Click for next page ( 28


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 27
28 though trip lengths in statewide models exceed the duration NHTS or NPTS (5) of a peak period. Although some states intend to investigate ATS (4) traffic microsimulation for their statewide models, actual ap- Special long distance travel survey(s) (3) plications of microsimulation have not yet been reported. Employment data from private vendor(s) (2) Employment data from public source(s) (2) Given the limitations of the available traffic assignment Roadside surveys (2) algorithms, most states have chosen to ignore the peak period Tourism economic or attendance data (2) or do simple factoring of 24-h traffic into a peak. Here are the U.S. Census (1) adopted methods. Economic data from a private vendor(s) (1) FAA data (1) Factored by percent of traffic in peak from National and state park attendance database (1) traffic counts (7) MPO survey data (1) Peak period assigned directly (6) Borrowed data from another model (1) No factoring into peak (5) Borrowed parameters from another model (1) Post-processed in another manner (2) Own long distance survey (1) Own household survey (1) No state overtly includes peak spreading. Massachusetts Seasonal traffic counts (1) and Ohio reported having time-of-day models. Ohio's model includes travel time as a variable in its utility expression; therefore, there is some sensitivity to traffic congestion. FREIGHT COMPONENTS Freight components of statewide models do more than Every model uses speed and volume curves, such as the simply complement passenger components, as is typical for BPR curve, for delay calculations. urban models. Indeed, the driving forces behind statewide model development in many states are economic develop- Special Treatment of Long Distance Trips ment issues that cannot be fully analyzed without a good freight component. In addition, freight is more easily ana- Beyond computational and data problems associated with the lyzed statewide than for urban areas because the scale of detail of networks in statewide models, the greatest obstacle the geography is more compatible with available freight for forecasting is good representation of long distance trip data sources. Thus, freight components for statewide mod- making. Obtaining good information on long distance trips eling have evolved to a level of sophistication well beyond has become more difficult as the 1995 ATS has aged; there- what is seen within MPO models. Freight components fore, states have discovered numerous ways to work around sometimes include commercial vehicles that are not carry- this limitation. ing a commodity. All states with models are cognizant of the need to include The following discussion includes Montana's freight long trips, and a little less than two-thirds of the states re- component that is part of HEAT. HEAT is primarily a tool ported taking special actions to model long distance travel. A for economic forecasting, but contains a commodity-based solid majority of these states create special trip purposes for truck model. long distance travel. There are three approaches: (1) seg- menting existing trip purposes into short and long distance categories; (2) creating separate trip purposes, such as recre- Nature of Freight Components ation/tourism, to capture long trips; and (3) Fratar factoring an OD table of long distance trip purposes. There are two fundamentally different styles of freight fore- casting: (1) direct forecast of vehicle flows without refer- Some other fixes were necessary, depending on the state. ence to commodities or (2) forecasting of commodities, then Delaware, Maine, and New Hampshire, in particular, reported using the commodity flow forecast to estimate vehicle the need to account for tourism during the summer months. flows. Of the 16 states reporting freight components as a California found it necessary to introduce k-factors during trip part of their statewide travel forecasting model, 12 base their distribution; to use composite impedances for input to the forecasts on commodities. Although they are much more gravity expression and to modify friction factors to account complex, commodity-based models have a greater sensitiv- for long distance travel. ity to economic conditions and to state policies toward in- dustrial development. A variety of data sources, cited here, were used specifi- cally to model long distance travel. There is no consensus as Commodity flows are derived from data sources in either to the preferred data sources. Notably, Ohio performed a 2- tons or dollars. Finding the effects of freight on the trans- week household survey of trips in excess of 50 mi. Four portation system requires that commodity flows be converted states are still using the 1995 ATS. to trucks, rail cars, shiploads, aircraft, barges, or containers.

OCR for page 27
29 The correct conversion requires knowledge of how much of If a model is commodity-based, it is likely that states a commodity is carried by a particular vehicle. These payload would need data on commodity flows for calibrating their factors (tons per vehicle) can be obtained from several trip generation and distribution steps. Slightly more then half sources, as listed here. of the states with freight components purchased the TRANSEARCH database from Reebie to understand com- VIUS (Florida, Georgia, Michigan, Montana, modity flows. Three states were able to use the CFS instead. Ohio, Wisconsin (6) Oregon performed its own shipper and carrier survey. Commercial freight data vendor (Kentucky, Louisiana, Tennessee, Texas) (4) None of the major sources of commodity flows are com- Rail Carload Waybill Sample (Georgia, Indiana, plete. Some states have adopted different methods of dealing Ohio) (3) with missing commodities or industrial sectors and empty Data from another state or from an MPO trucks, but the main objective is to adjust for the error by com- (Kentucky, Virginia) (2) paring assignment results to truck counts. Indiana, Louisiana, Truck intercept studies (Georgia) (1) and Virginia use OD table estimation from traffic counts to bridge the missing commodities and account for empties. VIUS pertains only to trucks, and the Rail Carload Way- Florida ignores the missing categories and Kentucky lumps all bill Sample only to railroads. However, the Rail Carload the missing categories into one catch-all commodity group. Waybill Sample can provide estimates of the density of com- Ohio did its own establishment survey, which included all modities, which can then be applied to other modes. commercial vehicle movements, not just freight shipments. Some freight components are closely tied to models of Freight components that are commodity-based usually re- economic activity (e.g., Ohio's new model and Oregon's quire that commodity production totals be estimated for each model) that account for commodity flows in units of dollars. commodity category for each zone. Almost all states with To forecast vehicular flows there is an additional need for a this requirement derived commodity productions from em- conversion between dollars and tons. Two states, Georgia ployment estimates and commodity output per employee. and Indiana, reported that their principal source of data on Kentucky obtained its production totals directly from Reebie. dollars per ton is the CFS. Similarly, freight components that are commodity-based Many sources of freight data give commodity flows as usually require that commodity consumption totals be esti- yearly totals. For single-day forecasts (or peak periods with mated for each commodity category for each zone. Estimat- a single day) it is necessary to determine the fraction of ing consumption is more difficult that estimating production, yearly commodities transported in a day. This fraction can be because (1) the commodities consumed by an industry are not obtained implicitly through OD table estimation techniques obvious by looking at the nature of the industry and (2) house- or explicitly by calculating the number of truck days in a holds consume a large fraction of the commodities. One year. The number of truck days ranges from 261 in Kentucky method of understanding commodity consumption is IO to 365 in Texas and Virginia. analysis (as suggested in NCHRP Report 260); however, only Michigan, Montana, Ohio, Oregon, and Vermont use IO. All The distribution of commodities from zone to zone is han- states (except Kentucky) with a need to estimate commodity dled by three methods, alone or in combination: (1) Fratar consumption by zone do it through employment estimates factoring a vehicle or commodity OD table that was created along with consumption per employee or through household from data, (2) a gravity expression, or (3) a logit expression. estimates and commodity consumption per household. Most states use a gravity expression. The new models in Ohio and Oregon, because of their economic activity and Commodity flow databases are often reported for fairly land use underpinnings, use logit expressions. large spatial units such as counties or states. There is a need in some states to expand the flow matrices to cover much smaller Five states (Indiana, Louisiana, Ohio, Tennessee, and Vir- spatial units. Half of the states with freight components created ginia) reported using techniques of OD table estimation from procedures for disaggregating their commodity flows. The ground counts for improving their truck forecasts; however, method most often cited by states was to factor county-to- Indiana used OD table estimation only for non-freight truck county flows into zone-to-zone flows using employment cat- traffic and Ohio will be abandoning these techniques when egories and population totals. its new model is completed. Commodity flows must be divided among modes. Only Four states (Florida, Indiana, Michigan, and Vermont) re- Florida, Ohio, and Oregon reported using mode split expres- ported using "quick response" methods, such as the ones sions (such as logit) to allocate commodities to modes. The from the Quick Response Freight Manual, to supplement remaining states use fixed shares from data. Indiana varies their freight forecasts. For example, Florida used these tech- these shares by the distance of the shipment, which is facili- niques only for non-freight truck trips. tated by the way data are reported from the CFS. A model

OCR for page 27
30 with fixed shares does not necessarily mean that the propor- Air freight (5) tion of tonnage carried by each mode remains constant. Total Deep water shipping (4) mode shares can shift as commodity production and con- Inland water shipping (3) sumption patterns change in the future. However, fixed-share Less than truckload and truckload (1) models are insensitive to changes in shipping costs that may give an advantage to one mode over another. Florida, Ohio, and Wisconsin reported having all five of these major modes. All states with freight components have A few commodity-based components further calculate the at least a truck network or a passenger network that has been fraction of commodities carried by each truck type. All states modified for trucks. Ohio and Texas have networks for other reported using fixed shares, derived from the CFS, Reebie's modes besides trucks. Almost all states worked with all TRANSEARCH database, expert judgment, or the VIUS. trucks together or just worked with heavy trucks. Michigan and Ohio divided trucks into heavy, medium, and light cate- Traffic assignments usually involve the mixing of passen- gories, similar to the categories in the Quick Response ger and freight traffic. States have adopted two methods of as- Freight Manual. Montana's HEAT divided trucks between signing a mix of traffic: (1) preloading trucks onto highway truckload and less-than-truckload. links, and then performing a passenger car assignment or (2) loading trucks and passenger cars together. Preloading is often All statewide models with a freight component do a 24-h done with an all-or-nothing traffic assignment. When trucks truck forecast. Five states also reported the ability to do a and passenger cars are assigned together, a static user-optimal peak-period truck forecast. equilibrium traffic assignment is preferred. The decision be- tween the two methods (preloading or together) revolves It is desirable, but not necessary, that the in-state zone sys- around the question of whether truck routing is heavily influ- tem for a freight forecast correspond to the zone system for enced by traffic congestion, which is essentially ignored in an a passenger forecast. All states reported consistent zone sys- all-or-nothing assignment. When trucks are assigned together tems except Texas, which has a coarser zone system for with passenger cars, a multiclass traffic assignment algorithm freight. Kentucky, Ohio, and Virginia use subzones or grid is required to account for the mix of vehicles on each link and cells to increase the spatial detail where necessary. Because to ensure that trucks are assigned only to legal routes. Because of the ready availability of freight OD data for the whole trucks have a greater impact on congestion than passenger cars, United States, a majority of statewide freight components it is further necessary to weight truck volumes by a passenger- cover most or all of the continental United States rather than car-equivalent factor when calculating delays from a multi- relying on external stations at the state borders. Half of the class traffic assignment. For example, Louisiana derived a statewide freight components cover parts of either Canada or statewide average factor of 1.83 from the Highway Capacity Mexico. Manual, a value that is appropriate for the terrain in that state. The following is a list of the ways in which zones are de- Only two states (Ohio and Wisconsin) explicitly handled fined for out-of-state portions of the freight component. transshipment of commodities. None of the states chose to use national transportation analy- sis regions. Some models used multiple sources of zones, de- Four states used FHWA's FAF for network development. pending on how far the area is from the state border. A majority of states obtained network information from a GIS and used GIS for freight network storage. Counties or aggregations of counties (6) BEA regions or aggregations of BEA regions (6) Freight Component Level of Detail States or aggregations of states (6) TAZs (2) Freight components can be either multimodal or concentrate External stations (1) on a single mode. No state reported concentrating on a single Multistate regions (1) mode other than trucks, even though numerous railroad-only models have been described in the literature. Cited here are Freight components use special generators sparingly, and the modes reported as forecasted by statewide models. None most models do not have any. New Jersey has the most with of the models dealt directly with truckrail intermodal, or 200. Special generators include rail yards, airports, seaports, indeed, any other intermodal pairings. Models also ignored truck terminals, warehouses or distribution centers, pipeline categories of trucks that would be related to the economic terminals, and regional shopping malls. structure of the trucking industry, such as a for-hire truck or a private truck. All truck networks have links that are coded to the same highway functional classes as passenger car networks. A Truck, general (15) state's truck network has about the same number of links as Rail freight (5) its passenger car networks.