III. Freight Identification Trends and Strategies

A. The Long Term Trend in Freight Identification Technology

Two revolutions swept through the movement of freight in the last forty years: the introduction of standardized intermodal containers and the application of information technology. Those revolutions brought dramatic changes in business practices, reductions in transit times, lower real costs of transportation, and they contributed to the explosion of world trade and prosperity. Although containerization is relatively mature, information technology seems to be changing at an accelerating rate as service alternatives multiply and the real price/performance ratios continue to drop for computing power, data storage, and telecommunications.

Both revolutions raised the importance of and increased the spending on information about freight shipments, including information about their identity, location, condition, and status. With some irony, hard-won improvements in the information flows have often been accompanied by great frustration as expectations move ahead more quickly than the improvements. However, expectations notwithstanding, there is still much to be improved in the timeliness, accuracy, completeness, and integration of freight data.

The long-term trend in freight identification technology is moving towards automatic dependent surveillance of materiel movements and freight shipments. Automatic Dependent Surveillance (ADS) is a term used in air and vessel traffic control for on-board equipment that automatically determines location and other relevant information without intervention from crew or network managers. ADS is a necessary kernel for the possible development of "free flight" air traffic control, where most flow control functionality would be distributed to individual aircraft. Today's over-land air traffic control systems are still somewhat short of ADS since ground-based connectivity is essential: the air traffic control system records digital data taken from radar readings and on-board radar transponder signals that uniquely identify individual aircraft.[5]

The critical feature of automatic dependent surveillance in regard to freight shipments is that the identification and data recording processes become fully automatic byproducts of managing the operation. The movement or traffic management process provides an inherent trigger for identification data collection. Human beings are out of the loop. A secondary, more distant feature of automatic dependent surveillance for freight would be that a tag on or with the freight shipment would manage the identification and reporting process on its own.

For most of the century prior to World War II, freight shipment status information was a rare exception that required the exchange of telegrams or teletype messages, often with multiple railroads. In the decades after the war, we moved to telephones and faxes to supplement what were at best semi-automatic status reports.[6] Long haul truck drivers, for example, depended on pay phones to report their status to dispatchers.

The early 1990s were watershed years. Aviation implemented automated tracking systems as described above, although the focus remains on traffic management of the aircraft themselves, not their passengers or cargo.[7] For surface movements, Qualcomm pioneered in combining Global Positioning System (GPS) data and two-way mobile digital satellite communications. Schneider National was the first motor carrier to deploy Qualcomm's OmniTracs service to its entire fleet. Schneider received automated real-time position reports on the location of its tractors, then integrated that data into its customer service, load planning, driver assignment, and maintenance management systems. In 1991, the Class I US railroads agreed to a mandatory Association of American Railroads (AAR) standard for an Automatic Equipment Identification (AEI) system based on radio frequency identification (RFID) tags. Today, nearly 98% of rail equipment in interchange service has two AAR standard RFID tags affixed to it.[8]

These changes in aviation, trucking, and railroading represent a major turning point. All took people out of the loop and derived status information from other activities. This trend is far from finished. There are at least three ways the trend towards automatic dependent surveillance will continue to play out: greater penetration of freight markets (especially motor carriers); tighter and more transparent integration of freight or lading information with transportation equipment information (especially railroads and air freight); and the development of new capabilities and services—akin to free flight—through information technology.

B. Freight Identification Strategies

It is useful to think of three business process strategies or models or architectures connected with freight identification. The first is on its way to being history; the second represents today's best practices; and the third holds promise for the near future. Each model draws on the metaphor of human communication. After all have been discussed, Table 1 compares their characteristics.

Table 1 - Characteristics of Freight Identification Strategies*

	"Mute Freight, Deaf Networks"	"Talking Freight, Listening Networks"	"Smart Freight, Smarter Networks"
Freight Items	Stencils Paper Punch cards	Barcodes RFID "license plate" tags	Barcodes on low value goods Intelligent microprocessors on higher value goods Auto-networking tags
Trailers, Containers	Paper	Barcodes RFID tags – AEI "license plates" – content data	Untethered location monitors Auto-networking tags Intelligent Microprocessors
Power Units (Tractors, Trains)	Paper	AEI RFID DSRC GPS + Fleet Mgmt	AEI RFID DSRC GPS + Fleet Mgmt. On-board intelligence
Networks	Manual and semi-automatic data entry Pre-set rules Limited flexibility, slower to respond	Automatic data capture and entry Intelligence resides on the network, at its hubs	Automatic data capture and entry Sentinels and agents collect status data, adjust operation Richer, more distributed decision support

* Each cell suggests possible combinations of items or technologies.

1. Mute Freight in Deaf Networks

This is the traditional model, used prior to adoption of automated identification techniques. Freight items might be identified with stenciled shipment information, attached packets of paper documentation or punch cards. Freight is mute, requiring a person to capture information from it, usually by writing it down or pulling off a piece of paper or a card. Networks are deaf, depending on manual or semi-automatic data input. Deaf networks operate largely according to pre-determined rules, and they require significant manual intervention to adjust schedules or flows.

This model or strategy was the state of art before and through the early stages of the information technology revolution. It persists in surprising ways. For example, in the port of Charleston, containers are managed effectively with chalk marks on each container showing its voyage number and whether it is empty, light, or heavy (E, L, or H).[9]

2. Talking Freight in Listening Networks

Talking Freight. The dominant feature of this strategy is the use of shipment identification technologies that can be identified or read automatically—think of them as variations of license plates that can talk to readers. ("License plate" is used simply to convey the one-of-a-kind nature of the freight identification technology, not any aspect of government regulation). A unique identification number is associated with each shipment and each conveyance and moves with it. The license plate identifiers are the keys to building and maintaining nesting relationships about shipments in distributed databases that belong to the shippers, carriers, and regulatory agencies, such as Customs.

"Talking" license plate technologies come in a range of sophistication.[10] The oldest and simplest are printed linear barcodes followed by their more robust two-dimensional descendants. When implemented well—starting, for example, with good quality printers—barcodes are part of an effective and relatively inexpensive system. They are the keys to what some call a barcode-and-database strategy that will continue to be used in appropriate high quality logistics systems even after more sophisticated tools become cost-effective.

Intermediate license plate technologies include simple read-only reflective RFID tags, such as those used by the railroads. These tags are unique, permanent freight car identifiers, and all relationships to specific shipments are established and maintained in remote databases. When implemented correctly, reflective or passive RFID systems are highly reliable. They can be thought of as simple talkative tags.

Data-rich read/write RFID tags are more sophisticated. These tags carry unique identifiers and allow users to write data about details of individual shipments. However, in transportation operations, these tags are used primarily in license plate applications roughly comparable to read-only tags.[11]

The most sophisticated license-plate-like technologies in use today are seldom thought of as such: the real-time location determination systems used mostly in trucking. These systems contain or pass along unique unit or tractor identifiers to permit network management systems to recognize them and associate their location and status information with the proper shipments, etc. At a minimum, transmissions from these systems include time and location information in addition to the license plate identifier.

Barcodes, image recognition, and RFID tags generally communicate over short distances to a fixed-site signpost reader. Users and networks know only where and when a bar-coded or tagged shipment was read; they are limited to inferences about status changes after the last report. On the other hand, real-time location determination systems use mobile long range communications to provide nearly continuous status updates and higher degrees of confidence.

A note of caution: applying freight identification technologies adds new risks to the shipment process. As a simple example, while barcode systems can be very effective, illegible or damaged barcode labels will disrupt shipment processing. For a more complex example, consider the use of relatively sophisticated RFID tags on shipments. Automated systems really track automated media, not the freight or conveyance itself. As a result, the potential for error multiplies since automated media may fall off a shipment, be attached to the wrong shipment, not be updated when the shipment status changes, or be disrupted by battery failure.[12] In any of these circumstances a shipment could be handled correctly while control systems reflect incorrect status information. In order to manage the added risk of error, application of sophisticated freight identification technologies increases the need for effective system design and strict system discipline

Listening Networks. The minimum common denominator for the networks within which license-plate-like technologies operate best is that the networks can listen or read: they can automatically capture transaction data with little or no human intervention. (Handheld barcode and RFID readers are automated devices that still require some human action).

Another common feature of listening networks is that intelligence resides in the network management system and with the people who operate it. Intelligence does not travel with the unit of freight. Warehousemen, dispatchers and operations centers often can use the automatically collected status information to assess some or all of their operation and initiate corrective action when necessary. In its most advanced form, significant decision aids are built into the network control systems to use the license-plate-related information in powerful ways. One example is the operating efficiencies achieved by truckload carriers such as Schneider National when they integrate their license-plate-like location status information with other data and models.

3. Smart Freight in Smarter Networks

Looking ahead, one can anticipate significant changes in the nature of freight identification technologies and important complementary extensions to current trends in freight networks. This section sketches these changes. Section VI outlines the key technology developments that are likely to be the catalysts for broader adoption of smart freight and smarter networks.

Smart Freight. Smart freight means smart tags attached to or built into the freight. The tags will combine at least four elements: a license-plate-like identifier, read/write storage capacity for shipment-specific data, portable processing power or decision capacity, and communications capability. A fifth element will probably be real-time location determination for the freight or conveyance. The license-plate-like capability would continue to be used in a default mode as described above for today's talking freight applications. The read/write data storage would both enable value-added activities in the field and provide some of the grist for on-board decision support activities. The portable processing capability would establish the potential for the tag to initiate specific actions when status changes trigger programmed thresholds.

These capabilities exist today, although their high cost and package size limits them to specialized applications. One example is the Army's combination of battery or variable-powered deployable boxes with laptops, GPS receivers and Inmarsat transceivers, with and without RFID tag interrogators, for tracking high priority shipments.[13] Perhaps the best indication of what lies ahead is a new offering by HighwayMaster, the number two provider of GPS-based fleet management systems. Their "Rolling ETA" capability resides on the truck, and it seems fair to assert that other kinds of automated on-board or "on-freight" analysis will develop in the future.

"With the Rolling ETA software option, the Series 5000's onboard microprocessor automatically monitors a truck's progress along the route you specify. Using the truck's reported speed and cost-free GPS satellite location reports, the system continually calculates the truck's estimated time of arrival. If a truck is running behind, the system immediately delivers an alert to the dispatcher so he can begin addressing the problem right away. If the dispatcher does not receive an alert, he can rest assured that his trucks are running on time. As a result, dispatchers can focus their time on late trucks, so they can manage more trucks and loads…Rolling ETA helps contain communication costs because…you only incur a charge when an alert is transmitted."[14]

Smarter Networks. We have already discussed some contemporary developments in adding intelligence and decision aids to freight network management systems. This trend seems likely to accelerate. Smart networks would take as a given the capability to gather and maintain automated status information. They would also include much tighter and more transparent integration of data sources, models, decision support tools, and users. The Defense Advanced Research Projects Agency (DARPA), in its Advanced Logistics Program, is emphasizing new capabilities such as intelligent agents and sentinels which would add to the richness and flexibility of logistics network management capabilities.[15] Looked at in terms of its analytic tools, logistics is rocket science and will become even more so.

previous | next