Cracking codes

Once hailed as a game-changing technology, RFID was expected to revolutionise supply chains, so what happened? Barney Cox investigates


'The Thing' sounds like the sort of 1950s B-movie monster whose presence is accompanied by the eerie sound of a Theremin to add extra spookiness. However, in this instance it refers to the Great Seal bug, an eavesdropping device that the Russians sneaked into the US Embassy in Moscow. Presented by schoolchildren, The Thing was a cunning device designed to pick up what was being said and transmit it by radio for interception. By coincidence, it was designed by Léon Theremin, the man behind the eponymous instrument notorious for adding spooky sound effects to B-movies and shrill sounds to Beach Boys' hits.

Also, because of the way it worked, it is often considered a precursor of the radio frequency identification (RFID) tag.
A few years ago RFID tags were very much ‘the thing' for solving all manner of supply chain and product identity problems. Their ability to uniquely identify the product or package they were attached to when read by a radio frequency receiver was predicted to be the way to keep tabs on products everywhere from the factory floor, through the supply chain and to the end-user, whether via retail or commercial distribution.

"People thought that RFID was a solution that covered the whole infrastructure," says Domino Printing Sciences global account manager Craig Stobie. "It didn't and the infrastructure is the hard part."

His views are echoed by the findings of technology research firm Gartner in its report RFID in the 2009 Supply Chain: Overview and Best Practice Guide for Maximum Investment Value, which states: "RFID is best-known as the thing that was going to revolutionise supply chain management. The hype asserted that tagging everything in the supply chain would solve all the control and visibility issues that plague modern supply chains." The report goes on to add that "there is no magic quadrant for RFID; it doesn't make sense to look at RFID as a homogeneous technology or market".

The applications where RFID has made inroads include high-value goods and pharmaceuticals, asset tracking and medical devices. And even in those areas it is not a solve-all technology, cautions Stobie.

Best practices
Gartner has defined a checklist of seven best practices for identifying whether RFID is relevant. They are: start with the business case and realistically examine the alternatives; focus on the business process; align hardware capability with the environmental situation; decide on closed-loop versus open-loop deployments; don't neglect data quality; select vendors that understand your problem; and lastly, narrow the scope, address security and simplify the architecture.

One of the drivers behind the adoption of RFID technology has been demand for automated traceability of products through what is known as mass serialisation. Mass serialisation means identifying individual products down to a package level instead of having a barcode that identifies something as a particular type of product. For example, on a 16-tablet pack of a certain dose of a branded drug, a mass serialisation code would carry more detailed information, such as batch number, manufacturing date and use-by date as well as a further unique part of the serial number that identifies that particular packet.

The move towards mass serialisation is being driven by healthcare legislation, including the Californian Pedigree legislation, and efforts from the pharmaceutical industry to impose its own standards rather than face a myriad of different and costly country-specific legislation.

Another area where it is becoming important is in tobacco production where it is used to manage parallel trade, to make sure that duty is being paid on products in the country in which they end up being consumed.

Tagging options
When it comes to mass serialisation it's necessary to have a machine-readable code on each item, but that doesn't have to be an RFID tag. Instead it could be a 1D or 2D barcode. When it comes to selecting the most appropriate data carrier the criteria include cost and read reliability.

"Generally RFID is de-selected because of cost," confirms Stobie. "And in many situations RFID is only 90% accurate whereas the only thing that is 99% accurate is an ECC200 2D data matrix."

RFID does have a place when products are ganged together at pallet level and above, at which point its reliability and cost become viable. This means that there needs to be what Stobie terms "aggregation systems". This is the workflow and software that enables you to link information about individual items to the packets they are in and information about those packets to the pallets they are on and so forth across the supply chain.

Below pallet level there is still a need for the infrastructure to read data and for each pack to be identified by a unique number, but for reasons of cost and reliability these tend to be 2D barcodes.

Common technologies
Although 2D barcodes are cheaper than RFID tags and can be printed using a number of established variable-data printing techniques, such as lasering and inkjet, to generate unique product IDs, they do place more demands on the print process.

"Machine-readable codes mean production has to be consistent," says Stobie. "Print areas and colours need to be carefully controlled if codes are going to be read reliably. If you can't print reliably at line speed, your business will suffer." In this instance, there are two parts to that printing: one is the printing of the graphics on the carton or label, the second is the later overprinting at the packing line with the unique identifier.

The first printing will have effects on the effectiveness of the second when it comes to machine reading that demand greater consistency than ever before.

"Gone are the days of a carton just having to look nice and have a human-readable best-before date," he says. "If the best-before date was a bit grey, it wouldn't stop the customer from reading it."

Because of the adaptability of the eye we can account for very big differences in appearance and still make out the necessary information. On the other hand a camera or scanner may have a much narrower tolerance and therefore changes in the print's density, colour or surface gloss may have a big impact on that import read reliability. This is another driver towards tighter print production standards using objective data.

The Thing that is RFID may have started out as a solution looking for a problem. As a result firms have focused on their supply chains, often to find that while RFID could be an answer to their questions there are often better and cheaper answers available


CODE BREAKING: MAKING SENSE OF THE JARGON

RFID tags are only one way of encoding information about an object onto the object itself or its packaging. These machine-readable codes (MRC) are also classed under another acronym, AIDC, which stands for Automatic Identification and Data Capture, and include barcodes, optical character recognition (OCR) and magnetic stripes as well as the three types of RFID.

Picking the most appropriate technology depends on the requirements of each application in terms of cost, available space, amount of data that needs to be stored and how the data is read.

Barcodes and OCR need scanners or cameras that can ‘see’ them in situ to be read. However, they have the advantage of being able to be printed using standard processes including laser and inkjet.

One of the key features of RFID, because it is read using radio waves, is that sensors don’t need to be in the line of sight of the tag to read it. This can be an advantage or a disadvantage depending on the application. While it makes it possible to scan multiple items in the vicinity of the reader, for some applications such as passports, it has raised privacy concerns.

Active RFID
This is the most expensive type of tag. It includes a battery and broadcasts its information – good for long range uses.

Passive RFID
The simplest and least expensive kind, which relies on the power of the receiver to transmit its information. This means the distance a tag can be read from is limited.

Barcodes

1D barcodes These are the familiar barcodes used to identify your shopping at the checkout, and are widely used for identifying products.

2D barcodes
These are also known as 2D data matrices and are becoming more common for postal and logistics uses as well as for product identification. The advantage is they can hold more data, up to 2kb, making it possible to give individual products a unique identity number for traceability.

The ability to carry more data, such as a website URL, means they can also be used for advertising, marketing and promotion, linking print to other media. Examples of this include QR Codes, which are widely used in Japan, with the software beginning to appear on smart phones in the West.


2D ART: HELLO WORLD
What’s a wheat field got to do with machine readable codes? In this instance it’s an example of how versatile a 2D data matrix can be, and the range of imaging technologies that can produce them isn’t limited to printing. German artist Bernd Hopfengärtner’s Hello World project involved mowing a 160x160m 2D data matrix into a wheat field near the town of Ilmenau in Thuringia, Germany in May 2006 with the intention that it could be read and decoded from the aerial photographs used by Google Earth.

The data encoded in the pattern of the 2D data matrix reads Hello World, which is a phrase that programmers commonly reproduce in their first lines of code when they are learning a new language.

The Hello World project is the biggest example of a 2D data matrix; at the other end of the scale some microelectronic devices have had the matrices etched into their surfaces that are just 300 micrometres across.