National Academies Press: OpenBook

Guide to Truck Activity Data for Emissions Modeling (2019)

Chapter: Section 4 - Speeds

« Previous: Section 3 - Age Distributions
Page 21
Suggested Citation:"Section 4 - Speeds." National Academies of Sciences, Engineering, and Medicine. 2019. Guide to Truck Activity Data for Emissions Modeling. Washington, DC: The National Academies Press. doi: 10.17226/25484.
×
Page 21
Page 22
Suggested Citation:"Section 4 - Speeds." National Academies of Sciences, Engineering, and Medicine. 2019. Guide to Truck Activity Data for Emissions Modeling. Washington, DC: The National Academies Press. doi: 10.17226/25484.
×
Page 22
Page 23
Suggested Citation:"Section 4 - Speeds." National Academies of Sciences, Engineering, and Medicine. 2019. Guide to Truck Activity Data for Emissions Modeling. Washington, DC: The National Academies Press. doi: 10.17226/25484.
×
Page 23
Page 24
Suggested Citation:"Section 4 - Speeds." National Academies of Sciences, Engineering, and Medicine. 2019. Guide to Truck Activity Data for Emissions Modeling. Washington, DC: The National Academies Press. doi: 10.17226/25484.
×
Page 24
Page 25
Suggested Citation:"Section 4 - Speeds." National Academies of Sciences, Engineering, and Medicine. 2019. Guide to Truck Activity Data for Emissions Modeling. Washington, DC: The National Academies Press. doi: 10.17226/25484.
×
Page 25
Page 26
Suggested Citation:"Section 4 - Speeds." National Academies of Sciences, Engineering, and Medicine. 2019. Guide to Truck Activity Data for Emissions Modeling. Washington, DC: The National Academies Press. doi: 10.17226/25484.
×
Page 26

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.

21 4.1 Overview For project-scale analysis, MOVES accepts inputs of average speed by source (vehicle) type on the project link, where average speed = average travel time/distance (length of link). Travel time should account for the average delay attributable to traffic signal operation. A single average speed is input for all source (vehicle) types; if different speeds for different source types are to be modeled, the user must create a separate link for each source type. The average speed should be specific to the analysis hour(s). For county-scale analysis, MOVES accepts an average speed distribution. This distribution is a set of 16 fractions that sum to 1, which represents the distribution of vehicle hours traveled among 16 speed bins. MOVES requires this information for every combination of 13 source (vehicle) types, 4 road types, 24 hours of the day, and 2 types of days (weekdays/weekends). Table 4.1 illustrates the input format. Average speeds in MOVES are intended to capture the overall driving experience of vehicles (time taken over a distance driven during trips) and not just the nominal speed (speed limit) or level-of-service speed. Differences in how speed is deter- mined can have a significant effect on the average speed value. 4.2 MOVES Embedded Data Defaults are not available for either county- or project-scale analysis. Speeds can vary sub- stantially by analysis area, and the EPA notes that for SIP and conformity analysis, states are expected to develop and use local estimates of average speed. MOVES2014 does include default distributions for national-scale analysis that are based on light-duty probe vehicle data collected by TomTom. The speed distributions do not vary by state, but do vary by road type and hour of the day. Single-unit short-haul and long-haul trucks (source types 52 and 53) share the same MOVES default speed distributions as light-duty vehicles. Distri- butions for combination short-haul and long-haul trucks (source types 61 and 62) were adjusted to reduce the average speed by 8 percent, on the basis of research on car versus truck speeds on California freeways (U.S. EPA, 2016). For the 2014 NEI, the EPA developed average speed estimates using data provided by StreetLight (Eastern Research Group, Inc., 2017). County-level distributions are based on state distributions in many cases where local samples are inadequate. 4.3 Sensitivity/Importance The impact of speed on emissions can range from modest to significant, depending upon the speed range, vehicle type, and pollutant. One analysis found that reducing truck speeds by 5 mph compared to a standard regional speed distribution increased fine particulate matter (PM2.5) S E C T I O N 4 Speeds

22 Guide to Truck Activity Data for Emissions Modeling by 6 percent, nitrogen oxide (NOx) by 2 percent, and VOC by 1 percent (Porter et al., 2014c). Another analysis, comparing speed distributions submitted to the 2011 NEI across states, also found modest differences across states (Eastern Research Group, Inc., 2013). In contrast, another analysis looked at changes in individual source type emissions when speeds were varied from congested [level of service (LOS) F] through uncongested conditions (LOS B) and points in between (Noel and Wayson, 2012). The emission changes depend on source type and pollutant, but in many cases emissions change by 50 percent or more in this scenario. Although this was a regional analysis, because the same logic and underlying data are used at the regional and project scale to model different average speeds, this analysis is applicable to project-level runs too. Another analysis conducted by the NCHRP Project 08-101 team compared the impact on truck emissions of using National Performance Management Research Data Set (NPMRDS) versus MOVES default speed distributions, on data from Virginia. PM2.5 emissions based on the NPMRDS ranged from 20 to 28 percent lower on urban unrestricted access roads to 11 to 16 percent higher on rural unrestricted access roads, compared to emissions based on MOVES distributions. Differences in NOx emissions were typically less than 8 percent. Case Study #5 dis- cusses the results in more detail (see Appendix E, available in NCHRP Web-Only Document 261). Results of recent sensitivity analyses are shown in Table 4.2. Analysis of NPMRDS data in Virginia found substantial differences between speed dis- tributions observed in the NPMRDS and MOVES defaults (Cambridge Systematics, 2017). Figure 4.1 provides an example speed comparison for urban unrestricted access roads across all hours of the day and days of the week (see Case Study #5 for other road types). The default distributions are flatter with higher fractions of vehicle hours of travel (VHT) occurring in the lowest speed bins compared to the NPMRDS distributions. A possible explanation for this Bin Description (not an input field) sourcetypeID roadtypeID hourdayID avgSpeed BinID avgSpeed Fraction speed < 2.5 mph 52 2 185 1 0.0000000 2.5 mph <= speed < 7.5 mph 52 2 185 2 0.0000000 7.5 mph <= speed < 12.5 mph 52 2 185 3 0.0000000 12.5 mph <= speed < 17.5 mph 52 2 185 4 0.0000000 17.5 mph <= speed < 22.5 mph 52 2 185 5 0.0000000 22.5 mph <= speed < 27.5 mph 52 2 185 6 0.0000000 27.5 mph <= speed < 32.5 mph 52 2 185 7 0.0212651 32.5 mph <= speed < 37.5 mph 52 2 185 8 0.0027255 37.5 mph <= speed < 42.5 mph 52 2 185 9 0.0000000 42.5 mph <= speed < 47.5 mph 52 2 185 10 0.0000000 47.5 mph <= speed < 52.5 mph 52 2 185 11 0.0000000 52.5 mph <= speed < 57.5 mph 52 2 185 12 0.0000000 57.5 mph <= speed < 62.5 mph 52 2 185 13 0.3890143 62.5 mph <= speed < 67.5 mph 52 2 185 14 0.0590208 67.5 mph <= speed < 72.5 mph 52 2 185 15 0.5279743 72.5 mph <= speed 52 2 185 16 0.0000000 Table 4.1. Example average speed distribution input.

Speeds 23 Source Type Road Type Delta VOC Delta NOx Delta PM2.5 Reduce Truck Speeds by 5 mph in Average Speed Distributiona All 1% 2% 6% Increase Speed Distribution from 90th to 10th Percentile of Statesb All All All 12% 18% 18% Change Speed Distribution from LOS B to LOS Fc Single-Unit Truck Urban Restricted 93% 64–65% 64–66% Combination Truck Urban Restricted 78–84% 27% 70–73% MOVES2014a Default to VA NPMRDS Speed Distributiond Single Unit Urban Unrestricted N/A −18% −28% Short-Haul Truck Urban Restricted 0% 5% Rural Unrestricted 5% 11% Rural Restricted 4% 17% Combination Urban Unrestricted −14% −20% Long-Haul Truck Urban Restricted −4% −7% Rural Unrestricted 0% 16% Rural Restricted −3% 5% aPorter et al., 2014c (using MOVES 2010). bEastern Research Group, Inc., 2013 (using MOVES 2010). cNoel and Wayson, 2012 (using MOVES 2010). dCambridge Systematics, 2017 (using MOVES2014). Table 4.2. Impact on emissions of changing speed inputs. Source: Cambridge Systematics (2017) using NPMRDS data from 2014. Includes all hours of day and days of week. 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 2.5 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 Fraction Speed Bin (mph) NPMRDS Truck Default Single-Unit Truck Default Combination Truck Figure 4.1. Urban unrestricted access speed distribution—Virginia, 2014.

24 Guide to Truck Activity Data for Emissions Modeling discrepancy is that compared to defaults, NPMRDS data might be collected mainly on arterial and limited-access roads, where speeds tend to be higher than collector or local streets. How- ever, the extent to which these differences reflect differences between state and national data, differences in data collected in different years, or the inclusion of only National Highway System (NHS) data in the NPMRDS data set is not known. Other researchers have found that early versions of the NPMRDS data over-represented lower-speed vehicles (Hallenbeck et al., 2016). The Virginia comparison also examined truck versus car speed distributions from the NPMRDS. In a comparison of average truck speeds to average car speeds on the same roadway route, average truck speeds appear about 7 percent lower than average car speeds across all road- way types. However, in a comparison of the ratio of weighted average speeds across segments, average truck speeds appear closer to average car speeds. This appears to be due to the higher amount of truck travel on higher-speed roads. This information is included to show some of the considerations that need to be taken into account when using a fleetwide average speed distribu- tion, which is primarily weighted to cars, to represent truck speeds. In total, the combined sensitivity analyses suggest that small changes in average speed are not very influential on overall daily, regional, emissions estimates, but that correctly identi- fying when and where congestion occurs will have a significant impact on hourly emission estimates, especially for project-scale analysis. At a regional scale, the choice of data source also may have a moderate impact on emissions estimates. 4.4 Generating Local Data Volumes 1 and 2 of NCHRP Web-Only Document 210 note the following options for generating local speed data (Porter et al., 2014a, 2014b): • Monitoring (base/historical year data only): – Study-specific field surveys. – Public agency ITS systems (loops, radar, video). – Vehicle-based monitoring systems (GPS, cell phone, Bluetooth). • Modeling (historical and forecast year): – Stand-alone volume-delay functions. – Travel demand forecasting model/speed postprocessing methods. The options for obtaining speed data specific to trucks are more limited and include the following: • Field survey methods that can distinguish cars and trucks. • Vehicle-based monitoring systems in which the vehicle type is identified. • Adjusting an average speed or speed distribution for all vehicles based on observed differences between truck and car speeds. Vehicle-based monitoring data that are distinct for cars versus trucks (and possibly, medium- versus heavy-duty trucks) can be acquired from private data providers, and examples are discussed here. These data are coming into widespread use and are more likely to be utilized than custom field surveys. 4.4.1 NPMRDS Source, Availability, and Cost NPMRDS is a nationwide data set of roadway speeds by road segment and time period for the NHS. FHWA funds the NPMRDS and makes it available to states and metropolitan plan- ning organizations (MPOs) to support performance measurement. The passenger car and truck

Speeds 25 speed data in the NPMRDS are provided to FHWA by Inrix. Truck data are collected from trucks in fleets with GPS data loggers. Level of Data-Processing Effort This analysis requires modest effort to identify link speeds for project-scale analysis or moderate effort to develop county-scale speed distributions. Data-Processing Steps: Project Scale The steps of the analysis include the following: 1. If there is one (or a fraction of one) traffic management center (TMC) link corresponding to a project segment, average the travel time on this link across all time periods in the hour. 2 If there are multiple TMC links corresponding to a project segment: a. For each time period, sum the truck travel time on each link across the corresponding links to obtain a total travel time. If a link partially extends outside the project boundary, its travel time should be reduced in proportion to the length outside the project boundary. b. Average the total travel time for each time period over all time periods in the hour. 3. Compute speed for the hour = Sum(TMC lengths)/average travel time. This is equivalent to computing the length-weighted harmonic speeds across all time periods and TMCs for an hour. Data-Processing Steps: County Scale County-scale processing is somewhat more complicated due to the need to match NPMRDS road types with MOVES road types and to match NPMRDS travel time data with traffic volume data and segment lengths so that travel times by segment can be weighted appropriately. This is the most challenging issue and requires use of segment information from the Highway Per- formance Monitoring System (HPMS) and volume information from the Traffic Management Analysis System (TMAS), both of which are maintained by FHWA. A memorandum prepared for the Virginia Department of Transportation (DOT) (Cambridge Systematics, 2017) describes the following approach to processing NPMRDS data for develop- ment of MOVES speed inputs: 1. Join the HPMS database with truck volume data from TMAS based on functional classifica- tion ID and urban/rural code. 2. Limit NPMRDS records to truck speed records only. 3. Discard NPMRDS records with extremely high or low speed values. Thresholds of 3 mph and 90 mph were used in the example, but a threshold lower than 3 mph might be warranted in areas experiencing significant amounts of extreme congestion. 4. Aggregate NPMRDS travel time data from the 5-minute level to the hourly level. 5. Join the NPMRDS data set with the HPMS database according to the TMC link ID and roadway name, begin mile post, and end mile post. 6. Calculate hourly volumes from truck annual average daily traffic (AADT), functional class from HPMS, hourly volume distribution from TMAS, and hours from NPMRDS. 7. Aggregate hourly volumes from HPMS segments to TMC segments. 8. Aggregate NPMRDS speeds from TMC segments to routes as provided in the TMC descrip- tion table (this step may not be necessary if segment-level data are retained.). Space mean speeds are calculated based on aggregated route lengths and travel times, and TMC volumes are aggregated to route volumes using distance as weight. Check aggregated speed values to ensure they fall within reasonable ranges.

26 Guide to Truck Activity Data for Emissions Modeling 9. Calculate VHT on each route by dividing the length of the route (miles) by speed (miles/ hour) and multiplying by AADT. 10. Tag each route with a MOVES speed bin based on the route speed. 11. Sum VHT by speed bin for each hour and weekday/weekend. Note that the segments are defined differently in the NPMRDS (which uses TMC segments) and TMAS (which reports volumes for HPMS segments), so the volume-weighting of speeds by segment will not be precise. The time periods for the speed and volume observations also may not match exactly. Applicability and Limitations The procedure described here could be applied in any state or metropolitan area with access to NPMRDS data. Because NPMRDS data include only the NHS, these data are most appropriate for representing higher functional class roads and may not be representative of the entire road system. The relative impact on emissions may differ depending upon the specific speed distribu- tions observed within a given state. 4.4.2 Other Resources Volumes 1 and 2 of NCHRP Web-Only Document 210 (Porter et al., 2014a, 2014b) discuss a variety of methods available for obtaining observed or modeled speed data for general vehicle populations (not specific to trucks). Case Study #5 (see Appendix E, available in NCHRP Web-Only Document 261) summarizes findings from a state DOT’s analysis of NPMRDS truck and car speeds, suggests procedures and implications for local application, and provides a sensitivity analysis showing how the use of different truck and car speed distributions affects emissions. As of this writing, FHWA is undertaking research to evaluate the nationwide use of NPMRDS for developing both county-scale and project-scale emissions estimates.

Next: Section 5 - Drive Schedules and OMDs »
Guide to Truck Activity Data for Emissions Modeling Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

TRB’s National Cooperative Highway Research Program (NCHRP) Research Report 909: Guide to Truck Activity Data for Emissions Modeling explores methods, procedures, and data sets needed to capture commercial vehicle activity, vehicle characteristics, and operations to assist in estimating and forecasting criteria pollutants, air toxics, and greenhouse gas emissions from goods and services movement.

Goods movement is a vital part of the national economy, with freight movement growing faster than passenger travel. The growth in freight traffic is contributing to urban congestion, resulting in hours of delay, increased shipping costs, wasted fuel, and greater emissions of greenhouse gas and criteria pollutants. The limited national data on urban goods movement are insufficient for a thorough understanding of the characteristics of the trucks operating in metropolitan areas and the complex logistical chains that they serve.

For instance, there are at least three different segments of urban freight—long haul, drayage, and pickup and delivery. It is believed that truck fleet characteristics differ between the segments, but only local registration data exist at a level of detail needed to support regional transportation plans, transportation improvement plans, and state implementation plans. The lack of data on all types of commercial trucks affects model estimation and results in inaccurate base year emissions inventories, limiting the ability to design and implement effective policies to reduce freight-related emissions.

NCHRP Research Report 909 enumerates various sources of truck data and how they can be obtained and used to support emissions modeling.

NCHRP Web-Only Document 210: Input Guidelines for Motor Vehicle Emissions Simulator Model (Porter et al., 2014a, 2014b, 2014c) provides guidance on developing local inputs to the MOVES mode. It covers all vehicle types, but is not specific to trucks. NCHRP Research Report 909 supplements NCHRP Web-Only Document 210 by describing the use of various data sources to obtain truck-specific inputs.

Appendices A through G to NCHRP Research Report 909 are published as NCHRP Web-Only Document 261 and contain seven case studies that serve as the basis for much of the guidance provided in NCHRP Research Report 909.

NCHRP Research Report 909 is also supplemented by three MS Excel files that contain data from the case studies:

Case Studies #1 and #7

Case Study #2

Case Studies #3, #4, and #6

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!