- mine the impact on VMT, VHT, and emissions. The results are summarized in Tables 44 through 46. This case study illustrates the impacts of the optimization of signal coordination for a 0.5-mile section of urban arter- ial. The improvements provide a 9-percent reduction in travel times for the peak direction of the peak periods and a similar reduction for both directions during the off-peak period. Daily traffic on the facility is predicted to increase by 27 percent. The net effect on regional travel is to reduce regional VMT by 0.01 percent and to reduce regional emissions by 0.02 to 0.04 percent.

55 USER’S G UIDE Project Figure 13. Case Study 7: Montblake Boulevard, Seattle. Period Scenario NB SB V/C Speed (mph) Time (min) V/C Speed (mph) Time (min) AM Peak Before 0.39 17.0 5.30 0.39 18.5 4.31 After 0.42 18.6 5.10 0.64 20.4 3.92 Difference 0.02 1.7 -0.20 0.25 1.9 -0.39 % Difference 6.00% 9.72% -3.77% 63.10% 10.00% -9.05% PM Peak Before 0.50 18.5 4.33 0.43 16.2 6.60 After 0.54 20.3 3.94 0.45 17.6 6.68 Difference 0.04 1.8 -0.39 0.02 1.4 0.08 % Difference 8.91% 9.88% -9.01% 5.07% 8.47% 1.21% Off Peak Before 0.21 18.5 4.33 0.20 18.5 4.31 After 0.22 20.3 3.94 0.32 20.4 3.92 Difference 0.01 1.8 -0.39 0.12 1.9 -0.39 % Difference 5.13% 9.88% -9.01% 58.64% 10.00% -9.05% AM Peak NB 2,167 2,297 130 6.00% SB 2,262 3,666 1,405 62.10% TOT 4,428 5,963 1,535 34.65% PM Peak NB 2,801 3,048 248 8.84% SB 2,401 2,538 137 5.68% TOT 5,202 5,586 384 7.38% Off Peak NB 7,091 7,455 364 5.13% SB 6,719 10,697 3,978 59.20% TOT 13,811 18,152 4,341 31.43% Total NB 12,059 12,800 741 6.15% SB 11,382 16,901 5,519 48.48% TOT 23,441 29,700 6,260 26.70% Period Direction Before After Difference %Difference TABLE 44 Case Study 7: Travel time changes on the facility TABLE 45 Case Study 7: Volume changes on the facility

56 US ER ’S G UI DE Scenario Period Before AM Peak 12,152,900 381,540 31.9 PM Peak 15,261,700 518,222 29.5 Off Peak 37,208,800 1,179,200 31.6 Total 64,623,400 2,078,962 31.1 45,000,186 714,064,881 46,356,879 After AM Peak 12,150,900 381,266 31.9 PM Peak 15,259,400 517,865 29.5 Off Peak 37,204,700 1,176,500 31.6 Total 64,615,000 2,075,631 31.1 44,983,926 713,840,873 46,349,753 Difference -8,400 -3,331 0.0 -16,260 -224,008 -7,126 % Difference -0.01% -0.16% 0.15% -0.04% -0.03% -0.02% VMT VHT Speed THC CO NOX (mi) (hrs) (mph) (gm) (gm) (gm) TABLE 46 Case Study 7: Regional results

57 USER’S G UIDE CHAPTER 17 CASE STUDY 8:TRANSIT IMPROVEMENT Case Study 8 involves doubling the frequency of bus line 7B (PSRC Model Line 4007), providing bus service on Broadway between Downtown Seattle and South Rainier Beach. This bus line extends for 23.55 route-miles (see Fig- ure 14). 17.1 APPLICATION Before improvement, the bus service operated at 30-minute peak-hour headways, with a mean speed of 15 mph. After improvement, the peak-hour headway was cut from 30 min- utes to 15 minutes. Note that the service improvements were meant to apply to both the AM peak and PM peak hours. However, in the PSRC model, transit trips outside of the AM peak hour are not assigned to specific transit lines. Conse- quently, the test of the PM peak-hour service improvements could not be made with the available PSRC model database. (However, one option for overcoming this limitation of the database would have been to take the AM peak-hour impacts of the project and double them to approximate the combined AM and PM peak-hour impacts.) 17.2 CASE STUDY RESULTS The NCHRP 25-21 methodology was used to compute the impacts of the traffic-flow improvement on the 2020 base case trip tables by time period (AM peak, PM peak, and off peak) and by mode (SOV, HOV, and transit). The revised trip tables were then reassigned to the improved network to determine the impact on VMT, VHT, and emissions. Facility-specific results were not computed. The regional impacts of the transit ser- vice improvements were found to be negligible within the precision of the reported results. There were no significant differences in VMT or emissions.

58 US ER ’S G UI DE Project Seattle Figure 14. Case Study 8: Transit improvement.

59 USER’S G UIDE CHAPTER 18 CASE STUDY 9: REMOVE PARK-AND-RIDE LOT Case Study 9 looks at the impacts of removing a bus rapid transit park-and-ride lot from a critical freeway facility feeding downtown Seattle. The park-and-ride lot is located on SR 520 at Hunts Point, about 2.5 miles west of I-405 (see Figure 15). 18.1 APPLICATION The park-and-ride lot is located at node 890 in the PSRC model. The lot is effectively removed from the model by dis- allowing the use of mode i (auxiliary auto mode) on the cen- troid connector from Node 890 to Node 3126. 18.2 CASE STUDY RESULTS The NCHRP 25-21 methodology was used to compute the impacts of the traffic-flow improvement on the 2020 base case trip tables by time period (AM peak, PM peak, and off peak) and by mode (SOV, HOV, and transit). The revised trip tables were then reassigned to the improved network to determine the impact on VMT, VHT, and emissions. The results are summarized in Table 47. Facility-specific results were not tabulated. The results show that eliminating the park-and-ride lot would increase daily VMT by 0.14 percent. Regional vehicle emissions would be increased by slightly less than 0.1 percent. These results show that in this case, construction of a park- and-ride lot at this location would result in net reductions in VMT and vehicle emissions.

60 US ER ’S G UI DE Project Seattle Figure 15. Case Study 9: Remove park-and-ride lot. Scenario Period Before AM Peak 12,152,900 381,540 31.9 PM Peak 15,261,700 518,222 29.5 Off Peak 37,208,800 1,179,200 31.6 Total 64,623,400 2,078,962 31.1 45,000,186 714,064,881 46,356,879 After AM Peak 12,184,800 382,704 31.8 PM Peak 15,288,700 518,357 29.5 Off Peak 37,242,100 1,180,300 31.6 Total 64,715,600 2,081,361 31.1 45,041,575 714,679,954 46,393,679 Difference 92,200 2,399 0.0 41,389 615,073 36,800 % Difference 0.14% 0.12% 0.03% 0.09% 0.09% 0.08% VMT VHT Speed THC CO NOX (mi) (hrs) (mph) (gm) (gm) (gm) TABLE 47 Case Study 9: Regional results

61 USER’S G UIDE CHAPTER 19 CASE STUDY 10: LONG-RANGE REGIONAL TRANSPORTATION PLAN Case Study 10 involves the implementation of 30 years of transit and highway improvements in the Seattle region. The improvements include all improvements that actually occurred between 1990 and 2000, plus the planned improvements con- tained in the 20-year transportation plan from 2000 to 2020. 19.1 APPLICATION Two PSRC highway networks (1990 and 2020) were com- pared to obtain the transportation system improvements that occurred between 1990 and 2000, plus the improvements planned between 2000 and 2020. The 2020 highway network has 11 percent more centerline-miles of road and 13 percent more capacity than the 1990 network (see Table 48). The 2020 highway network has the following traffic-flow improvements over the 1990 network: • New freeway HOV lanes, • Freeway mixed-flow lane additions, • New freeway sections, • Urban street lane additions, • New urban streets, • Rural road lane additions, and • New rural roads. The 2020 transit network has 32 percent more transit vehi- cles and 21 percent more route-miles than the 1990 network (see Table 49). The 2020 transit network has the following transit service improvements over the 1990 network: • New transit lines, • Frequency increases for existing service, • Extensions of existing transit lines, • New park-and-ride lots, and • New stations. 19.2 RESULTS OF PSRC MODEL RUNS The base PSRC model was run on three scenarios: • 1990 demand loaded on 1990 network, • 2020 demand loaded on 1990 network, and • 2020 demand loaded on 2020 network. The resulting VMT and VHT are shown in Table 50. The PSRC model predicted that the 2020 highway and transit net- work improvements would result in a 4-percent increase in daily VMT and a 9-percent reduction in daily VHT for the region. 19.3 NCHRP 25-21 METHODOLOGY RESULTS The NCHRP 25-21 methodology was used to compute the impacts of not building the 30-year improvement program. The impacts on the 2020 base case trip tables of retaining the 1990 network were predicted by time period (AM peak, PM peak, and off peak) and by mode (SOV, HOV, and transit). The revised trip tables were then reassigned to the 1990 net- work to determine the impact on VMT, VHT, and emissions. The results are summarized in Table 51. In contrast with the standard PSRC model, which predicts that the 30-year improvement program would increase VMT, the NCHRP 25-21 methodology predicts that the 30-year pro- gram of transportation improvements would decrease VMT by 0.7 percent. The NCHRP 25-21 methodology furthermore predicts that emissions would be reduced by 6 to 7 percent.

62 US ER ’S G UI DE Network Centerline-Miles Lane-Miles No. of Links Capacity-Miles Mean Free-Flow Speed (mph) 1990 10,266 15,704 15,171 17,802,716 21.2 2020 11,388 17,390 17,711 20,194,252 19.9 Difference 1,122 1,686 2540 2391536 -1.3 % Difference 10.9% 10.7% 16.7% 13.4% -6.1% Network Transit Vehicles Lines Route-Miles 1990 972 448 8,065 2020 1,286 542 9,716 Difference 314 94 1,651 % Difference 32.3% 21.0% 20.5% Scenario Demand Network Period VMT VHT MPH #902 1990 1990 AM 3hr Peak 12,049,800 384,443 31.3 #903 PM 3hr Peak 15,085,000 498,400 30.3 #904 Off Peak 37,113,300 1,175,500 31.6 Total Day 64,248,100 2,058,343 31.2 #1002 2020 1990 AM 3hr Peak 18,459,100 696,708 26.5 #1003 PM 3hr Peak 26,472,000 1,088,500 24.3 #1004 Off Peak 56,319,200 2,081,600 27.1 Total Day 101,250,300 3,866,808 26.2 #2002 2020 2020 AM 3hr Peak 19,363,700 630,847 30.7 #2003 PM 3hr Peak 27,670,300 977,495 28.3 #2004 Off Peak 58,668,900 1,927,400 30.4 Total Day 105,702,900 3,535,742 29.9 Difference 4,452,600 -331,066 % Difference 4.4% -8.6% Source: base case 2020 EMME2 databank, module 6.11. Impact of Difference = (the results for the 2020 demand loaded on the 2020 network) – (the results for the 2020 demand loaded on the 1990 network). Scenario Period 2020 AM Peak 12,152,900 381,540 31.9 Demand PM Peak 15,261,700 518,222 29.5 On Off Peak 37,208,800 1,179,200 31.6 2020 Total 64,623,400 2,078,962 31.1 45,000,186 714,064,881 46,356,879 Network 2020 AM Peak 12,313,300 446,992 27.5 Demand PM Peak 15,432,600 642,538 24.0 On Off Peak 37,331,500 1,200,800 31.1 1990 Total 65,077,400 2,290,330 28.4 48,285,393 761,290,211 49,348,300 Network Difference -454,000 -211,368 2.7 -3,285,207 -47,225,330 -2,991,421 % Difference -0.70% -10.17% 8.59% -7.30% -6.61% -6.45% VMT VHT Speed THC CO NOX (mi) (hrs) (mph) (gm) (gm) (gm) TABLE 48 Case Study 10: Highway improvements 1990 to 2020 TABLE 49 Case Study 10: Transit service improvements 1990 to 2020 TABLE 50 Comparison of baseline VMT and VHT estimates by PSRC model TABLE 51 Case Study 10: Regional results