Click for next page ( 78


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 77
Methods for Detailed Analysis G-77 Element 3 Handling of Reliability Effects As the volume of traffic rises beyond 80 percent and toward 100 percent of the facility design capacity, there tends to be an exponential increase in the incidence and severity of delays due to non-recurring and unpredicted events such as accidents, mechanical breakdowns, special events, or hazardous materials spills. Various studies have found that such incidents account for over half of total congestion delays on both freeways and arterial roadways. Various other studies have shown that diminishing reliability (increasing variation) in travel time has a particularly high cost for truck traffic, since it affects vehicle delivery schedules. Penalty factors have been developed for application to average time delays in situations where travel time reliability also degrades. Those factors can be used with travel demand models to effectively increase the valuation of time sav- ings benefits for congestion reduction scenarios that also improve travel time reliability. There is a full discussion of the measurement and modeling of travel time reliability, its valuation, and application with travel demand models, in the previously cited NCHRP Report 463: Economic Implications of Congestion. To avoid redundancy, readers of this guide are referred to that document for a more complete discussion of methods to account for reliability impacts of congestion. Read- ers are also referred to the previously cited Portland report for case study examples of the impact of congestion-induced travel time reliability degradation on business scheduling costs. 5.1.6 Required Resources In general, travel demand models involve data such as Forecasts of trip generation rates by households and businesses; Forecasts of car/truck/bus/rail mode split; Model specification of road system links and nodes; Model specification of traffic control data at intersections and junctions; Observed traffic volumes (counts) on road links (daily or peak/off-peak); Observed travel time and speed data; and Observed traffic delay and queue data. When considering rail freight solutions for traffic congestion, it becomes particularly impor- tant to be able to distinguish truck shares of traffic on the key congested areas and corridors and to forecast changes in congestion during peak periods for those areas and corridors. 5.2 Identify Carrier Cost and Service Levels 5.2.1 Overview This step identifies the carrier costs and service capabilities of rail, truck, and intermodal options for moving freight. An understanding of carrier costs is necessary to understand how new projects and facility investments, changes in operations, or new public policies can affect carrier costs. The outcome of this step is then used later to calculate the broader logistics cost associated with use of truck and rail alternatives by freight system users (shippers). 5.2.2 Components The analysis of carrier costs and service features is based on a classification of different types of freight carriage, each of which has its own set of cost and service features. The classes are Truckload freight service; Intermodal (rail/truck) freight service;

OCR for page 77
G-78 Guidebook for Assessing Rail Freight Solutions to Roadway Congestion Unit train freight service; and General rail freight services. 5.2.3 Background An understanding of relative costs and prices of the various transport options is essential for anyone trying to identify useful projects. In recent years, cost and service models have been devel- oped at many different levels of detail. Many of the costs shown in this section were originally developed as estimates of transportation costs and/or rates as of the year 2000 or earlier. Costs are somewhat higher as of the year 2006 and will likely increase in future years. Rough estimates of costs, operations, and resource utilization can be extremely helpful in initial planning stud- ies. These estimates are not intended to be used for any specific shipment, and they certainly should not be used as indications of future cost or price levels. 5.2.4 Factors The carrier cost and service features are very different for the various classes of truck and rail options, and those differences are reflected in both the rules-of-thumb methods and the more sophisticated costing model methods discussed next. However, all of these methods key off of common factors that serve to distinguish the various freight transportation options. Those fac- tors are Length of the average shipment (miles or km); Per mile line haul operating costs; Size of the average shipment (whether it is truckload or less than a truckload, a full train or less than a full train); Additional terminal or transfer costs associated with less than truckload or less than trainload shipments; and Frequency and speed of shipment. 5.2.5 Methods Element 1 Truckload Freight Movement Many different kinds of trucks are to be seen on the highways, but only the largest carry freight that might be divertible to rail. Tractor-trailer combinations that can carry 20 or more tons of freight are commonly used for long-distance trucking. Specialized trucks are used for moving automobiles, chemicals, bulk commodities, and other heavy products that might also be rail competitive. The routes taken and the miles traveled by smaller trucks might relate to the loca- tion of industrial plants, warehouses, and retail establishments, so there could be a long-term relationship between the use of rail, the location of such facilities, and the nature of local truck movements. However, discussions of diverting freight to rail must focus on the larger trucks. Large trucks might be carrying freight to a single customer (truckload or TL) or freight destined to multiple customers (less-than-truckload or LTL). Both TL and LTL are divertible to intermodal and possibly to rail carload services. For LTL movements, railroads can be involved in the movement of a trailer or container of consolidated shipments from one trucking terminal to another; railroads are no longer competitive in terms of the pickup and delivery of the indi- vidual LTL shipments. The truck traffic of interest is therefore either TL or the linehaul portion of LTL. Approach 1 Overall Rules of Thumb. Distance is an important cost factor. For bulk traffic, as discussed above, railroads can handle even very short trips using a very efficient mini-train.

OCR for page 77
Methods for Detailed Analysis G-79 For general merchandise traffic, rail is competitive only for hauls of at least a couple of hundred miles. For these rail-competitive movements, the trucking operation is straightforward: drive to the customer's loading dock, load the truck, drive to the destination, and unload the truck. While costs also vary with the specifics of the journey, trucking costs per mile are predictable. For many years, they have been on the order of $1 to $1.35 per milei for general freight moving in standard equipment for distances over 300 to 400 miles. Costs for shorter haul movements depend to a great extent on the time required to load and unload and the average speed of the highway trip. These factors determine the number of loads per day that can be handled by a truck driver--this is a more important measure than distance for truckload costs. Interviews with truckers indicated that they need to charge a total of about $500 to 525 per day to cover their costs in short-haul service. The trucking market is highly competitive; prices were deregulated in 1980; and prices have been close to costs ever since. For the preliminary analysis, it is probably sufficient to assume that truck costs for a standard tractor-trailer combination range from $1 to 1.35. However, truck rates have recently been rising, and somewhat higher costs may be needed in future studies. For specialized trucking, a recent study estimated costs of $1.35 per mile for system loads and $2.60 for restricted loads. These costs were obtained from a larger, efficient tank carrier in 2003. System loads were shipments where the company would be able to reload the truck because (1) commodities did not contaminate the trailer and limit its next use and (2) the length of haul was long enough to make it worth seeking a back-haul load. Restricted loads were the opposite: the nature of the commodity limited reuse or the distance was too short to do other than return empty to base. The distance limit was defined by what a driver could do in an out-and-back run, which was about 250 miles each way. Thus, the cost of tank shipments under favorable reload conditions could also be on the order to $1.35 per mile, but would likely be considerably higher. Approach 2 Modeling Cost Components. More precise estimates of trucking costs can be obtained with further knowledge of operating conditions and current costs. The major compo- nents of trucking costs fall into the following categories: Truck driver costs are on the order of $0.35 to $0.40 per mile for long-distance, non-unionized truck drivers who drive in excess of 100,000 miles per year; $40,000 or more per year applies for unionized truck drivers, who are typically involved in LTL or specialized operations. Both costs have been rising in recent years and are expected to rise higher than the historical rate in order to attract and retain drivers. Ownership costs for the tractor-trailer combination are on the order of $100,000; the tractor might have a useful life in long-haul service of 5 or more years and the trailer should last 10 or more years. Operators of big fleets typically will sell equipment before the end of its useful life. It is an advantage to them to sell tractors after 3 to 4 years, i.e. before the normal 400,000-mile engine warranty expires. Costs are higher and also rising for specialized equipment, such as refrigerated units or tank-trailers. The purchase prices can be transformed into a cost per trip as follows: Calculate the equivalent annual ownership cost over the expected life of the vehicle, assum- ing a reasonable discount rate (e.g., the weighted average cost of capital for the trucking industry); Estimate the number of days per year that the equipment will be used; Divide the equivalent annual ownership cost by the expected number of days the equipment will be used to get the daily cost of the equipment; i These figures include fuel costs, but could be significantly higher during periods of fuel price spikes.

OCR for page 77
G-80 Guidebook for Assessing Rail Freight Solutions to Roadway Congestion Divide by the typical number of hours used per day to get hourly cost; and Estimate the cycle time, which is time required for the trip, taking into account the hours avail- able for working each day (which can be as high as 10 to 11 hours per day for 6 days per week) and the time required for each activity. Note 1: Positioning the equipment for loading: For most TL traffic, empty repositioning is small, on the order of 10 percent of total miles. For efficient bulk operations, large trucks cycle between a given origin and a given destination, and the empty miles equal the loaded miles. For general freight, it may be necessary to travel empty 50 to 75 miles to pick up the next load, independent of the length of haul. Empty mileage tends to be higher for specialized equipment. Note 2: Loading, Loaded movement and Unloading: Allocate costs to a particular trip by multiplying the hourly or daily cost by the hours or days required. Maintenance costs for equipment. Some maintenance costs will vary with time, others with mileage. Fuel: Large trucks typically achieve 5 to 7 mpg (as of the year 2005). Fuel can be allocated on a per-mile basis along with maintenance. Tolls can be allocated on a per-mile basis using typical values for a generic trip or actual tolls for a specific route. Fees and taxes will vary by state; most can be included in daily equipment costs. Insurance costs have driven some smaller fleets out of business. Some modifications to truck costs can be considered. The rate/ton-mile is what is important relative to mode split, and this can be estimated by dividing the rate by the shipment size or typical payload of the truck. (The 15 tons/load figure cited above is a good factor for general merchandise, but payloads will be higher for commodities that have higher density.) In most states, it is possible to use multiple trailers of various kinds, which will reduce the cost per ton on the order of 10 to 20 percent. The service provided by rail-competitive trucks is also easy to estimate, since most long- distance trucking uses the interstate system. A trip of up to 500 miles can be done overnight. A trip of 1,000 miles can be done in 2 days. Faster service (1,000 or more miles per day) can be provided by using two drivers. Long-distance trucking is very reliable compared to rail, so it is usually not necessary to worry about the distribution of trip times. Low empty mileage and efficient equipment utilization are the keys to low operating costs. In order to maximize loaded-miles and the total weight carried, carriers will sometimes consolidate several loads, often from one origin to several destinations or from multiple origins to a single destination. Congestion is a major concern for truckers, because congestion increases travel time, reduces utilization, and limits the amount of work that can be completed in a day. On a very short-run basis, costs of drivers and fuel are the most important, as the cost of equipment is not a day-to-day issue. Element 2 Intermodal Freight Movement Intermodal freight involves a combination of trucking for the pickup and delivery ends (drayage), transferring at an intermodal terminal for rail movement for the longer distance linehaul travel. In general, intermodal is faster and more reliable than general rail service and cheaper than truckload service. Intermodal service levels are generally similar to those for truck. Additional time is required in the terminals, but trains generally can move traffic further in a day than can trucks. For long hauls, therefore, intermodal can be faster than truck, while for shorter hauls, trucks can be faster than intermodal service. The differences are likely to be on the order of hours, not days. Truck operations also tend to be more reliable, and pickup and delivery times are more flexible--two aspects of service that can be important to some customers.

OCR for page 77
Methods for Detailed Analysis G-81 To evaluate the relative cost of intermodal freight compared to other options, it is necessary to separately consider the three distinct elements of intermodal service: (1) drayage, (2) termi- nals/hubs, and (3) linehaul. Drayage. Costs of drayage consist largely of the costs related to the driver and the tractor, both of which are largely proportional to the time required per dray. Hence, draymen focus on the number of trips per day that they can make, and they are concerned about taking excessive time to pick up or deliver a container. In addition, there are public concerns with traffic con- gestion near the intermodal terminal, vehicle-miles traveled within the congested area, and the related effects on noise, air quality, and quality of life along the routes used to access the terminal. Drayage costs can be below $50 for short hauls, but up to $500 or more for trips more than 200 miles from the terminal; $150 per trailer or container is a typical figure. Drayage costs can be modeled in detail using the same approach described in the previous sec- tion for trucking costs. Terminals/Hubs. Terminal costs include the costs related to the gate operation, lifting con- tainers and trailers on and off the trains, storage of containers, and management of empty equipment. There are economies of scale in intermodal terminals, so that railroads and ter- minal operators try to concentrate the workload at a few high-volume facilities rather than at more, smaller volume but perhaps better located facilities. Depending on the nature and size of the operation, terminal costs can be $50 to $150 per lift. Some intermodal terminals also act as hubs where intermodal traffic is transferred between trains. The transfer operation adds to operating costs, but using hubs makes it easier to consolidate traffic and increase train frequencies in key lanes. Linehaul. Variable linehaul costs include the costs of operating the train, the equipment costs for locomotives and freight cars, maintenance costs for the right-of-way, and costs related to communications and control. As of the year 2005, these costs range from $0.70 to $0.80 per container-mile or trailer-mile for TOFC (truck on flatcar) or COFC (container on flatcar), but only $0.40 to 0.50 per mile for double-stack trains. This compares to trucking costs of $1.00 to 1.35 for TL or the linehaul portion of LTL. Many of the basic concepts of the competition between truckload and intermodal freight options can be understood in terms of a very simple cost comparison: Intermodal - Truck Cost = Drayage + Terminal + Intermodal Linehaul Truck Linehaul = Drayage + Terminal + (Trk$/mile Int$/mile*(1+circuity))*Distance Since the linehaul costs are fairly constant for competitive distance, the basic question is whether or not the intermodal savings per mile are sufficient to offset the added costs for drayage and ter- minal costs. Generally, the trip must be several hundred miles before the linehaul savings from intermodal shipment becomes larger than the added costs associated with drayage and terminals. Factors Affecting Carrier Costs. Lane density is also an important consideration for inter- modal operations. The higher the density, the easier it is to provide frequent service to customers and the easier it is to fill up trains. The geography of the region and the location of the intermodal terminals in respect to the lane also are significant cost factors. A substantial drayage of a hundred miles or more is not necessarily a problem, so long as it is in the general direction that the ship- ment is moving; the distance-related portion of drayage costs would not be much different than if the move were the first portion of a TL move. If the drayage required backtracking a hundred miles or more, then the drayage costs would be a much more significant burden in competing with the direct TL move. Double-stack has grown dramatically because the linehaul savings are so much greater than they are with TOFC. Instead of a linehaul service that offers a modest saving over TL, double-stack

OCR for page 77
G-82 Guidebook for Assessing Rail Freight Solutions to Roadway Congestion cuts the linehaul costs to less than half of truck costs. Railroads and, in some cases, public agen- cies have invested in increasing clearances along the right-of-way in order allow operation of double-stack trains. Other technological changes are further reducing intermodal costs or flexibility. The Road- Railer technology allows specialized trailers or containers to be hauled in very efficient trains that can be assembled in small terminals without using expensive equipment. Two primary concepts have been used in specialized services. The original concept was to use trailers that had a rail axle as well as the traditional highway axles; the axles could be lowered or raised hydraulically in order to assemble and disassemble trains. The extra weight of the rail axle proved to be a competitive burden, as it reduced the load that could be carried on the highways. The newer concept was to use rail "cars" that were basically a pair of axles with a shelf that could hold up a trailer. A forklift could move these bogies around to facilitate train assembly, and the trailers would not have the extra weight of the wheels. A RoadRailer train is remarkable for its low wind resistance, which improves fuel efficiency at higher speeds, and for its very low loss and damage rate. This type of equipment is not quite as efficient as double-stack trains, so it has been used in specialized traf- fic lanes that lack the volume to support double-stack service. Other innovations that may also change intermodal costs are the Expressway and Rolling Highway classes of equipment and similar rail systems. These provide what is effectively a long, articulated platform for hauling any kind of trailers, containers, or even tractor-trailer combi- nations. Like the RoadRailer technology, no specialized terminal lift equipment is necessary and very little space is needed for loading or unloading. This type of technology has been used in Canada for tank and flatbed highway trailers as well as vans, and for many years in Europe to shuttle trucks through tunnels in the Alps. The ability to carry tractor-trailer combinations means that this technology could support other kinds of shuttle services that take highway trucks off the road for movement through metropolitan areas. For example, Chicago Metropolis 2020 has recently recommended that "intermodal bypass service should be developed to shuttle trucks 100 to 400 miles through and around the region."ii Element 3 Unit Train Freight Movement Unit train costs are straightforward. The main cost elements are (1) equipment, (2) opera- tions, and (3) track maintenance. Equipment costs are generally considered to be the cost of ownership and maintenance, and they are allocated based on time (for ownership and some maintenance) or distance traveled (for most maintenance). Equipment costs can be allocated to a shipment based on the cycle time required for the trip and the distance traveled. The modeling approach is the same as described above for TL operations. Operating costs include the costs of the crew, fuel, and communications and control. Crew and fuel costs are most important. Crew costs are determined by complex labor agreements, but, for most unit train services, they will be approximately proportional to train-miles. Fuel costs vary with gross tonnage and the terrain. Track maintenance includes the costs of installing, inspect- ing, and maintaining rail, ties, ballast, and structures. These costs generally vary with the gross tonnage carried. Costs will be somewhat higher if axle loads or operating speeds are higher. Administrative costs and most other costs are commonly assumed to be fixed costs that can be allocated on the basis of tonnage or shipments. Unit train service is generally easy to understand, as it operates similarly to truckload service. The train operates on a continuous cycle between a shipper and a receiver, making 5 to 10 or ii "The Metropolis Freight Plan: Delivering the Goods," Chicago Metropolis 2020, 2004, p. 28

OCR for page 77
Methods for Detailed Analysis G-83 more cycles per month depending on the distance and the time required to load and unload the train. While average speed might only be 20 mph, this allows unit trains to travel on the order of 500 miles per day, which is competitive with trucks for longer distances. Since terminals are usu- ally bypassed, unit train service is reasonably reliable. Element 4 General Freight Train Movement General freight service is used when a shipper uses one or more railcars but less than a full train. The resulting service is more complicated to operate than intermodal or unit train service. The main difference is that freight must go through a series of rail yards; at each yard, freight cars are sorted and assembled into trains. The variable costs of yard operations are on the order of $25 to $100, depending on the size of the facility, the complexity of the operation, and the amount of traffic. There are economies of scale and of density, and railroads have long attempted to expand the geographic coverage of their networks while consolidating switching operations into fewer, larger yards. General freight service requires the railroad to serve the customer directly. Placement of an empty car for loading, picking up the loaded car, delivering the load, and picking up the empty car tend to be time-consuming operations performed by crews handling short trains on light- density lines or in terminal areas. Important trade-offs are embodied in two fundamental deci- sions: how often to provide service on a branch line and how well to maintain the branch line. Higher frequency of service increases crew costs, but lower frequency service leaves cars at customers' sidings for extra days and increases car costs. Lower quality maintenance reduces train speeds and adds to the time required for switching, but better maintenance can greatly increase costs if there is only a small amount of traffic using the line. The trend since the 1920s has been for railroads to reduce the number of branch lines in order to avoid the high costs of operations and maintenance. The size of the shipment is a key factor for general freight service. A rail car can typically carry 3 to 5 times as much freight as a truck, yet the linehaul costs will be similar (i.e., on the order of $1 per mile for boxcar service). Even with the added costs associated with terminals and branch- line operations, costs per ton can therefore be much lower than for TL or intermodal. However, the costs are very situation specific: a move involving many terminals and very light-density branch lines can easily be more expensive than going by truck. Also, a move involving shortline railroads can be cheaper than an equivalent move involving one of the larger railroads, because they may have a much different cost structure related to train crews, track maintenance, and other cost factors. While low cost is a benefit for merchandise traffic, service quality is a major problem. It takes approximately a day for each terminal, and there is a chance of missing a connection at each ter- minal because of delays or lack of room on the outbound train. As a result, service is slow and unreliable; the typical 600-mile trip takes 6 to 8 days, which is much longer and far less reliable than truck service. During congested periods, service deteriorates dramatically. Additional Rail Performance Models Besides the rules-of-thumb estimation approaches and more detailed cost component approaches discussed in the preceding text, computer models also can be used for rail system cost, performance, and supply adequacy. They include Train performance calculators. A TPC calculates train performance (speed and energy con- sumption) as a function of train and route characteristics. Dispatching models. These models predict the movement of trains along a route, taking into account the need for trains to use passing sidings on single-track routes, and the need to allow