Performance Metrics Used by Freight Transport Providers
Wayne D. Cottrell, Ph. D., P. E.
Civil Engineering Department
California State Polytechnic University, Pomona
Pomona, California 91768
wdcottrell@ csupomona. edu
prepared for the
Leonard Transportation Center
California State University San Bernardino
Project 2007- SGP- 1011
September 30, 2008
Maritime Administration, Annual Report to Congress, Fiscal Year 2007 ( www. marad. dot. gov) W. Cottrell Cal Poly Pomona
1
Table of Contents
Section Page
Abstract....................................................................................................................... .................... 1
Executive Summary........................................................................................................................ 1
Industry Overview....................................................................................................................... ... 3
Freight Transport Providers
Trucking....................................................................................................................... ...... 3
Railroads...................................................................................................................... ...... 4
Maritime....................................................................................................................... ...... 4
Pipelines...................................................................................................................... ....... 5
Aviation....................................................................................................................... ...... 7
Other Modes..................................................................................................................... 10
Summary........................................................................................................................ .. 10
Freight Performance Measures: Literature Review....................................................................... 11
Industry Performance Measures and Concerns
Commercial Trucking and Multimodal............................................................................ 15
Railroads...................................................................................................................... .... 19
Ports and Ships.................................................................................................................. 21
Pipelines...................................................................................................................... ..... 22
Aviation....................................................................................................................... .... 24
Other Modes..................................................................................................................... 26
Conclusion..................................................................................................................... ............... 26
References..................................................................................................................... ................ 27
Appendix ( Tables A1 and A2)....................................................................................................... 31 W. Cottrell Cal Poly Pomona
Performance Metrics Used by Freight Transport Providers
Wayne D. Cottrell, Ph. D., P. E., Civil Engineering Department
California State Polytechnic University, Pomona
ABSTRACT
The newly- established National Cooperative Freight Research Program ( NCFRP) has allocated $ 300,000 in funding to a project entitled “ Performance Metrics for Freight Transportation” ( NCFRP 03). The project is scheduled for completion in September 2009. According to the project’s background discussion, “ public and private decisions related to the freight industry should be based on a thorough analysis of the impacts of those decisions. These analyses are routinely made in the private sector but less commonly in the public sector. As the demand for freight movements outstrips the capacity of the nation’s highway, rail, waterway, air, and port systems, the effects are felt as congestion, upward pressure on freight prices, and longer and less reliable transit times. These indicators of distress in the freight transportation system result in increased supply costs for manufacturers, higher import prices, and higher inventory levels. Ultimately, these costs add up to a higher cost of doing business for firms, a higher cost of living for consumers, and a less productive and competitive economy. Such indicators need to be quantified to be useful to decision makers as well as for public education on freight issues. Establishing consistent performance metrics for the freight system will be very helpful in conducting and comparing analyses of the freight system, particularly by identifying the critical data that are needed to assess system performance.” This report investigates freight transportation performance metrics from one perspective; that is, that of the freight transport providers. In combining the findings of this study with those of the NCFRP study, and other efforts, it may be possible to develop a basis for national and international goods movement performance measurement. One objective is for the measures to be used to better understand freight transport issues, and to relieve some of the industry’s “ distress,” while facilitating the economic growth that is facilitated by an efficient goods movement system.
EXECUTIVE SUMMARY
A total of 19 billion tons of freight, having a total value of $ 13 trillion, were moved in the U. S. in 2002. The dominant freight transport modes, in terms of ton- mileage, were trucks, railroads, pipelines, and ships. Aviation had the fifth greatest modal contribution. Trucks accounted for 70% of the value of all shipments in the U. S. Railroads dominated the long- distance ( greater than 500 miles) freight market. Aviation dominated the international shipment of high- valued goods. Freight transportation’s leading providers were Schneider National Carriers and United Parcel Service in the trucking industry, Union Pacific in the railroad industry, Ingram Barge Company in waterborne shipping, FedEx Express in air freight, El Paso Natural Gas in natural gas pipeline throughput, and Enbridge Energy in oil pipeline throughput. The national freight infrastructure was served by extensive highway, railroad, waterway, and pipeline networks, as well as large port and airport systems. The backbone of the highway system is the National Network, an extensive truck system that is essentially equivalent to the 46,871- mile Interstate System. The railroad network encompasses 141,698 miles, of which 95,663 miles are owned by the Class I railroads. The waterborne shipping industry is supported by the nation’s 300- plus ports, the largest of which serve ocean- going vessels in the Gulf of Mexico ( South Louisiana), Pacific Ocean ( Los Angeles- Long Beach), and Atlantic Ocean ( New York City). The 9,300- mile inland commercial waterways system stretches into the interior of the central U. S.; locks, dams and levees are in need of upgrading to sustain the viability of this aspect of the maritime industry. The ( mostly) underground flow of goods is supported by 1,414,200 miles of natural gas pipelines, and 131,353 miles of oil pipelines. There are over 5,200 public- use airports in the U. S., many of which are equipped to accommodate air freight. John F.
1
W. Cottrell Cal Poly Pomona
Kennedy International Airport is the largest air freight gateway in the U. S., and is the busiest freight terminal ( the Ports of Los Angeles and Long Beach are the second busiest). The “ smallest” freight sector, local messenger and delivery services, involves bicycles, motorcycles, and other small delivery vehicles – generally in the core areas of large cities. The industry earned a not- insignificant $ 7.9 billion in 2006.
Performance measurement in the freight transport industry has attracted two realms of interest: that of the public sector, and that of the providers. The public sector is keenly interested in measures that justify policy decisions, such as asset productivity, total shipments, total flow, and so forth. The public sector is also interested in measures that indicate how well regulations and standards are being met. These would include environmental and safety measures, such as total fleet emissions of criteria pollutants, employee injuries, and fatalities. The providers have an interest in economic measures, such as aspects of financial performance, along with equipment, load, haul, employee, and customer service measures. The two realms cross over in only a few areas; for example, productivity, load and haul are all related, are of interest in the public and private sectors. Otherwise, there are significant distinctions. For example, while the public sector is interested in fatalities, injuries and spills, the freight providers seem to be more interested in the effects of these incidents on insurance costs, tort and liability.
There is little uniformity in performance measurement in freight transportation, particularly across the five modes. Some measures, by necessity, are pertinent to only one or two modes, such as “ barrels per day” in the oil pipeline industry, and “ carloads originated” in the railroad industry. Also, there is little agreement on the “ best” or “ most critical” performance measures, even among individual providers. One motor carrier, for example, was using about 300 measures to represent different aspects of its operations and resources. For many of the providers, a large number of measures are financial, with multiple versions of revenue, expenses, and revenue- related ratios, along with the before and after effects of taxes, interest, depreciation, insurance and other costs. Despite the lack of uniformity and consensus, six measures seem to be represented in all five of the freight transport modes:
􀂃
Average length of haul
􀂃
Operating ratio
􀂃
Revenue per ton- mile
􀂃
Tonnage ( total, all loads)
􀂃
Ton- miles or barrel- miles
􀂃
Terminal dwell time or empty miles factor
The average length of haul is a measure of productivity that is useful for separating short- distance from long- distance shipments. Increases in this measure may be indicative of expansion, and possibly service optimization. The operating ratio is one of the simplest measures of financial performance. The operating ratio is simply the total expenses divided by the total revenue. Revenue per ton- mile, tonnage and ton- miles ( or barrel- miles) are all load- and haul- related measures. Ton- miles are used as a key benchmark of freight movement activity in private industry and government. The four main freight modes – trucks, railroads, ships and pipelines – are surprisingly well balanced in the U. S. in terms of their proportions of total ton- mileage. It is possible that a national freight transportation plan or system, incorporating all modes, would aim for a load- haul balance. ( It is possible that such a system might be supported by a well- developed performance measurement system). Terminal dwell time and the empty miles factor are measures of “ non- productivity.” Freight transport providers probably aim to reduce these measures, as a means of improving efficiency and productivity.
The literature on freight transportation and logistics is extensive. Many of the authors suggest how certain aspects of goods movement could be optimized. Some of the authors also recommend measures that could serve as optimization criteria. Optimization is often associated with idealized spatial distributions of activity, which may be difficult to apply in practice. Similarly, many of the recommended measures are theoretically sound, but are difficult to compute or replicate with existing data collection strategies. It is anticipated that the NCFRP Project 03 will be a major step forward in the understanding of freight transport performance measurement, and data collection.
2
W. Cottrell Cal Poly Pomona
INDUSTRY OVERVIEW
According to Freight in America ( BTS 2006) 19 billion tons of freight, having a value of $ 13 trillion, were moved within the U. S. in 2002. The cumulative weight and distance moved was 4.4 trillion ton- miles. On a typical day, 53 million tons of goods were being moved about 12 billion ton- miles within the U. S. transport network ( 68 tons and 15,310 ton- miles per capita). In terms of ton- mileage, the dominant freight transport modes were trucks ( 34%), railroads ( 31%), pipelines ( 16%), and ships ( 11%). The remaining 8% were carried by aircraft and other modes. Trucks were the dominant mode for shipment distances of less than 500 miles, and rail was dominant over longer distances. California led the nation in total commodity flows, with 11% of the total value. The Los Angeles- Long Beach- Riverside metropolitan area led the nation, compared to other metro areas, in commodity flows and weight.
The freight transport industry is complex and intermodal, such that more than one mode may be used to transport a good during its origin- destination journey. Truck plus rail, truck plus water, and rail plus water are common multimodal combinations. Overall, trucking is the most frequently used mode for goods transport, accounting for 70% of the value and 60% of the weight of all U. S. shipments in 2002. On the fringes of the freight transport provider industry are the intermediaries, or “ freight forwarders,” who act as brokers between shippers and carriers. The freight forwarders can generally be categorized as either 3PL or 4PL. Third- party logistics service providers ( 3PL) offer a number of functions, including order processing, warehousing, tracking and payments. Fourth- party logistics service providers ( 4PL) are multifaceted organizations that may link several 3PL companies, while managing worldwide trading systems.
As would be expected, performance measures and concerns vary according to the freight transport market sector. There can also be variation within the given mode, according to the magnitude and scale of operations. The following sections discuss the freight transport providers within the various freight modes.
FREIGHT TRANSPORT PROVIDERS
Trucking
The trucking industry can be classified into three types of carriers: parcel, full truckload ( FTL), and less- than- truckload ( LTL). FTL carriers typically carry fully or partially- loaded containers from a shipper to a single destination. The freight is not handled en route, since all contents are bound for the same point. LTL carriers collect freight from different shippers, consolidating the goods into containers for line- haul to a terminal. The freight may be further sorted or consolidated at the terminal for continued hauling. LTL shipments typically weigh between 100 and 10,000 lb. Some FTL carriers have a 10,000 lb minimum shipment. Parcel carriers typically carry shipments weighing no more than 150 lb, although some carriers were starting to move heavier packages.
Commercial vehicle carriers are commonly classified according to “ small” and “ large.” Small carriers have annual revenue of $ 30 million or less, while large carriers have annual revenue of $ 30 million or more. The Bureau of Transportation Statistics uses a three- part classification scheme, in which Class I carriers have an annual operating revenue of more than $ 10 million, Class II carriers have annual revenue between $ 3 and $ 10 million, and Class III carriers have annual revenue of less than $ 3 million. Just over one million carriers were registered with the Federal Motor Carrier Safety Administration as of 2006. The nation’s “ top 100” motor carriers, in 2004, are listed in Table A1 ( Appendix). Major FTL carriers included Schneider National Carriers ($ 3.20 billion in revenue in 2004), Swift Transportation Company ($ 2.83 billion), J. B. Hunt Transport ($ 2.79 billion), Werner Enterprises ($ 1.68 billion), and Landstar Carrier Group ($ 1.45 billion). Major LTL carriers included United Parcel Service ($ 36.58 billion in revenue in 2004), Federal Express Express ($ 19.49 billion), DHL Worldwide Express ($ 8.57 billion), Federal Express Ground ($ 4.68 billion), and Federal Express Freight ($ 3.22 billion). None of
3
W. Cottrell Cal Poly Pomona
these providers was headquartered in California, although all had satellite offices nationwide. The largest FTL carriers headquartered in California were Pacer International ($ 406.0 million) and Beneto Bulk Transport ($ 73.4 million). The largest California- based LTL carriers were GI Trucking ($ 215.5 million) and Dependable Highway Express ($ 79.5 million). The industry is under a constant state of flux, with mergers, acquisitions, and market- oriented modifications.
Major parcel carriers had traditionally included United Parcel Service ( UPS) and Federal Express ( FedEx). Both carriers “ upgraded” to LTL status, however, with UPS’ purchase of Overnite Corporation in 2005, and FedEx’ acquisition of Viking Freight ( in 1998) and American Freightways ( in 2001). The third- largest parcel carrier, DHL, is also classified as LTL. The Parcel Shippers Association ( www. parcelshippers. org) membership list includes 59 companies.
Combination trucks, which perform the bulk of truck shipments, are served by the National Network ( NN). The NN is essentially identical to the Interstate System. In most States, additional truck routes facilitate penetration into areas not served by the Interstate System. In California, for example, Terminal Access, Service Access, and California Legal routes enable large, legally- sized trucks to access terminals, authorized service routes, and non- Interstate highways. Combination trucks traveled 143.66 billion miles on the National Network and other truck routes in 2005 ( BTS 2007).
Railroads
As of 2002, there were 552 railroad carriers in the U. S., operating over 141,698 miles of track, and earning $ 36.92 billion in revenue. There were four categories of railroad: Class I, Regional, Line- haul, and Switching & Terminal. Class I railroads had revenue of at least $ 347 million in 2006. The seven Class I railroads earned 92% of the U. S.’ freight railroad revenue in 2002. The Class I railroads were:
􀂃
Burlington Northern & Santa Fe
􀂃
CSX Transportation
􀂃
Grand Trunk Corporation ( Canadian National, Grand Trunk Western, Illinois Central, Wisconsin Central)
􀂃
Kansas City Southern
􀂃
Norfolk Southern
􀂃
Soo Line
􀂃
Union Pacific
Regional railroads covered at least 350 route miles and had annual revenue between $ 40 million and the Class I threshold. There were 31 Regional railroads in 2002. Line- haul railroads generally provide point- to- point service within a single State, operating over fewer than 350 miles and with annual revenue less than $ 40 million. Switching & Terminal railroads perform pickup and delivery services for one or more connecting line- haul carriers. There were about 300 Line- haul and 200 Switching & Terminal railroads in 2002. Class I railroads dominate the industry in terms of freight revenue. Some 40,000 miles of track are owned and used by non- Class I railroads, however.
Maritime
Waterborne transportation was involved in the movement of about 9% of freight in the U. S. in 2004. This amounted to 2.4 billion tons of goods in 2003. Container ports handled over 65,000 TEUs ( 20- foot equivalent container units) per day in 2004. Most container units were involved in a form of intermodal exchange, either between ship and rail or ship and truck. Freight transport providers within the maritime sector can be divided into two major groups: ports, providing the infrastructure for freight movement and exchange, and the marine vessels that actually move the freight. Maritime differs from the other modes in that the freight infrastructure ( ports) may be owned by government or independent authorities ( in trucking, the infrastructure is generally government- owned, while in railroads, the infrastructure is private- 4
W. Cottrell Cal Poly Pomona
Figure 1. Class I Railroads ( BNSF = Burlington Northern & Santa Fe; CN/ GTW = Canadian National/ Grand Trunk Western; CP/ SOO = Canadian Pacific/ Soo Line; CSX = CSX Transportation; FXE = Ferrocarril Mexicano; KCS/ KCSM = Kansas City Southern; NS = Norfolk Southern; UP = Union Pacific). SOURCE = American Association of Railroads.
ly- owned). The port authorities are separate from the companies that operate the marine vessels. There are over 300 ports in the U. S. Table 1 lists the top 20 ports in the U. S. in 2005, by total annual shipment weight, and by total number of 20- foot equivalents ( i. e., number of container equivalents). Many of the busiest ports by total weight of shipments were along the Gulf Coast, where the primary goods were oil and petroleum products. The busiest container ports were at Los Angeles and Long Beach, with Oakland the fourth- busiest. California has 12 ports of various capacities levels of activities, three of which are among the five busiest ports in the U. S.
The U. S. Army Corps of Engineers maintains a 200+ page list of marine vessel companies ( WTLUS 2006), thereby suggesting the size and number of participants in the industry. These companies operate barges, cargo carriers, container ships, pushboats, tugboats, and other types of freight vessels. The five largest marine vessel companies in 2006, in terms of the number of vessels operated, included Ingram Barge Company ( 4,210 vessels; based in Nashville, Tennessee), American Commercial Lines ( 3,266; Jeffersonville, Missouri), American River Transportation Company ( 2,267; Ama, Louisiana), AEP Memco ( 1,770; Chesterfield, Missouri), and Kirby Inland Marine ( 1,090; Houston, Texas).
Pipelines
Pipelines in the U. S. carry energy commodities, including oil and petroleum products, and natural gas. A total of 868 billion ton- miles of oil and gas were moved by pipeline in the U. S. in 2003. The oil and gas pipeline networks are each divided into three functions: gathering, transportation or transmission, and dis-
5
W. Cottrell Cal Poly Pomona
Table 1. Top 20 U. S. Water Ports by Shipment Weight & Container TEUs: 2005
Port by shipment weight
Short tons ( millions)
Port by container TEUs
Full TEUs ( thousands)
South Louisiana, LA
212.2
Long Beach, CA
5,200
Houston, TX
211.7
Los Angeles, CA
4,375
New York, NY and NJ
152.1
New York, NY
3,581
Huntington- Tristate, WV- OH- PA
83.9
Oakland, CA
1,561
Long Beach, CA
79.9
Tacoma, WA
1,545
Beaumont, TX
78.9
Charleston, SC
1,514
Corpus Christi, TX
77.6
Savannah, GA
1,486
New Orleans, LA
65.9
Seattle, WA
1,443
Baton Rouge, LA
59.3
Norfolk, VA
1,436
Texas City, TX
57.8
Houston, TX
1,290
Mobile, AL
57.7
Honolulu, HI
856
Los Angeles, CA
54.9
Miami, FL
778
Lake Charles, LA
52.7
San Juan, PR
726
Tampa, FL
49.2
Port Everglades, FL
591
Plaquemines, LA, Port of
47.9
Jacksonville, FL
582
Duluth- Superior, MN and WI
44.7
Baltimore, MD
487
Valdez, AK
44.4
Anchorage, AK
293
Baltimore, MD
44.1
New Orleans, LA
177
Pittsburgh, PA
43.6
Wilmington, DE
162
Philadelphia, PA
39.4
Boston, MA
160
Total, top 20
1,558
28,241
Total, all ports
2,528
30,059
NOTE: TEU = 20- foot equivalent container. ( BTS, Pocket Guide to Transportation, 2008).
tribution. The transmission lines are the heart of these networks. According to a 2002 survey, 51 U. S. and 10 Canadian transmission lines carried 85% of the natural gas in North America. Pipeline & Gas Journal regularly ranks pipelines according to various aspects of performance. A summary of their year 2006 rankings leaders is provided in Table 2. The National Natural Gas Pipeline Network, as of 2000, is shown in Figure 1. As shown, there are heavy concentrations of pipelines in the Gulf Coast and Louisiana, western Oklahoma, and western Texas. Major gas lines extend into all of the continental United States.
Table 2. Leading Gas & Oil Pipeline Companies ( 2006)
Aspect of Performance
Company
Amount
Gas piping
Southern California Gas
95,603 miles
Gas sold
Public Service Electric & Gas
2,043,083 MMcf
Gas throughput
El Paso Natural Gas
5,641,319,000 Dth/ y
Gas operating revenues
Duke Energy Field Services
$ 12,335,000,000
Liquids ( oil) piping
Magellan Pipeline
8,583 miles
Crude oil deliveries
Enbridge Energy
553,528,000 Bbl
Liquids ( oil) operating revenues
Colonial Pipeline
$ 764,100,000
SOURCE: Pipeline & Gas Journal, Nov. 2007.
NOTE: MMcf = millions of cubic feet; Bbl = barrels; Dth/ y = ?
6
W. Cottrell Cal Poly Pomona
Figure 1. National Natural Gas Pipeline Network, 2000
The major trunklines of the U. S.’ crude oil pipeline network are shown in Figure 2. A large concentration of pipelines emanates from the ports adjacent the offshore drilling platforms along the Gulf Coast. Trunklines extend from the Texas coast to Illinois, Ohio and Indiana. One trunkline runs north- south through central California, extending from Los Angeles to the Bay Area. The network extends into Canada to incorporate several trans- border pipelines. The network of refined ( petroleum) products pipelines is shown in Figure 3. This network is separate from and in addition to the crude oil pipelines. A heavy concentration of the refined products pipelines is in the midwestern U. S., particularly Illinois, Indiana, Missouri, Iowa, Kansas, and a few other States.
Aviation
Aircraft were used to transport about 4% of the value and 1% of the tonnage of goods in the U. S. in 2002. A total of 37 billion ton- miles of goods were moved by aircraft in the U. S. in 2004. Despite the large value, aviation is a “ distant fifth” to the four primary freight modes – trucks, railroads, maritime and pipelines – in terms of goods movement in the U. S. Aircraft do, however, carry over 25% of the value of all U. S.- international merchandise. Similarly to maritime, aviation’s role in freight transport can be separated into two groups: airports and air carriers. Also, as with maritime, the airport and air carrier owners and operators are different.
7
W. Cottrell Cal Poly Pomona
Figure 2. Crude Oil Pipelines Network – Major Trunklines ( 2001)
Figure 3. Major Refined Products Pipelines ( 2001)
Despite aviation’s secondary role in freight movement in the U. S., an airport – John F. Kennedy International ( JFK) in New York City – is the U. S.’ busiest freight gateway. In 2004, more U. S.- international freight revenue ($ 125.3 billion) moved through JFK than through either the Port of Los Angeles or the Port of Long Beach ( which were the second and third busiest freight gateways). Other than JFK, Los Angeles International Airport ($ 68.7 billion), O’Hare International Airport in Chicago
8
W. Cottrell Cal Poly Pomona
($ 65.4 billion), San Francisco International Airport ($ 54.6 billion), and Dallas- Fort Worth Airport ($ 31.2 billion) were also busy air freight gateways. Miami International Airport claimed leadership in international freight tonnage ( 2.1 million) in 2007. Domestic air freight was dominated by Federal Express, United Parcel Service and DHL. Figure 4 shows the airports that were most used by these three parcel carriers. Memphis International Airport had the most domestic air freight activity in 2004, followed by Louisville International Airport and Indianapolis International Airport.
Figure 4. Major Air Parcel Hubs ( 2004)
Forster and Regan ( 2001) characterized the air freight industry as two organizational structures: integrators, who own all assets of production from shipper to consignee, and non- integrators, who forward, carry and deliver cargo. As of 2000, there were some 25,000 forwarders and 700 air cargo carriers worldwide, with 1,500 forwarders and 100 air carriers in the U. S. Leading integrators were FedEx and United Parcel Service ( UPS), while airlines such as Lufthansa and United were forwarders. The most active air freight carriers ( domestic) in the U. S. in 2004 were FedEx Express, which moved 8.984 million tonne- km of freight, followed by UPS ( 4.260 million tonne- km), Northwest Airlines ( 0.949 million tonne- km), China Southern ( 0.860 million tonne- km), and American Airlines ( 0.576 million tonne- km). Notice that the ranking blends integrators and non- integrators.
9
W. Cottrell Cal Poly Pomona
Other Modes
In addition to the primary freight modes of trucking, railroads, maritime and pipelines, as well as aviation, goods are also moved in the U. S. by bicycle couriers, foot messengers, cars, and motorcycles. Limited statistics are available on these “ other modes.” IBIS World reported that local delivery and messenger services were a $ 7.9 billion industry in 2007, with about 220,000 employees and about 177,000 business establishments. A total of 73.4% of the revenue were earned by pickup, van, and small car delivery services. The remaining 26.6% of the revenue were earned by bicycle and foot messengers, and motorcycle, moped, and small- box truck delivery services. Although this sector is a vital component of the freight transport industry, particularly in the central business districts of cities, the revenue generated by local delivery and messenger services represents less than 0.1% of the total U. S. freight transport industry revenue.
Yet another category is the “ virtual” shipping of documents by e- mail, fax and the internet. These electronic modes have not yet been incorporated into the scope of transportation engineering. That is, freight transportation statistics do not reflect “ electronic” goods, although the impact of these virtual modes of transport on the traditional modes has been discussed. De Jong et al. ( 2006), for example, predicted that freight transport would increase in the so- called “ e- economy.” Similarly, Smith et al. ( 2002) predicted rapid growth in “ e- business,” and the potential for freight transport to aid this growth. Although the indication is that freight transport activity will increase in the e- economy, the impact on local messengers and couriers has not been identified.
Summary
The diversity of the U. S. freight transportation industry is evident in the data shown in Table 3. Air, road, water, rail and pipeline modes are used to move goods. Trucking dominates the amount of freight revenue generated in the U. S., but railroads are competitive with trucks in terms of total shipment weight. Aviation makes only a minor contribution to domestic freight, but is a major player in international shipments. Performance measures and standards are diverse, to fit the needs of the various modes and categories within the modes, as well as the modal providers. A few performance measures are common to all modes, such as revenue and ton- mileage. A full understanding of the measures, measurement needs, and standards of freight transport providers requires an investigation of the separate modes and, to a certain extent, the providers themselves. The industry is competitive, primarily as a consequence of governmental deregulation of the various modes. It is a challenge to “ tap into” the data and statistics of the industry given the proprietary aspects. The Surface Transportation Board and the U. S. Department of Transportation require that certain statistics be reported, however, enabling an investigation of this information. This study focused on learning from the readily- available data. A more extensive – and expensive – investigation would involve industrial contacts, and possibly a survey, to acquire a fuller understanding of freight transport provider performance. The following sections of this report feature a review of the literature on freight transport performance measures, followed by a discussion of performance within each of the freight modes.
10
W. Cottrell Cal Poly Pomona
Table 3. Overview of U. S. Freight Transportation Industry
Industry- Wide
Mode
Category( s)
Leading Provider ( 2002- 2006)
Revenue1
Ton- Miles2
Aviation
Air carriers
Airports
FedEx Express
JFK ( international)
Memphis ( domestic)
$ 265.0
16,451
City streets
Local messengers
NA
$ 7.9
--
Highways
Trucks
Schneider National Carriers ( FTL)
United Parcel Service ( LTL)
$ 6,235.0
1,314,616
Maritime
Marine vessels
Ports
Ingram Barge Company
South Louisiana ( by weight)
Long Beach ( by TEUs)
$ 89.3
621,170
Pipelines
Pipelines
El Paso Natural Gas ( throughput)
Duke Energy Field ( gas revenue)
Enbridge Energy ( oil throughput)
Colonial Pipeline ( oil revenue)
$ 149.2
938,013
Railroads
Railroads
Union Pacific
$ 310.9
1,684,461
TOTAL
$ 8,397.2
4,574,711
1 Modal totals, in billions of dollars ( 2002). Air revenue includes truck connections.
2 Modal totals, in millions ( 2004).
SOURCE: BTS ( 2007).
FREIGHT PERFORMANCE MEASURES: LITERATURE REVIEW
Morash ( 2000) explained that there are five categories of freight performance measurement: asset management, cost, customer service, productivity, and quality. Further, there are four categories of freight stakeholders: the freight infrastructure provider ( e. g., transportation agency, port), the producer of goods ( i. e., manufacturer), the shipper, and the customer. Each category of measurement is subject to a different interpretation by each of the stakeholder groups; also, some of the categories may not be applicable to certain stakeholder groups. Thus, it is possible to develop a three- dimensional matrix with up to 20 cells, each containing a set of performance measures related to the given measurement category and stakeholder group. This research concentrates on “ freight transport providers;” i. e., the shippers, and the freight infrastructure providers. Hence, the scope of the investigation is limited to a proportion of the potentially 20 cells.
Overview of Literature Findings
The literature on freight performance measurement has become more active in recent years than in years past. For example, a review of several databases revealed as many articles published since 2000 as were published in previous years. The heightened activity is because, in part, of an increasing reliance of economies on the reliable movement of goods. From the pre- 2000 references, Boisjoly ( 1979) reported on 20 motor carrier performance measures, two of which were given special attention: revenue per ton- mile and the ratio of operating expenses to revenue ( the operating ratio). Miller ( 1990) concentrated on customer service, using five measures: request date, first acknowledgement, published interval, last acknowledgement, and last positive acknowledgement. Each measure was associated with a shipping timeline extending from pickup to dropoff, and final acknowledgement of a shipment’s receipt. Mentzer and Konrad ( 1991) listed a collection of efficiency and effectiveness performance measures in five categories: transportation, warehousing, inventory control, order processing, and logistics administration. Their measures are listed in Table 4. One of the authors’ arguments was that measures needed to address
11
W. Cottrell Cal Poly Pomona
both efficiency and effectiveness, rather than one or the other, partially because of the difficulty in defining “ 100% efficiency” or “ 100% effectiveness.”
Ferreira and Sugut ( 1992) identified three major performance measures for road- rail container transfer facilities: customer service, operational efficiency, and terminal productivity. The authors noted that an underlying constraint in a performance measure system would be the total capital and operating costs ( i. e., budget). Australia’s Bureau of Industry Economics ( 1992) suggested two types of indicators for the road freight industry: customer service and operational efficiency. Measures within each category were obtained from a survey of ( mostly) Australian freight transport providers. The report identified four key customer service measures: on- time pickup (% of pickups), on- time delivery (% of deliveries), loss and damage rate, and proportion of claims paid. Six operational efficiency measures emerged as the most common among the providers surveyed:
􀂃
total kilometers per vehicle per year
􀂃
total ton- kilometers per vehicle per year
􀂃
kilometers traveled empty as a proportion of total kilometers traveled
􀂃
average actual load as a proportion of full load capacity
􀂃
number of kilometers per driver per year
􀂃
fuel usage by vehicle type
Stewart ( 1995) discussed four “ keys” to unlocking “ supply chain excellence:” delivery performance, flexibility and responsiveness, logistics cost, and asset management. His suggested performance metrics were as follows:
􀂃
Delivery performance: % of orders fulfilled on or before the customer requested date; % of orders fulfilled on or before the original schedule or committed date.
􀂃
Flexibility and responsiveness: supply chain response time ( a sum of four components, including communications to end- product and feeder plants, product sourcing, and lead time).
􀂃
Logistics cost: order management cost; materials acquisition cost; inventory carrying cost; supply chain finance, planning and management information systems ( MIS) cost.
􀂃
Asset management: cash- to- cycle time (= total inventory days- of- supply + days- sales- outstanding – average- payment- period to suppliers).
Appfel, et al. ( 1996) described a methodology for determining freight terminal capacity. Two types of freight terminals were identified: flow processing components and stock holding components. Flow processors did not store cargo, and were involved only in transferring goods. Two measures were developed for the two terminal types:
􀂃
Dynamic capacity of flow ( tons per year) = effective transfer rate ( tons per day) * effective working time ( days per year)
􀂃
Dynamic capacity of stock component ( tons per year) = effective static capacity ( tons) * effective turnovers ( per year)
The above measures could be adapted by freight transport providers to their inventory control concerns. Lawrence, et al. ( 1997) categorized a broad spectrum of “ infrastructure industries” into four areas of performance: price, service, labor productivity, and capital productivity. All freight modes were considered, as well as several public utilities. The measures developed, all of which were supported with industry data, were:
􀂃
Price: average revenue per net ton- kilometer; waterfront charges per twenty- foot equivalent container ( TEU); waterfront charges per ton; standard dry bulk vessel operating costs; long- haul cents per ton- kilometer.
12
W. Cottrell Cal Poly Pomona
Table 4. Freight Logistics Performance Measures ( Mentzer and Konrad, 1991)
Transportation
Warehousing
Inventory Control
Order Processing
Logistics Admin
Vehicles loaded or
unloaded per labor
hour
Equivalent vehicles
loaded or unloaded
per labor hour
Activity per labor
hour
Activity per labor
hour
Activity per labor
hour
Weight loaded or
unloaded per labor
hour
Weight loaded or
unloaded per labor
hour
Cost savings per
labor hour
Activity per facility
per day
Activity cost
Miles driven
Lines, cases, orders
or units per labor
hour
Total order activity
Activity cost per unit
Driving hours
Dollar value per
labor hour
Accuracy
Miles per driving
hour
Weight unloaded per
dock door per day
Labor hours used
Labor hours
Labor hours
expended
Transit hours per trip
Equivalent vehicles
unloaded per dock
door per day
Cost
Cost
Cost
Cost
Total cost per unit
Weight, orders, lines
or units throughput
per labor hour
Total cost per unit
Total cost per unit
Equivalent cost of
outside substitute
Weight, units or
pallets throughput
per total warehouse
cost
Downtime
Downtime
Equipment downtime
Downtime
Equipment downtime
Equipment hours
Weight, units or
pallets throughput
per hour
Equipment hours
Equipment hours
Units per hour
Lines, units or
orders per square
foot
Units per hour
Units per hour
Units per hour
Fuel use per mile
Fuel use per tonmile
Fuel use per stop
Units, weight, lines,
orders or dollars
throughput per
square foot
Miles driven per
gallon
Transactions processed on time
Transit time
Replenishment cycle
time
􀂃
Service: claims for loss or damage in cents per $ 100 revenue; hours to move 600 boxes; percentage of late deliveries; percentage lost and damaged.
􀂃
Labor productivity: millions of net ton- kilometers per employee; TEUs per employee; thousands of tons per employee; manning level of small dry bulk vessels.
􀂃
Capital productivity: millions of net ton- kilometers per railcar; millions of net ton- kilometers per locomotive; throughput/ capacity (%); crane rate in moves per hour; thousands of tons per kilometer per year.
13
W. Cottrell Cal Poly Pomona
Some of the measures were extracted from the BIE ( 1992) report, discussed earlier. Stainer ( 1997) emphasized productivity measures as being, perhaps, the most meaningful indicators of logistics performance. He noted that productivity could be divided into three types, each with a different measurement scope:
􀂃
Partial productivity: ratio of total output to a single input, such as labor, materials or capital.
􀂃
Total factor or value- added productivity: total sales less bought- in goods, materials and services.
􀂃
Total productivity: ratio of total output to total input.
Duma ( 1999) argued that the ton- kilometer, although widely used, was not a powerful enough measure to differentiate between freight transport activities, or to characterize the importance of transport modes. Although the author did not recommend any measures, the following were suggested for consideration:
􀂃
Weight of transported goods
􀂃
Transport distance
􀂃
Transport tariff revenue
􀂃
Transported units
􀂃
Number of vehicles
􀂃
Capacity measurements ( no examples given)
􀂃
Operation time/ haulage time
􀂃
Fuel & energy consumption
􀂃
Utilization/ crowd indexes ( no definition provided)
􀂃
Artificial indexes ( to be defined by the user)
A Transportation Research Board conference ( TRB, 2001) brought together Federal Highway Administration ( FHWA) and State Department of Transportation officials to “ hash” out several performance measurement issues, including freight movement. Although no definitive measures were identified or recommended, a heavy emphasis was placed on the data needed to compute key measures. One conclusion is that any performance metric, for it to be useful, must be tractable. Also, a list of feasible metrics may represent the constraints of data availability. Gunasekeran, et al. ( 2001) identified multiple measures in four supply chain categories: plan, source, make- assemble, and delivery- customer. Their proposed measures are listed in Table 5.
Table 5. Supply Chain Performance Metrics ( Gunasekeran, et al., 2001)
Plan
Source
Make- Assemble
Delivery- Customer
Total order cycle time
Customer order path
Level of supply chain
partnership
Capacity utilization
Effectiveness of
scheduling techniques
Productivity of human
resources
Actual vs. planned
throughput
Inventory levels
Manufacturing cost
Delivery- to- request date
Delivery- to- commit date
Order fill lead time
Percentage of goods in
transit
Number of faultless
notes invoiced
Flexibility of delivery
systems
Logistics distribution cost
Customer query time
Customer perception of
service
Total logistics cost
Total cash flow time
Total inventory cost
14
W. Cottrell Cal Poly Pomona
Lai, et al. ( 2002) separated the supply chain process into two segments: “ customer facing” ( i. e., customer service) and “ internal facing” ( i. e., operations). The primary concerns of customer facing were identified as reliability, flexibility and responsiveness, while the main concerns of internal facing were costs and assets. Their suggested performance measures were:
􀂃
Customer facing: delivery performance, order fulfillment performance, and perfect order fulfillment ( reliability); supply chain response time and production flexibility ( flexibility and responsiveness).
􀂃
Internal facing: total logistics management costs, value- added productivity, and return processing cost ( costs); cash- to- cash cycle time, inventory days of supply, and asset turns ( assets).
Lai, et al. ( 2004) extended this discussion by adding measures related to shippers’ needs, as well as the needs of consignees. Holguin- Veras, et al. ( 2004) developed an experimental economics approach to urban goods modeling. To evaluate their model, the following measures were used: number of tours required to meet freight needs, total profits, total number of stops, profits per tour- hour, profit per tour per unit freight, and profit per tour- hour per unit freight. A “ tour” included the travel, loading and unloading time of a pickup and delivery.
Finally, Jones & Sedor ( 2006) summarized the efforts of the FHWA to facilitate the development of reliability measures for freight travel. The authors pointed out the Department of Transportation’s recognition that the “ timely and reliable movement of freight is critical to the Nation’s economy.” Hence, the FHWA effort concentrated on reliability. The following measures were proposed: fill rate, delay, travel time, travel time reliability ( speed & buffer time index), profitability, and return on investment. The latter two measures did not pertain to reliability per se, but recognized the importance of solvency to the freight industry. Fill rate was defined as the percentage of orders delivered on time ( i. e., no later than the delivery day requested by the customer).
INDUSTRY PERFORMANCE MEASURES AND CONCERNS
A key distinction between the performance measures “ suggested” in the literature, and those actually applied in practice, is the availability of data to compute the measure. Another distinction is the performance measure that can be “ influenced by the public sector,” and the measure that is “ meaningful to stakeholders in the private sector” ( Jones and Sedor 2006). Performance measurement experts have also noted that the measures of interest depend on the role ( i. e., users, shippers, carriers, authorities) and the geographic scale. The FHWA has ascertained that speed of travel and travel time reliability are two measures that are of interest to both the private and public sectors, particularly for highway- based modes. Several research efforts have addressed these two measures, as well as the technology needed to track the location of trucks; the vehicle location technology is needed for the compilation of travel speeds and times. Another factor is the extent to which the measure addresses a critical industry issue. The following discussion reviews freight performance measures, by mode, that are found in readily- available publications, or that are implied by discussions in industry- related documents.
Commercial Trucking and Multimodal
The use of performance measures in the trucking industry, and perhaps in all freight modes, is vast and extensive. For example, USA Truck, an FTL carrier, indicated that their annual self- assessment involved the use of performance measures in “ 300 statistical areas.” Some performance measures are common to many carriers – regardless of mode – while others are common to carriers within a specific mode. Still other performance measures are customized to one or a few carriers, although many of these are derivatives of a common base ( such as “ revenue” or “ load”). A review of one FTL carrier ( USA Truck), one LTL carrier ( US Xpress), and one carrier offering both FTL and LTL services ( Frozen Food Express)
15
W. Cottrell Cal Poly Pomona
revealed that the greatest proportion of performance measures was financial. Other performance measurement categories were either equipment- or load and haul- related. Notably, none of the carriers included a congestion- or speed- based measure, although the effects of any shipping delays would ultimately be reflected in financial statements. Three measures were emphasized as being “ extremely important” to the industry:
􀂃
Average length of haul
􀂃
Empty miles factor
􀂃
Operating margin or ratio
The average length of haul is self- explanatory. One definition of the empty miles factor is the total number of miles traveled between loads as a percentage of the total miles traveled. The operating margin or ratio is simply a company’s operating expenses divided by the operating revenue. Other performance measures being used, in the financial, equipment and load categories, are listed in Table 6. Performance standards vary by carrier; this research did not investigate the different possible criteria.
Table 6. Performance Measures Used by Commercial Vehicle Operators
Financial ( annual or year- end)
Equipment
Load and Haul
% of revenue from S& P 500
Average age of revenue equipment
Average length of haul
% of revenue from top customers
Equipment utilization rate
Empty miles factor
After- tax return on equity
Tractor operating life
Freight volume
Annual revenue growth rate
Trailers in service ( trailer fleet)
Hundredweight
Average shares outstanding
Trailer operating life
Loaded miles per load
Book value per share
Tractors in service ( tractor fleet)
Loaded miles
Claims costs
Pounds per shipment
Debt
Revenue per business day
Debt- to- equity ratio
Revenue per hundredweight
Earnings per share
Revenue per loaded mile
FTL or LTL revenue
Revenue per shipment
FTL/ LTL % of revenue
Shipments
Insurance costs
Shipments per business day
Internal rate( s) of return
Market value per share
Net capital expenses
Net income or loss
Operating expenses
Operating margin or ratio
Pre- tax margin
Return on capital
Revenue
Shareholders’ equity
Working capital
SOURCES: Measures used by Frozen Food Express, US Xpress, and USA Truck.
Other measures in use were customer- oriented, including the number of customers, the proportion of all customers who were returnees, customer duration or dedication, and average collection time ( i. e., time period from billing to receipt of payment). Still other measures, not fitting into any of the above categories, included the number of employees, the number of drivers, and the maximum tractor speed. The latter measure was incorporated into one carrier’s safety measures. That is, a limit on tractor speed
16
W. Cottrell Cal Poly Pomona
was being used as a mitigation against excessive highway speed, thus serving as a potential crash prevention action.
The commercial carriers studied tended to address safety in terms of claims costs, liability, and mitigating factors, rather than with crash- or incident- related measures. Claims and tort were, perhaps, consequential measures of incidents, whether vehicle- or load- related. It can be argued that crashes and incidents, along with crash and incident rates, are the most direct measures of safety; it can also be argued, though, that these measures are best recorded on a broad scale ( e. g., statewide or nationwide), rather than on a per carrier basis.
For another perspective on performance among commercial carriers, the American Transportation Research Institute identified the trucking industry’s critical issues in 2007, based on a survey of trucking companies. The issues point toward performance measures that might be used to assess how well the needs of the trucking industry are being addressed. The critical issues can be grouped into eight subject areas:
•
Hours of service regulations
•
Driver availability and shortages
•
Fuel costs
•
Highway congestion
•
Toll costs
•
Tort and other liability matters
•
Environmental controls
•
On- board technology
In general, the trucking industry is concerned with heightened regulations that restrict operational flexibility, rising costs ( in several areas), the costs of satisfying regulations, the effects of congestion, and improving safety ( an outcome of which is tort and associated legal matters). Each of these issues suggests one or more performance measures, as follows:
Hours of service. Fundamental performance measures are the hours of service per driver, and the total hours of service per selected time period, such as a week, month or year. Driver- related measures, determined as an average per driver per selected time period, are the hours of sleep, on- duty hours, and off- duty hours. Other example time periods might be the 60- or 70- hour periods specified in the hours of service legislation.
Driver Availability and Shortages. A rising concern in the trucking industry is driver turnover – annual rates reportedly approach 100% for some carriers. Relevant performance measures include the annual driver turnover rate, driver retention rate, and annual driver recruitment and training costs. Measures related to driver workload include the miles per driver per day, tour length, average circuity per load, and first dispatch empty miles. A “ circuity” is a ( presumably roundabout) tour; the first dispatch empty miles represent the numerator of the empty miles factor as applied to the first pickup or delivery of the day.
Fuel Costs. The costs of fuel are typically incorporated into a carrier’s operating expenses. Rising fuel costs demand a separate consideration, however. Fundamental measures would include the average amount paid per gallon of diesel fuel, the total annual fuel expenses, and fuel efficiency ( i. e., average miles per gallon). Similar measures for gasoline or other types of fuel may be applicable for certain truck companies having non- diesel vehicles.
Highway Congestion. As noted above, the FHWA has identified the average speed of travel and travel time reliability as two critical freight performance measures. Many carriers use 47 mph as the average speed at which freight will be transported, regardless of actual traffic conditions. Table 7 shows the car
17
W. Cottrell Cal Poly Pomona
and truck freeway speed limits in States in which there is a difference. California has the greatest car- truck speed difference ( 15 mph), although Alabama has the same differential for hazmat trucks. Actual travel speeds vary according to levels of congestion, the time of day of travel, terrain and topography, truck size and load, and truck driver behavior. Peak period congestion can be avoided with travel during non- peak periods, but scheduling and hours of service needs may necessitate travel during the peak. The diagram in Figure 5 shows freeway speeds along the I- 10 freeway westbound in the Los Angeles area during the morning peak on a weekday in September 2000. It is evident that the 47 mph assumption would not be applicable to travel along this route at this time of day. The indication is that a more flexible measure of travel speed should be used. One recommendation would be to use different peak and non- peak speeds, along with speeds that are reflective of different geographical areas. Regarding the latter, Table 8 shows year 2005 peak period freeway speeds ( calculated, not empirical) in selected very large, large, medium and small urban areas in the U. S. Of the 85 urban areas listed, 23 had freeway speeds less than 47 mph, with the lowest being San Francisco- Oakland ( 39.4 mph), Chicago ( 39.1 mph), and Los Angeles- Long Beach- Santa Ana ( 34.7 mph). Since these data are compiled annually as part of the Urban Mobility Report series ( e. g., Schrank and Lomax 2007), this is accessible information that could rather easily be adopted by trucking companies.
Table 7. Truck Speed Limit Differentials
State
Statutory car speed limit
Statutory truck speed limit
Alabama
70
55 ( hazmat only)
Arkansas
70
65
California
70
55
Idaho
75
65
Illinois
65
55
Indiana
70
65
Michigan
70
60
Montana
75
65
Ohio
65
65 on Ohio Turnpike, 55 on all other freeways.
Oregon
70* never implemented, 65 or less still in effect[ 96]
5 mph differential, effectively 60[ 96] although 55 is still posted in most locations
Texas
70- 80 mph day/ 65 mph night
70/ 65 night
Texas ( Farm- to- Market roads only)
70 mph day/ 65 mph night
60 day/ 55 night
Washington
70
60
SOURCE: http:// en. wikipedia. org/ Wiki/ Speed_ limits_ in_ the_ United_ States. < accessed on Sep. 17, 2008>
Toll Costs. Trucking companies have expressed concern about the effects of an increasing number of toll facilities on overall operating expenses. Toll roads have existed for some time in the eastern U. S., but there is a growing number of toll facilities in the western U. S. Fundamental measures might include the proportion of operating expenses devoted to tolls, the total toll costs paid, total toll road mileage, and total toll road mileage as a percent of total mileage.
18
W. Cottrell Cal Poly Pomona
Figure 5. I- 10 Westbound Speed Profile, 7: 30 am, 9/ 14/ 2000 ( PeMS data, Univ. of California, Berkeley)
Tort and Liability, Environmental Mitigation, and On- Board Technology. Performance measures associated with these issues were not found in any documentation. Since these were identified as critical issues, the development of performance measures is warranted. Further research, beyond the scope of this study, might identify appropriate measures.
Railroads
As discussed above, the railroad industry is dominated by the Class I railroads, which earned 92% of all freight rail revenue in 2002. As such, the industry performance measures reported are dominated by those pertaining to Class I Railroads. The American Association of Railroads ( AAR) serves as a clearinghouse for industry statistics. Statistics that also serve as performance measures include the following. Note that average length of haul and operating ratio, along with revenue and expenses measures, are used by other modes:
•
Average length of haul
•
Average tons per carload
•
Average tons per train
•
Carloads originated
•
Containers transported
•
Employees
•
Freight cars in service
•
Freight revenue
•
Freight revenue per ton- mile
•
Locomotives in service
•
Net income
•
Operating expense
•
Operating ratio
•
Operating revenue
•
Railroad market share
•
Return on average equity
•
Ton- miles of freight
19
W. Cottrell Cal Poly Pomona
Table 8. Estimated Average Freeway and Arterial Speeds in Urban Areas, 2005
SOURCE: Schrank, D. and T. Lomax, Urban Mobility Report 2007 ( speeds are in mph).
•
Tons originated ( by commodity)
•
Tons originated ( total)
•
Trailers transported
Individual railroads regularly report three performance measures to the AAR:
•
Railcars on line
•
Terminal dwell time
•
Train speed
20
W. Cottrell Cal Poly Pomona
“ Railcars on line” is the average daily online inventory of freight railcars. Terminal dwell time is the average time a railcar resides at a specified terminal. The train speed is calculated by dividing the train- miles by the total operating time, excluding terminal time. Train speed represents line- haul movement between terminals.
All railroads, including freight and passenger, report their incident data to the Federal Railroad Administration ( FRA). Year 2005 freight railroad safety data are summarized in Table 9. Note that the FRA reports incident data by railroad for Class I and Regional operators; such detailed reporting is not done in the motor carrier industry. Safety statistics, which could also serve as performance measures for the individual railroads, include those listed in the table. Incident rates can be used to compare railroads.
Table 9. Safety Statistics for Selected Class I Railroads, 2005
Total
Hwy.- Rail Grade Xings
Railroad
Fatalities
Incident Rate
Fatalities
Incident Rate
Employee Injuries Rate
Yard Incidents Rate
Damaged Consists
Hazmat
Releases
BNSF
153
3.41
70
2.36
1.53
25.68
128
10
CSX
112
4.39
47
4.67
1.73
24.25
53
5
KCS
13
14.07
7
12.36
2.23
53.50
21
0
Norfolk
NA
NA
55
4.84
1.21
14.66
22
3
UP
167
4.77
63
2.81
2.10
34.84
95
12
NOTES: BNSF = Burlington Northern & Santa Fe; CSX = CSX Transportation; KCS = Kansas City Southern; Norfolk = Norfolk Southern; UP = Union Pacific; Total = all incidents; Incident rates are per million train- miles; Employee injuries rate is per 200,000 ( work) hours; Yard incidents rate is per million yard switching train- miles; NA = not available.
Ports and Ships
Chung ( 1993) noted that the primary performance indicators used by ports are the vessel turnaround time, and the tonnage handled per ship day in port. The vessel turnaround time is the length of stay from time of arrival to time of departure. A variation of turnaround time is dwell time, which is the number of days that a ton of cargo ( as opposed to a vessel) remains in port. Chung also noted that port productivity is measured by tons per gang hour, and TEUs per crane or hook hour. The former measure applies to general, non- container cargo, in which a work station is referred to as a “ gang.” The latter measure applies to containers – cranes and hooks are the equipment used to move and place containers. Ports are also concerned with financial performance, such as operating surpluses, operating expenses, possibly as ratios to the tonnage of cargo handled. Total TEUs, total tonnage, trade values ( total and by type of cargo), commodity values and volumes, and market shares are also used by the ports to establish benchmarks.
The Maritime Administration ( MARAD) maintains a statistical database on the U. S.’ 300 or so ports. Each of the statistics, also reflective of performance, measures the level of activity at each port. The port rankings in Table 1 use two of these measures: annual shipment weight and annual TEUs. The other annual measures used include:
•
Total calls ( all vessels)
•
Total capacity of all calls ( total metric tons of all ships loaded to water line)
•
Total tanker vessel calls and capacity
•
Total product vessel calls and capacity
•
Total crude oil vessel calls and capacity
•
Total container ship calls and capacity ( in TEUs)
•
Total dry bulk cargo vessel calls and capacity
•
Total ro- ro ( roll- on roll- off container) vessel calls and capacity 21
W. Cottrell Cal Poly Pomona
•
Total motor vehicle vessel calls and capacity
•
Total gasoline carrier vessel calls and capacity ( in cubic meters)
•
Total combination vessel calls and capacity
•
Total general cargo vessel calls and capacity
Marine vessel operators were using performance measures that reflected volume, solvency, efficiency, and safety. Kirby Inland Marine ( Holcomb 2004), for example, reported on:
•
Delay days ( i. e., total delay, measured in days)
•
Revenue per ton- mile
•
Ton- miles
•
Towboats operated
American Commercial Lines reports on its financial and business objectives in its financial statements, as required by the U. S. Securities and Exchange Commission. The performance measures used to determine whether the objectives had been met, in their April 2008 report, were:
•
Earnings before interest, taxes, depreciation and amortization ( EBITDA)
•
Earnings per share ( net income divided by outstanding shares)
•
Average working capital as a percent of revenue
•
Environmental responsibility
•
Safety incident rate ( recordable injuries * 200,000 divided by number of employee- hours worked)
•
Sales, general and administrative expenses as a percent of revenue
•
Stationary days reduction for covered hoppers
•
Turn rate per 10,000 liquid barges
The stationary days reduction for covered hoppers is analogous to Kirby’s delay days, except that the measure concentrates on the amount by which delay was reduced in comparison to a previous period. Two environmental responsibility measures were used: the number of notices of violation received from a State agency, and the number of releases ( spills) entering into a river. The turn rate is the number of days in the year divided by the average number days required to move a lift on a 10,000 liquid barrel barge. It is interesting to note that the commonly used motor carrier and railroad performance measures of average length of haul and operating ratio are not specifically used by marine vessel companies.
Pipelines
All pipeline operators track their financial performance in terms of standard measures ( revenues, expenses, earnings, etc.). The Pipelines and Hazardous Materials Safety Administration ( PHMSA) requires all owners of gas transmission pipelines to report on four overall measures of integrity ( each of which is related to pipeline safety) ( GAO 2006):
•
Pipeline miles inspected
•
Number of immediate repairs
•
Number of scheduled repairs
•
Number of leaks, failures and incidents
Mastio and Company, an independent, private group, annually performs customer satisfaction surveys of the gas industry ( Mastio 2008). Survey participants include local distribution companies, power producers, marketers, and various industrial users. The 2008 survey results were used to develop 29 22
W. Cottrell Cal Poly Pomona
attributes of natural gas pipelines; several of the attributes could be reformulated as performance measures, as follows:
•
Accuracy of contracts
•
Accuracy of scheduled gas volumes
•
Attitude of continuous improvement
•
Customer service orientation of company representatives
•
Ease of contacting right person ( accessibility and responsiveness of personnel)
•
Effectiveness of after- hours support
•
Execution of transportation requests
•
Expertise of personnel
•
Reliability of gas transportation
•
Timeliness of notification prior to restrictions
•
Timeliness of problem resolution
Several of the measures are, evidently, qualitative and based on opinions regarding customer service. This is a consequence of the role that many natural gas suppliers fulfill as public utilities. It is interesting to note that the Mastio surveys segregate natural gas pipelines into five categories, generally based on size or geography: mega, major, regional, intrastate, and major organizational groups. Mega pipelines had at least 3,500 miles of pipe and deliveries of at least 1 trillion cubic feet. Major pipelines had at least 3,500 miles of pipe and served at least three States. Regional pipelines had at least 3,500 miles of pipe, but served fewer than three States. Intrastate pipelines served one State, while major organizational groups were owner- operators. PG& E, for example, was in both the intrastate and major pipeline categories.
Similarly to natural gas pipelines, the PHMSA has identified several safety- related performance measures for oil or “ liquids” pipelines. These are, annually:
•
Hazardous liquid pipeline spills in unusually sensitive areas ( USAs)
•
Number of serious incidents
•
Pipeline incidents caused by corrosion
•
Pipeline incidents caused by excavation damage
•
Time required to close a Corrective Action Order after a safety sensitive incident
•
Unrecovered oil spill costs per costs for implementing IMP ( Integrity Management Programs) in USAs
Although the PHMSA requires liquids pipeline companies to report their safety- related incidents, only overall pipeline industry safety data are published. This approach is analogous to that in the motor carrier industry. That is, incidents are published for the entire industry as a whole, rather than on a per- carrier basis. In addition to the safety data, liquids pipeline companies such as Magellan and Enbridge Energy reported the following operating statistics to the U. S. Securities and Exchange Commission ( SEC):
•
Average barrels per day
•
Average haul ( miles)
•
Barrel- miles
•
Inland terminal throughput ( in millions of barrels)
•
Marine terminal average storage capacity utilized per month
•
Transportation barrels shipped
•
Transportation revenue per barrel shipped
23
W. Cottrell Cal Poly Pomona
Magellan, Enbridge Energy, and other companies also report standard financial measures to the SEC, including net income, earnings per unit, and EBITDA.
Aviation
An air cargo excellence survey is regularly conducted by Air Cargo World magazine, rating airports and air freight carriers. Airports are rated in four areas: performance, value, facilities, and regulatory operations. The performance measures include:
􀂃
Amount of allied services ( ground handling, trucking)
􀂃
Dependability
􀂃
Extent to which promises are fulfilled and contractual obligations are met
􀂃
Promptness and courtesy of customer service
Although the other three areas are not labeled “ performance,” aspects of performance are represented in all three. For example, value measures include competitiveness of rates, reasonableness of rates, and availability of value- added programs. Facilities measures refer to the size and adequacy of infrastructure, such as aprons, warehousing, accommodation of perishables, access to highways, and access to other modes of transport. Regulatory measures focus on the adequacy and integrity of customs, security, and a foreign trade zone. Airports are grouped into three categories: 1 million or more tons of freight annually, 500,000 to 999,999 tons, and less than 500,000 tons. Each airport is scored in each of the four areas by survey participants, each of whom rates the airport on a scale of one to five. The participants’ scores are summed and normalized ( within each of the three airport groups), such that the average air cargo terminal has a value of 100. Thereby, airports with a score of greater than 100 are above average. Table 10 lists the above average airports. The Memphis, Tennessee and Seoul Incheon ( South Korea) airports were the highest- rated in the one million and more tons of cargo category, with scores of 114; Osaka was highest in the 500,000 to 999,999 tons group, with Dallas- Ft. Worth the highest in the U. S. in this group; Nagoya was highest in the less than 500,000 tons group, with Houston Intercontinental the highest in the U. S. in this group. A total of 51 airports worldwide were “ above average,” 17 of which were in the U. S.
The air cargo excellence survey also rates air carriers, in five areas: customer service, performance, value, and information technology. Each airline is scored in each of the five areas in a similar manner as the airports, as explained above. Areas of performance addressed in the survey include:
􀂃
Accomplishment of scheduled transit times
􀂃
Dependability
􀂃
Extent to which promises are fulfilled and contractual obligations are met
Areas of customer service include expedience with which claims are handled, promptness and courtesy with which problems are solved, and professionalism and knowledge of sales force. Value measures are similar to those used to evaluate airports. Finally, information technology measures include the efficiency and capabilities of the air cargo carrier in tracking and tracing shipments, Internet ordering and processing, and electronic commerce. Air cargo carriers with a score of greater than 100 are above average. The world’s “ top 50 airlines” in the 2008 survey are shown in Table A2, in the Appendix. Lufthansa was the world leader, with an overall score of 124, and top scores of 131 in information technology and 125 in customer service. Swiss WorldCargo led in the performance category, with a score of 124, while Emirates SkyCargo led in the value category, with a score of 119. The top U. S.- based air cargo carrier was FedEx, with an overall score of 115.
24
W. Cottrell Cal Poly Pomona
Table 10. Above Average Air Cargo Terminals ( Air Cargo World, “ Air Cargo Excellence Survey,” 2008)
Airport
Performance
Value
Facilities
Regulatory
Overall
> 1 million tons annually
Memphis
116
115
115
111
114
Seoul Incheon
113
113
116
115
114
Amsterdam
107
109
108
112
109
Frankfurt
109
106
108
109
108
Anchorage
107
107
105
108
107
Hong Kong
106
105
105
103
105
Tokyo Narita
105
105
102
102
104
Louisville
100
104
103
102
102
Dubai
99
101
106
103
102
Chicago O’Hare
100
102
102
98
101
Singapore
99
99
104
101
101
500,000- 999,999 tons annually
Osaka
110
109
108
110
109
Shenzen
106
110
106
108
108
Dallas- Ft. Worth
110
106
109
105
107
Cologne- Bonn
107
104
111
105
107
Atlanta
104
104
105
104
104
Tokyo Haneda
106
98
100
99
101
< 500,000 tons annually
Nagoya
120
115
115
121
118
Munich
117
114
116
118
116
London Gatwick
110
114
114
113
113
Houston Intercontinental
112
113
112
112
112
Zurich
115
107
108
114
111
Vancouver
108
112
112
107
110
Helsinki
109
107
109
112
109
Frankfurt- Hahn
108
109
104
109
108
Abu Dhabi
110
102
106
106
106
Montreal Trudeau
106
105
104
104
105
London Stansted
105
106
104
106
105
Stockholm
104
106
107
102
105
Santiago
106
109
101
105
105
Seattle- Tacoma
103
105
107
103
104
Orlando
103
101
106
105
104
Copenhagen
103
107
108
100
104
Denver
103
102
103
103
103
Manchester
107
106
101
100
103
East Midlands
103
107
101
102
103
Liege
100
102
100
109
103
Salt Lake City
105
103
104
98
102
Ft. Worth Alliance
102
99
104
105
102
Detroit
103
101
102
103
102
Buenos Aires
104
98
103
103
102
Bahrain
105
101
102
100
102
Shanghai Honqiao
101
101
101
104
102
Minneapolis- St. Paul
104
103
100
98
101
Portland
104
99
99
101
101
Phoenix
99
100
104
100
101
Baltimore- Washington
98
103
98
103
101
Athens
103
95
100
108
101
New Delhi
101
102
98
103
101
Macau
99
101
101
103
101
Sharjah
97
99
106
102
101
25
W. Cottrell Cal Poly Pomona
OTHER MODES
The performance measurement practices of local messenger and small delivery services were not identified during the research conducted for this study. Only one publication, by IBISWorld ( an industry think tank), addressed this sector of the freight transport industry. The IBISWorld measures focused on industry performance, and were generally financial. A few measures that might be specific to the sector were discussed in the report, including:
􀂃
Disposable income levels in market service area
􀂃
Number of households in service area
􀂃
Number of businesses in service area
􀂃
Reliability of delivery ( particularly critical in this sector)
􀂃
Speed of delivery ( also critical in this sector)
􀂃
Accuracy of delivery ( also critical; the concern is with correct- address delivery)
􀂃
Internet connections ( a measure of competition with traditional messengers and delivery)
CONCLUSION
Performance measurement in freight transportation is practiced on a broad scale. No uniform guidance on performance measures in freight transport exists, so the measures and applications used in practice lack uniformity. This issue is currently being addressed, in part, by the NCFRP initiative described in the opening section of this report. One finding is that the performance measures used by freight transport providers are not the same ones in use by or of interest to the public sector. A fundamental reason for the different interests is that the industries surrounding the various freight transportation modes were deregulated some time ago, enabling open competition. Although the Surface Transportation Board, Federal Maritime Commission, Federal Aviation Administration, Federal Motor Carriers Safety Administration, Pipelines and Hazardous Materials Safety Administration, and other federal and State agencies regulate some aspects of the freight transport modes, the operators are generally free to function as businesses in a market. Thus, many of the performance measures used by the providers are financial and customer service oriented. Load and haul measures are also commonly used. Employee- related measures are used, as well, although there is little similarity in their forms. Travel speed measures are used, although some providers seem to use financial performance as a consequence of delay, rather than direct measures of delay. Nonetheless, the FHWA has identified average travel speed and travel time reliability as the two key freight transport performance measures. It is anticipated that these measures will “ catch on” in the industry, in all of the modes. Now that the measures have been identified, the FHWA has diverted its attention to the data collection technology needed to develop the speed and reliability measures. It is likely that the technologies will need further development before there is widespread implementation of speed and reliability measures.
A second finding is that there are modal differences that, by necessity, require different performance measures. For example, ports keep track of marine vessels that are specific to the type of cargo transported; thus, performance measures such as crude oil tanker calls and dry bulk cargo vessel capacity are used. In contrast, the trucking industry does not similarly record cargo by truck type – the emphasis is toward weight, distance and value. A third finding is that a handful of measures are used by many, if not most, of the providers in nearly all freight transport modes:
􀂃
Average length of haul
􀂃
Operating ratio
􀂃
Revenue per ton- mile
􀂃
Tonnage ( total, all loads)
􀂃
Ton- miles or barrel- miles
􀂃
Terminal dwell time or empty miles factor
26
W. Cottrell Cal Poly Pomona
Belman and White ( 2005) characterized the average length of haul as a measure of productivity, noting that there had been increases in several commercial vehicle sectors. Changes in this measure, according to these authors, could be related to load consolidation, route optimization, and- or the use of dedicated fleets. The measure can be negatively affected by congestion, particularly if a time constraint is associated with a haul. The operating ratio is one of the simplest measures of financial performance. Interestingly, many freight transport providers – particularly the motor carriers – function at very high operating ratios ( i. e., revenues and expenses are nearly equal). Railroad operating ratios tend to be lower, around 80%, and may suggest a potential for greater profitability. Given that how an operator defines revenue and expenses, or what is included in these, can vary, it is difficult to pinpoint an exact interpretation of the operating ratio in each individual case. The consensus, however, is that a high operating ratio is suggestive of an inability to cover costs.
Revenue per ton- mile, tonnage and ton- miles ( or barrel- miles) are all load- and haul- related measures. Ton- miles, in particular, are tabulated across all modes on national and international scales, and are used as key benchmark of freight movement activity. The four main freight modes – trucks, railroads, ships and pipelines – are surprisingly well balanced in the U. S. in terms of their proportions of total ton- mileage. It is interesting that the balance exists amidst deregulated industries, although it is not clear if deregulation precipitated the balance, or if there is a gradually evolving shift toward certain freight transport modes. Ton- miles, as a measure, has been criticized for not distinguishing between the affects of productivity and consist ( vehicle or vessel size) changes ( Levine 1985). The measure affects goods movement policy decisions despite this flaw ( Gerondeau 1996). Ideally policy and corporate decisions should also consider other performance measures.
Finally, terminal dwell time and the empty miles factor are measures of “ unproductive time.” Freight transport providers probably try to reduce these measures, to improve efficiency and productivity. If containers are used, then the manner in which they are stacked and stored becomes a function of dwell time ( Huynh 2007). Vachal and Bitzan ( 2005) noted that dwell time was positively correlated with transportation price, in part because of the negative effect of dwell time on system capacity. The empty miles factor is an alternative “ version” of dwell time, in that it is a measure of empty vessel movement. The measure is used primarily by motor carriers, perhaps because the other modes have greater control over the movement of empty vessels. Jordan and Burns ( 1984) noted that the empty miles factor could be reduced by backhauling ( carrying a load from the destination back to the origin). Ultimately, however, the empty miles factor might be minimized by optimizing terminal, plant and distribution center locations, relative to supplier selection decisions. It is not clear to what extent motor carriers attempt to optimize these aspects of their businesses.
This study is merely an overview and general assessment of current performance measurement practices in freight transportation. It is anticipated that the findings of NCFRP Project 03 will point toward the establishment of performance measures that are uniform and intermodal.
REFERENCES
Air Cargo World, “ Air Cargo World’s Air Cargo Excellence Survey,” 4th edition, Mar. 2008.
American Association of Railroads, Class I Railroad Statistics, Jul. 17, 2008, www. aar. org/~/ media/ AAR. < accessed on Sep. 9, 2008>
American Commercial Lines, 8- K Report, Quarterly filing with U. S. Securities and Exchange Commission ( Apr. 2, 2008 used), www. secinfo. com/ dsvRm. 1451. d. htm. < Accessed on Sep. 20, 2008>
American Transportation Research Institute ( ATRI), Critical Issues in the Trucking Industry – 2007, Arlington, VA, Oct. 2007.
27
W. Cottrell Cal Poly Pomona
American Trucking Association, STATS: Standard Trucking and Transportation Statistics, Vol. 13, Issue 1, 2006.
Appfel, Charles, Jay Jayawardana, Asaf Ashar, Kevin Horn, Robert McLaughlin, and Anatoly Hochstein, “ Freight Components in Louisiana’s Statewide Intermodal Transportation Plan,” Transportation Research Record 1552, Transportation Research Board, National Research Council, Washington, DC, 1996, pp. 32- 41.
Belman, Dale and Chelsea C. White, Trucking in the Age of Information, Ashgate Publishing, Hampshire, UK, Oct. 2005.
Boisjoly, Russell P., “ Effect of Mergers on Motor Carrier Performance,” Minutes of the Meeting: Pennsylvania Electric Association, Engineering Section, New Orleans, LA, Nov. 19- 21, 1979, Vol. 1, pp. 82- 84.
Bureau of Industry Economics ( BIE), International Performance Indicators: Road Freight, Research Report 46, AGPS, Canberra, Australia, Apr. 1992.
Bureau of Transportation Statistics ( BTS), Freight in America: A New National Picture, Research and Innovative Technology Admin., U. S. Dept. of Transportation, Washington, DC, Jan. 2006.
Bureau of Transportation Statistics ( BTS), National Transportation Statistics 2006, Research and Innovative Technology Admin., U. S. Dept. of Transportation, Washington, DC, 2007.
Bureau of Transportation Statistics ( BTS), Pocket Guide to Transportation, Research and Innovative Technology Admin., U. S. Dept. of Transportation, Washington, DC, Feb. 2008.
Chung, Kek Choo, “ Port Performance Indicators,” Infrastructure Notes, Transport No. PS- 6, Transportation, Water and Urban Development Dept., The World Bank, Dec. 1993.
Commercial Carrier Journal, “ The CCJ Top 250,” Aug. 2007.
De Jong, Gerard, Staffan Algers, Andrea Papola, and Robert Burg, “ Impact of E- economy on Traffic and Traffic- Related Indicators in Urban Areas,” Journal of Transportation Research Board: Transportation Research Record 1977, National Research Council, Washington, DC, 2006, pp. 286- 291.
Duma, Laszlo, “ Measurement of the Performance of Freight Transportation,” Periodica Polytechnica Transportation Engineering, Vol. 27, Nos. 1- 2, 1999, pp. 83- 92.
Fawcett, S. E., S. R. Smith, and M. B. Cooper, “ Strategic Intent, Measurement Capability, and Operational Success: Making the Connection,” International Journal of Physical Distribution and Logistics Management, Vol. 27, No. 7, 1997, pp. 410- 421.
Ferreira, Luis and Julius Sugut, “ Intermodal Road/ Rail Container Transfer Facilities: Performance and Planning Issues,” Proceedings, Conference of the Australian Road Research Board, Perth, Australia, Nov. 9- 12, Vol. 16, Pt. 7, 1992, pp. 85- 100.
Forster, Paul W. and Amelia C. Regan, “ Electronic Integration in the Air Cargo Industry: An Information Processing Model of On- Time Performance,” Transportation Journal, Vol. 40, No. 4, 2001, pp. 46- 61.
28
W. Cottrell Cal Poly Pomona
Frozen Food Express Industries, Inc., 2000 Annual Report to Shareholders, 2440- 2001- AR, Dallas, TX, 2001.
Gerondeau, Christian, “ Freight Transport in Western Europe: The Case for Using New Units of Measurement,” Transportation Quarterly, Vol. 50, No. 3, Summer 1995, pp. 51- 58.
Gunasekeran, A., C Patel, and E. Tirtiroglu, “ Performance Measures and Metrics in a Supply Chain Environment,” International Journal of Operations and Production Management, Vol. 21, Nos. ½ , 2001, pp. 71- 87.
Holcomb, Steve, “ Kirby Corporation Announces 2004 First Quarter Results,” Press release, Houston, TX, Apr. 2004, www. kirbycorp. com/ 3_ ir/ pr/ 042904. pdf. < Accessed on Sep. 20, 2008>
Holguin- Veras, Jose, Ellen Thorson, and Kaan Ozbay, “ Preliminary Results of Experimental Economics Application to Urban Goods Modeling Research,” Journal of the Transportation Research Board: Transportation Research Record 1873, National Research Council, Washington, DC, 2004, pp. 9- 16.
Huynh, Nathan N., “ Analysis of Effects of Container Dwell Time on Marine Terminal Throughput and Rehandling Productivity,” Proceedings, 87th Transportation Research Board Annual Meeting, National Research Council, Washington, DC ( on DVD).
IBISWorld, Local Delivery and Messenger Services in the U. S., IBISWorld Industry Report, Aug. 21, 2008.
International Air Transport Association, World Air Transport Statistics ( WATS), 52nd edition, 2007.
Jones, Crystal and Joanne Sedor, “ Improving the Reliability of Freight Travel,” Public Roads, Vol. 70. No. 1, Jul./ Aug. 2006.
Jordan, William C. and Lawrence D. Burns, “ Truck Backhauling on Two Terminal Networks,” Transportation Research B: Methodological, Vol. 18B, No. 6, Dec. 1984, pp. 487- 503.
Lai, Kee- Hung, E. W. T. Nagai, and T. C. E. Cheng, “ Measures for Evaluating Supply Chain Performance in Transport Logistics,” Transportation Research Part E, Vol. 38, No. 6, 2002, pp. 439- 456.
Lai, Kee- Hung, E. W. T. Nagai, and T. C. E. Cheng, “ An Empirical Study of Supply Chain Performance in Transport Logistics,” International Journal of Production Economics, Vol. 87, No. 3, Feb. 2004, pp. 321- 331.
Lawrence, Denis, John Houghton, and Anna George, “ International Comparisons of Australia’s Infrastructure Performance,” Journal of Productivity Analysis, Vol. 8, No. 4, Nov. 1997, pp. 361- 378.
Levine, Harvey A., “ Toward Practicality in Defining and Measuring Railroad Productivity,” Transportation Research Record, Transportation Research Board, National Research Council, Washington, DC, 1985, pp. 1- 5.
Mastio and Company, “ 12th Edition, Natural Gas Pipeline Report,” Press release, May 6, 2008, www. mastio. com/ press/ gsplpres. html. < accessed on Sep. 20, 2008>
Mentzer, John T. and Brenda P. Konrad, “ An Efficiency/ Effectiveness Approach to Logistics Performance Analysis,” Journal of Business Logistics, Vol. 12, No. 1, 1991, pp. 33- 62.
29
W. Cottrell Cal Poly Pomona
30
Mele, J., “ Solving Driving Turnover,” Fleet Owner, Vol. 84, No. 9, Sep. 1989, pp. 45- 52.
Miller, Bruce I., “ Analysis of Shipping Performance Measurements,” Production and Inventory Management Journal, Vol. 31, No. 1, 1990, pp. 13- 16.
Morash, Edward A., “ Linking Public and Private Performance Measurement,” Journal of the Transportation Research Board: Transportation Research Record 1729, National Research Council, Washington, DC, 2000, pp. 42- 50.
Pipeline & Gas Journal, 27th Annual 500 Report, Nov. 2007.
Schrank, David and Timothy Lomax, The 2007 Urban Mobility Report, Texas Transportation Institute, Texas A& M University System, College Station, TX, Sep. 2007.
Smith, Narida, Garland Chow, and Luis Ferreira, “ E- business Challenges for Intermodal Freight: Some International Comparisons,” Journal of the Transportation Research Board: Transportation Research Record 1790, National Research Council, Washington, DC, 2002, pp. 20- 28.
Stainer, A., “ Logistics: A Productivity and Performance Perspective,” Supply Chain Management, Vol. 2, No. 2, 1997, pp. 53- 62.
Stewart, G., “ Supply Chain Performance Benchmarking Study Reveals to Supply Chain Excellence,” Logistics Information Management, Vol. 8, No. 2, 1995, pp. 38- 44.
Taylor, G. Don, Santhanam Harit, and J. R. English, “ Hub and Spoke Networks in Truckload Trucking: Configuration and Operational Concerns,” Logistics and Transportation Review, Vol. 31, 1995, pp. 209- 246.
Transportation Research Board, Performance Measures to Improve Transportation Systems and Agency Operations, TRB Conference Proceedings 26, National Research Council, Washington, DC, 2001.
U. S. General Accounting Office ( GAO), Natural Gas Pipeline Safety: Integrity Management Benefits Public Safety, but Consistency of Performance Measures Should be Improved, GAO- 06- 946, Washington, DC, Sep. 2006.
Vachal, Kimberly and John Bitzan, U. S. Grain Rail Market Indicators, Upper Great Plains Transportation Institute, North Dakota State Univ., Fargo, ND, Nov. 2005.
Waterborne Transportation Lines of the United States, Volume 2 – Vessel Company Summary ( WTLUS). Institute for Water Resources, U. S. Army Corps of Engineers, Alexandria, VA, 2006.
http:// en. wikipedia. org/ wiki. Wikipedia ( online encyclopedia), various articles and entries.
www. nationalatlas. gov. “ Overview of U. S. Freight Railroads.” < Accessed on Sep. 8, 2008>
W. Cottrell Cal Poly Pomona
Table A1. Leading Commercial Motor Vehicle Carriers in 2006 ( from Commercial Carrier Journal, Aug. 2007) 31
W. Cottrell Cal Poly Pomona
32
W. Cottrell Cal Poly Pomona
Table A2. Top Air Cargo Carriers ( Air Cargo World, “ Air Cargo Excellence Survey,” Mar. 2008) 33