Home Internet of Things Aerospace Apparel Energy Defense Health Care Logistics Manufacturing Retail

The RFID Reader Architecture Debate

The question of whether to treat the EPC reader as simply another automation input device or as part of the IT infrastructure is a frequent topic of discussion. This article explores that debate.
Oct 05, 2005This article was originally published by RFID Update.

October 5, 2005—Most manufacturers subject to retailer mandates for EPC RFID continue to treat case and pallet tagging as an exception-based process that occurs outside the flow of normal production. Some Engineering Managers responsible for automation of both packaging lines and material handling equipment within these firms, however, have given considerable thought to the point down the road where EPC RFID is potentially incorporated into mainstream production and distribution, and how best to incorporate this technology into their automated architectures.

While the process itself is paramount due to the need to maintain production optimization, the question of whether to treat the EPC reader as simply another automation input device or as part of the IT infrastructure is a frequent topic of discussion. This debate can be likened to the "PLC (Programmable Logic Controller) versus PC (Personal Computer)" discussion in manufacturing engineering circles, but in the RFID realm it will ultimately drive the type of reader deployed and skill sets necessary to operate, maintain, and upgrade reader infrastructure.

Most of the leading CPG, food, beverage, and similar manufacturing companies subject to retailer RFID mandates already have sophisticated automation systems in place that control the high-speed packaging lines in their manufacturing operations and high-speed conveyors in their distribution facilities. These systems are typically controlled by Programmable Logic Controllers (PLCs), which are dedicated automation systems programmed and maintained by electricians or technicians.

In the 1990s, the introduction of Microsoft Windows-based products and Ethernet networks for use on the plant floor spurred numerous debates as to the personnel skill sets necessary to maintain them, plus the significantly shorter product lifecycles of the PC-based products and the consequent need to frequently upgrade or patch them. A similar debate is now brewing concerning incorporation of EPC RFID infrastructure into production equipment. This debate extends beyond the thick-versus-thin client reader debate but will similarly impact the type of reader/encoders purchased for each type of architecture.

Some forward-looking manufacturers are already attempting to automate the currently DC-centric tagging process by using photoeyes (photoelectric sensors, a typical PLC sensory input device) to detect a case that has a tag that needs encoding. When a case is detected that needs encoding, the existing PLC-based conveyor control system activates or turns on the RFID reader to encode the tag. This process is predicated on open-air tag encoding, however, a process that we don't anticipate will be universally viable until manufacturers have more experience with EPC technology.

In the more conventional scenario where a validated tag is placed on a case and then sent down a conveyor line, it is easy to see how the RFID reader can be incorporated into the existing automated infrastructure as a simple PLC input device. In this case we anticipate the PLCs and their supervisory systems will incorporate more of the tasks associated with embedded RFID middleware, including the new ALE (Application Level Event) functionality just standardized by EPCglobal. This last piece is not a necessity, however, as the PLC could still feed the reader inputs to higher-level IT systems for processing.

As RFID becomes more ubiquitous, is integrated with the mainstream production process, and reader prices drop, this PLC-centric architecture will have direct appeal to Engineering Managers interested in maintaining and upgrading the incremental RFID infrastructure using their existing personnel skill set. Several manufacturers we spoke to recently on this topic indicated that this scenario was a distinct possibility in their facilities if reader prices fell below US $100.

At the other end of the spectrum, the more IT-oriented manufacturing production people we've spoken with recently envision the possibility of RFID readers functioning as network appliances connected by Ethernet or ultimately wireless networking. Proponents of this approach tend to view the readers more as thick clients capable of edge computing, versus the more basic Auto-ID/sensory input perspective espoused by advocates of the PLC approach. The IT approach would also typically entail maintenance and support by IT or Manufacturing IT personnel versus the plant floor or DC work force.

Regardless of which option is pursued, proponents of both approaches expressed the desire to be able to remotely upgrade their readers via central administration. This sentiment is particularly strong among early adopters, many of whom have had to or will have to conduct firmware upgrades as EPC technology progresses. Harkening back to the Microsoft analogy, this need for frequent upgrades and patches relative to the stability of the automation systems has caused significant heartburn regarding the use of PC-based systems in production environments that often run 24/7. Manufacturers considering incorporation of RFID readers into their production infrastructure must similarly consider their upgrade strategy in order to not impact production.
  • Previous Page
  • 1
  • Next Page

Login and post your comment!

Not a member?

Signup for an account now to access all of the features of RFIDJournal.com!

Case Studies Features Best Practices How-Tos
Live Events Virtual Events Webinars
Simply enter a question for our experts.
RFID Journal LIVE! RFID in Health Care LIVE! LatAm LIVE! Brasil LIVE! Europe RFID Connect Virtual Events RFID Journal Awards Webinars Presentations