Overcoming RFID Limitations

By Chris Kapsambelis

Overview

Several supply chains are moving to the use of RFID (radio frequency identification) in logistics applications. RFID has many desirable capabilities but based on extensive testing and dealing with the laws of physics, there are real and practical limitations. In order to help those implementing RFID-based logistics systems, this white paper will identify some of limitations and propose ways to deal with them.

In October of 2004 at the Truth In Technology Forum sponsored by the AIDC 100 (www.aidc100.org), representatives from Energizer (a supplier to Wal-mart, a major proponent of RFID) and the other representatives of the Department of Defense (conducting large scale tests) as well as others, indicated that there are some significant limitations to the use of RFID in logistics applications. As of May of 2005, further testing continues to point to practical limitations to many of the anticipated RFID operation benefits.

RFID was selected over barcode as the technology to encode the Electronic Product Code (EPC), a product identification method providing a serial number for every product produced. The EPC is being advanced through GS-1 (formerly the Uniform Code Council (UCC)) because RFID is believed to offer superior reading performance characteristics. Among the important characteristics cited are the following:

  • Unlike barcode, RFID does not require Line-of-Sight between the reader and the tag.
  • Tags can be read at greater distances.
  • Groups of tags can be read all at once.

These properties permit a high level of automation in capturing data as goods flow between trading partners in many supply chains. The ideally accepted form of this automation employs RFID instrumented portals at shipping and receiving facilities. Such portals have been designed to capture all the data from palletized loads as they are being transported on forklifts in and out of each facility.

The efficiency of this type of operation is derived from the fact that it is not necessary to manually scan each and every case in order to receive it. Furthermore, as pallets and cases are dispersed throughout a given facility, strategically located readers will automatically capture the data needed to track pallets and cases.

Active RFID may possess properties that will permit the capture of all EPCs from a pallet of cases traveling through instrumented portals. The passive RFID tags that have been specified for use in several supply chains have been shown to have a number of limitations that reduce the read rate for pallet loads from 100% to a range between 50% and 80%, depending on the size and content of the cases involved.

Limitations

The limitations of passive RFID arise from the need of these tags to absorb enough power from the reader to transmit the stored tag data. In order to accomplish this, the tags must utilize directional antennas large enough to intercept the needed power from the reader. The view that RFID does not need a “Line-of-Sight” between the tag and the reader has to be modified as follows:

  • Tags must be facing the reader, and a direct line must exist between tag and reader unobstructed by any metallic or liquid object, or other tags.

The need for the tag to be powered by the reader’s RF radiation pattern further limits the distance between tag and reader. The amount of reader radiation is also limited by the need to avoid interference from adjacent reader installations as well as restrictions placed by the FCC, and the BRH.

Many people believe that future developments will overcome most of these limitations; however, field testing indicates that the read rate for pallet loads will always be less than what is acceptable for Shipping and Receiving applications, and other methods must be found to retrieve the lost data.

System Design

The limitations of RFID need to be considered when designing data collection systems for use by trading partners in the supply chain. Currently, orders have to be received by scanning barcodes on each case manually. The promise of RFID is that orders can be received automatically by reading all the tags for each case on a pallet load.

Given the less than perfect readability of tags because of everyday obstructions and interference from the physical properties of the products carrying the RFID tags, a higher level of data synchronization will be required in order to derive the missing tag readings. This can come about by requiring shippers to provide the additional data needed for synchronization in advance of the physical arrival of a given order. In addition to the currently supplied data for receiving an order, the receiver will need access to data that permits the reconstruction of all RFID case data associated with a tagged pallet. In this type of environment, the reading of as little as a single RFID tag, since it contains a unique serial number, can be used to derive all of the following data:

  • Location based on the physical location of the reader reading the tag.
  • Time and date based on a “time stamp” by the reader or the computer receiving the read.
  • Inbound or outbound status derived from lookup of the tag number or the associated pallet number in the synchronization database.
  • The pallet number and all case tags initially assigned to the pallet derived from lookup in the synchronization database.

With this additional capability, these low level systems will provide higher-level applications with the perfect transaction data needed for smooth operation.

Shipping Point Requirements

In order to comply with this extra need for data synchronization, hardware and software will need to be acquired and installed that can accomplish at least the following.

  • Prepare an RFID tag for application to each case and pallet scheduled for shipment to a trading partner.
  • At the point of tag application to a case, read the case tag and transmit the data to a system that can add all the needed fields to describe the contents of the case.
  • At the point where pallets are being loaded with cases, RFID readers are required to read tagged pallets and each case as it is being loaded onto the pallet. The readers transmit the data to a system that can add the pallet tag reading to a pre-existing record of the case tag.
  • A computer system that can receive data from the RFID readers, and build the database of case records and associated pallet tag numbers.
  • The capability to integrate the case/pallet data with other systems needed for Shipping/Receiving, Inventory, WMS, ERP, etc.
  • The capability to share the data with various trading partners for data synchronization.

Receiving Point Requirements

In order to automatically receive orders of palletized goods, a system will be required that has access to synchronized data from each trading partner. This system will require at least the following:

  • Strategically installed RFID scanners to read at least one RFID tag from each pallet.
  • Each RFID reading will be transmitted to a system that can derive the data needed for receiving all the cases on the pallet.
  • The resultant data will be configured into the necessary transactions for receiving by higher-level applications.

Conclusion

RFID and EPC have always been envisioned as requiring middleware to support data capture and derive transactions for higher-level applications. By requiring the linkage of a pallet EPC to case EPCs loaded on the pallet, RFID reader requirements can be simplified as follows:

  • Case tags are read one at a time, as they are loaded onto pallets.
  • Readers at receiving portals need only read any of the RFID tags on a pallet loaded with cases. The direction (inbound/outbound), and unread tag numbers can be derived from the synchronization database.

The need to develop RFID pallet readers that can read the pallet tag and 100% of the case tags on the pallet is not possible due to the limitations of RFID by the laws of physics (because of the material characteristics of the products and the interference or obstruction they produce. By linking pallet tags with associated case tags, and synchronizing the data between trading partners, the requirement for 100% readability of pallet loads is eliminated.