Inside GSK’s New RFID Pilot

May 5, 2006
The initial benefits will be modest, but GSK’s new pilot goes beyond what other manufacturers have attempted to do with RFID data, says IBM’s Paul Chang.

GlaxoSmithKline recently announced that it has begun putting radio frequency identification (RFID) tags on individual bottles of HIV drug Trizivir, which is packaged at its Zebulon, N.C. plant. According to the company, the main goal of the project is to better authenticate the product, which, according to the National Association of Boards of Pharmacy, is one of the 32 drugs most susceptible to counterfeiting or diversion. Another obvious benefit will be for the company to gain a better understanding of how to implement RFID and capture its myriad business benefits.

GSK’s pilot follows other high-profile forays into item-level RFID, namely those by Purdue Pharma (for Oxycontin) and Pfizer (for its “little blue pill”). GSK’s pilot represents a significant milestone, says Paul Chang, associate partner with IBM Business Consulting Services, which has counseled GSK on every step of the pilot. “This is the most complete RFID implementation in the pharmaceutical industry to date,” he says. The key difference, he says, is that the architecture will allow GSK not just to track product better, but also to use the data more effectively and integrate it with legacy applications.

IBM and GSK took a two-pronged approach to the project, Chang says. “The driving factor was product security,” he says. “But they saw the potential business benefits as well.”

Among those benefits GSK hopes to attain in due time:

  • Electronic product authentication and electronic pedigree documentation;


  • targeted recalls, or knowing precisely which products need to be recalled, and which don’t;


  • chargeback resolution;


  • improved inventory management, forecasting and planning;


  • better supply chain oversight, such as improved management of shipping receipt confirmations and returns processing.

Using RFID will allow GSK to place more faith in the data it collects from supply chain partners. “Right now they share data [with partners], but it’s assumed data,” Chang says. “RFID is actual physical data that you can trust.”

GSK has chosen to use high frequency (HF) tags carrying electronic product code (EPC) data, which has the manufacturer’s ID but not the product ID in order to protect consumer privacy. While there has been much debate within pharmaceutical circles about whether HF or UHF (ultra high frequency) tags work best for item-level applications, the prevailing pharmaceutical industry view is that HF is the better technology. In its testing, IBM has had better than 99% success rates in tag reading, and Chang sees no reason GSK will not get similar results. Most tag failures in the tests were due to tags that were, as he says, “dead on arrival” and not because of any inherent flaws in the setup of the RFID infrastructure.

Chang explains that the RFID data, while critical to improved security and supply chain optimization, is really just a facilitator of IBM’s hub-and-spoke EPCIS architecture. The architecture will allow the addition of various applications based upon EPC data — such as electronic pedigrees, or targeted recalls — to be added as spokes to the hub, which becomes a single point of integration of this data with legacy systems such as the ERP or a warehouse management system. (Click here to view a diagram of the IBM solution.)

The next frontier for pharma RFID, Chang says, is to make the entire supply chain RFID enabled, to allow for seamless integration of data and full optimization of the technology. IBM is also currently working with wholesalers such as Cardinal Health to implement a similar architecture as that at GSK. “We’d now like to get hospitals and retail pharmacies involved as well to complete the link,” says Chang.

Click here for more information on the GSK pilot.

About the Author

Paul Thomas | Managing Editor