Axicon Blogs

By John Nachtrieb 08/01/2020

Who Assigns the GTIN?

Who is responsible for assigning and managing GTIN’s on medical devices? Typically, the brand owner is responsible for every aspect of GTIN management. Where the medical device manufacturer is also the brand owner, they assign GTINs to all the products they market. Who assigns the GTIN when the medical device manufacturer is also the contract manufacturer? The same product sold under a different brand requires a different GTIN, which  the private label company assigns. This is because they must be able to distinguish and track that product through a potentially different supply chain, monitor inventory depletion and replenishment, and implement an efficient recall if necessary.

Is a Changed Product a New Product?

New products always require a new GTIN. What is a new product? The simple answer is a product that is new to the market. A new product is one that did not previously exist in the brand owner’s offerings, or was not marketed by that brand owner. If the brand is introducing an altogether new product to the market, or a product to compete with a similar product already on the market, a new GTIN is required.

If a product changes, is it necessary to assign a new GTIN? Yes and no. The short answer is, any change to the product that the brand owner will want or need to track requires a new GTIN assignment to the changed product. Such changes could include sourcing an ingredient from a different supplier, adding or removing an ingredient from the existing product, or changing the formulation in any other way. This might include adding or changing a flavor, a coating, or some other “innocuous” modification.

Changes to the packaging of an otherwise unchanged product can also require a new GTN assignment. Changing the size or weight of the package can require a new GTIN. Changing the color scheme or adding a certification logo can also require a new GTIN assignment.

What About Updated Packaging?

Introducing the identical product a package with a different language necessitates a new GTIN, but adding a second language to the original packaging does not require a new GTIN. In addition, minor design changes or package sizing does not require a new GTIN.

Any change to a product that changes the regulatory or compliance statement, or which effects the way the product is shipped, stored or received requires a new GTIN assignment. There are some exceptions. Contact us for advice.

Corporate mergers and acquisitions present a unique challenge to GTIN managers. How do you handle the GTINs owned by the acquired company? The correct answer may require some research. Although the GTINs of the acquired company are an asset, you must be clear that they are an asset that conveyed to the acquirer, along with the other assets that came with the transaction. Legal advice recommended.

What About GTINs Acquired from a Merger?

By John Nachtrieb 09/10/2019

oupon coding is complicated. GS1 Databar Coupon Code does not simplify that—but it solves many issues with coupon offers and the stores that honor them. Coupons are, by their nature, complex. GS1 Databar Coupon Codes handle just about any offer with its many stipulations and conditions. Best of all, the barcode does it all: store personnel just need to scan it. The register and the database behind it do the rest.

Coupons and the barcodes on them have evolved. A lot. In the early days, coupon barcodes were an addendum to the UPC symbol on the product. Coupons bring a lot of important data to the brands a store sells. Even more data was desired.  It quickly became apparent that the coupon needed to do much more than just offer the customer a simple discount. The barcode needed to do more.

Databar Solves Problems

The problem was data capacity. The more complicated the coupon offer, the more data the coupon barcode needed to include. But in the early days, the problem was more basic. Every company that applied for a UPC symbol was assigned the 6 digits on the left half of the symbol. Everybody from a Mom and Pop operation selling canned tomatoes to Proctor and Gamble selling hundreds of disparate items had the same range of coupon possibilities. All coupon codes included a UPC symbol with a prefix 5, and six digits with which to define the offer. That left room for a very few, simple coupon offers.

GS1-128 UPC addendum Code was an interim solution. It was that same 5-prefix UPC symbol with a trailing Code 128. It provided space for more offer arrangements but the barcode was very long. Considering where coupons are traditionally posted—magazines, newspaper ads and direct mail fliers—it was an interim solution at best. A 2D matrix code like QR could replace the addendum or even both barcodes. A great solution, but retailers understandably balked at the substantial cost and disruption of replacing all their scanners to read the 2D symbol.

Databar Handles Complex Coupon Offers

Plan B: Databar, a compact but high capacity 1D barcode that almost all scanning systems can read. It could replace both the UPC and addendum code but take up much less space by being stacked in rows. GS1, the standards organization for retail and supply chain systems, created a prefix for coupons in North America. Using this prefix, the UPC symbol identifying the purchased item is embedded in the coupon barcode along with the offer and conditions. This single barcode does it all.

GS1 Databar Coupon Code not only manages the offer, it tracks the conditions automatically. The offer can mix different brands: buy Product A and get 50% off on Product B. Offers can be very complicated: buy 1 and get 2; buy 1 and get 2 more at half price, etc. Databar Coupons can also define a start date and an end date.

By John Nachtrieb 10/07/2019

Quality Magazine, June 2019 published a great article entitled “The Current State of Quality,” by Michelle Bangert, Managing Editor. She makes the case that quality has changed over the years, both in terms of how it is done and what it means. In years past, the quality department was like a police force, looking for problems and often felt to be an obstruction. Today the quality department is mostly gone—now it is “everyone’s job.” Quality professionals are still necessary but they are more oriented to problem solving, not problem finding. That change rings true to me.

Standards have also helped redefine what we do and how we do it, regarding quality. When I started in this industry 45 years ago, I would not have believed that quality would ever change—or would ever need to change. But it has—and it has been good for the industry and for people like me. Those changes have created opportunities as well, for people and for the companies in which they work. It levels the playing field of quality versus price—now it is quality and price.

Time will Tell

One of the trends we are seeing in barcode quality is the proliferation of high speed, fixed mounted verification systems. Often these are integrated into scanning systems that also do other things. Do these systems really gauge barcode quality as well as they should? They are better than nothing but how good really?

One concern about inline systems is the disconnection they insert between the quality professional and the data they accumulate. The quality professional is still the problem solver but the data is aggregating—mountains of it—hands-off, automatically, somewhere. Is someone monitoring the data? Are subtle migrations in the grading of the ISO parameters being observed? Is the threshold of unacceptability being calculated? What do such systems gain? What do they lose? Often we have said that the best tool in barcode quality is a close, careful visual examination of the barcode.  More data does not equate to more certainty. When the problem solver is summoned, will he or she have anticipated and planned the next move? Time will tell.

More Data or More Certainty?

As a guy who enjoyed working on my car 45 years ago, I have personally experienced the disorienting sensation of trying to diagnose, let alone resolve a seeming simple problem with my modern car. All that technology has made it last longer and run more efficiently, but impossible to work on in the driveway, without a computer interface. Where do you start? When the red light on the light stack signals a barcode problem, and the press stops, is the person responsible for barcode quality going to know where to start? How to diagnose? What to do to resolve it

These are the Good Old Days

Don’t misunderstand—I am not nostalgic for the good old days. Carol King had it right—these are the good old days. I can see the possibility of some new, unanticipated problems arising and it is going to be fascinating watching it evolve.

By John Nachtrieb 07/05/2019

Where did ISO15416 for linear barcodes come from? Or ISO15415 for 2D symbols? What is the process that culminates in the development of an international standard? It is a fascinating story that varies by industry, with some shared aspects between different industries. The largest commonality is that of problem solving.  Barcodes exist and continue to be useful because they solve problems. The same is true for standards.

The process that led to the UPC symbol is well documented. Numerous applications and forms of automatic identification had been in use in other industries. A bullseye barcode was patented in 1949. A scanner to read the color-coded barcodes on railway cars was invented 1961. The development of the HeNe laser in 1969 made barcodes a viable way to identify items. Grocers saw these developments as applicable to problems they wanted to solve in their stores. Their industry trade association, the Grocery Manufacturer’s Association played a critical role in developing the Universal Product.

As barcode technology gained attention, many companies recognized the opportunity to get involved.  Competing developments made it obvious that the viability of the technology as a solution depended upon standardizing not only the scanners to read them, but the barcodes themselves. In 1971, the Material Handling Institute formed the Automatic Identification Manufacturers (AIM) association to hammer out standards for the fledgling barcode. In 1986, AIM became an independent association.

AIM has become the defacto step on the path to the international acceptance of most barcode standards. An important, little-known component of that process is the fact that these are open standards, as distinguished from closed or proprietary standards. Industry-wide or global systems must operate on open standards. All attributes of the barcode are disclosed in the standard—there are no secret, proprietary attributes. In addition to its formidable problem-solving capabilities as an inventory management tool, the open standard was key in propelling the UPC into a global symbology.

This same trajectory is how most global barcode standards evolve. In some cases, the players are different, but it always starts with a problem seeking a solution. A consortium of key leaders, often competitors, commit to developing a designated solution that will benefit the industry in a non-preferential way to individual companies. Not

Photo by Mark Duffel on Unsplash

all standards come through AIM and and become ISO standards, but many do.

Whatever the path, the goal is the same: a standard upon which member organizations can agree, and with which they can comply.

Compliance implies consequences for non-compliance. Non-compliance extends beyond the printed barcode. There are also standards for how scanners should perform. For example, a scanner should be able to read a barcode of a specified type, with bars and spaces that are neither smaller nor larger than a specified size, reading from a specified distance and angle.

When barcodes or scanners fail to perform as specified, operations are disrupted:

A manufacturing sequence is halted because the right part cannot be identified

An order cannot be fulfilled and shipped because an item cannot be picked

A shipment cannot be sent, tracked or received

An inventory cannot be replenished because a sold item is not debited

A patient cannot be dosed because the wristband does not match the medication

A patron cannot enter an concert or other entertainment venue because the ticket cannot be validated

Photo by Gabriel Matula on Unsplash

These, and innumerable other consequences, have costs, in both financial and experiential terms. Paused manufacturing and uncompleted fulfillment means late deliveries. Delayed shipments can cripple assembly lines and other time-sensitive functions. Missed medications or incorrect dosing can delay recovery, cause further injury or even death. Crowd control depends upon getting patrons into sports, concerts and other mass audience events.

Barcode standards are developed to solve problems and prevent consequences that range from inconvenient, expensive, disappointing or infuriating to life-threatening. They are relevant when they are created because they solve a problem.

By John Nachtrieb 29/03/2019

Not all barcode verifiers are the same, and the Axicon S range is both proof and prime example. It is so different and so much better than other 1D barcode verifiers, it is really a class of its own. The S range does what no other verifiers do and that not only makes it easier for the user, but it produces a significantly better report.

First the basics. As mentioned above, these are 1D only. They cannot test and grade 2D symbols such as Datamatrix and QR Code.  Like all Axicon verifiers, the S range are contact devices–you place the verifier in direct contact with the barcode. This does limit the size of the barcode the verifier can read and grade—but there is an important reason.

Non-essential light must be eliminated

The briefest possible explanation is that scanning (and therefore verification) works by detecting reflective differences in the barcode. It is of paramount importance the light and dark reflectance values be measured accurately, and to do that, the influence of extraneous light must be eliminated. Axicon’s solution is to position the verifier directly on the barcode.

Because the verifier cannot be moved in or out to adjust for shorter or longer barcodes, Axicon offers three versions of the S range.

Axicon 6025-S

2.6” maximum scan width (including quiet zones)

Ideal for GTIN12 and 13 and other smaller barcodes

Automatic aperture range from 3mil to 20 mil

Axicon 6525-S

4.9” maximum width (including quiet zones)

Ideal for Code 128 and other medium size barcodes

Automatic aperture range from 5 mil to 20 mil

Optional mount for thermal printer

Axicon 7025-S

7.6” maximum scan width (including quiet zones)

Ideal for GTIN14 and other very large barcodes

Automatic aperture range from 8 mil to 20 mil

Optional mount for thermal printer

These models and sizes reflect the scan widths and aperture ranges of the blue color “classic” Axicon verifiers that produce a single verification report with a push of the scan button. The recommended ten scan average is done by moving the verifier slightly, down the height of the barcode, taking 10 individual scans which the software is set to average into a final report. This is where the S range differs because it has a continuous scan mode.

A single push of the scan button with an S range verifier launches a scan sequence; the operator moves the verifier down (or up) through the entire height of the barcode.

The verifier captures up to 100 scans per second. The S range verifier tests and grades the full height of the barcode in a fraction of a second—hence a quicker, more complete and more accurate verification report.

Universal Mount

The 6525-S and 7025-S adapt to inline verification. The optional mount adjusts to virtually any thermal printer, and if the printer has an available I/O port, the verifier can pause the printer if a bad barcode or a no-ready is detected. The software can write verification reports to a csv file for a complete record of printer output.

The S range also makes verification of multi-row Databar symbols quick and easy. The user swipes the verifier down (or up) through all the rows and the software automatically stitches them all together and produces the verification report.

The continuous scanning capability is also very helpful when verifying extremely small barcodes, such as jewelry tags. “Finding” the barcode with an S range verifier is quick and simple. Position the verifier somewhere in proximity to the barcode, press the scan button and “wipe” the barcode through its field of view. The verifier will find and grade the barcode in one quick, simple sweep.

Other notable features of the S range:

No batteries! The S range communicates and takes power from its USB host

Portability is available with free Axicon Android software*

Costs very little more than the classic range

Standard 2 year warranty

Renewable annual maintenance contract available

Free software updates always available at www.Axicon.com

ISO compliant device comes with signed, dated Certificate of Compliance

Annual re-certification of ISO Compliance available at nominal cost

By John Nachtrieb 18/12/2018

Barcodes are everywhere. Thirty years ago, barcodes tracked inventories and enabled supply chains. In the ensuing decades, barcode use cases have exploded exponentially, and along with it, the importance of barcode quality.

Barcode verification is all about diligence: meticulously testing and confirming that the barcode performs correctly. Technology is a tool but does not replace diligence. For example, inline validation systems, many of which suggest they verify the barcodes, fail to provide the authentic quality testing they imply. More about this later.

New use cases bring barcodes into unanticipated, new scanning situations: surgical suites, outdoor activities, manufacturing, widely varied ambient lighting, airborne particulate, substrates and scanner types. Some newer barcodes are much smaller than in previously used. At the same time, newer scanners boast greater tolerance for poorly printed or damaged barcodes; ironically, the fuzzy logic involved also increases the frequency of misreads—apparently successful scans at the expense of accuracy.

High-speed inline validation systems are intrinsically not ISO compliant verifiers, although many of them can produce a report based on ISO parameters for barcode quality. The report is not what makes a scanner into a verifier: the testing instrument itself is what makes the verifier. To be an ISO compliant verifier, the device must demonstrate that it measures and grades to an international standard within published tolerances, under specific, controlled lighting conditions, scanning the barcode at a fixed, known distance and angle.

Compliant barcode verifiers eliminate the influence of ambient light; ten-scan averaging of 1D barcodes is recommended. All of the ISO parameters must be tested and graded; so-called “partial ISO grading” is simply not compliant.

Inline barcode validation is not without its merits but it does not replace verification. Knowing the grade for each ISO parameter and monitoring them over time is the only accurate way to determine authentic barcode quality—and that is the only way to predict barcode performance were it is used.

The successful scan of a barcode is not meaningful if “successful” means that data—some kind of data—was captured. Only an ISO compliant verifier can authoritatively decode and grade a barcode and predict future scanning success with virtually any type of scanner.

By John Nachtrieb 12/10/2018

For those who are uninformed, a chargeback is a fee or penalty imposed by a retailer for poorly performing barcodes on shipments received from vendors. Walmart is most often blamed—or credited—with inventing the chargeback, and that sounds plausible, since the Walmart business model built on logistics.

Although we do not know it for a fact, disgruntled Walmart suppliers have stated that the income stream from chargebacks is so considerable, Walmart no longer uses them to motivate vendors to improve their barcodes (if they ever did). Maybe so, but when barcodes don’t work right, logistics don’t work right and there is a cost to correcting the problem.

There is also a cost to preventing the problem, but whereas a chargeback will be an ongoing liability if the cause is not brought under control, the cost of prevention is more like an asset and it even has a return on investment. It is the chargeback pushback and it is called verification.

Consider the following actual case histories:

New Jersey clothing distributor saved $50,000 in prevented chargebacks with (2) verifiers that cost about 15% (at MSRP)

Medical device manufacturer was threatened with fines for non-compliance to FDA’s Unique Device Identification rule for non-working barcodes on hydration bags

Chinese garment manufacturer paid $20,000 in chargebacks on one 40 foot ocean freight container of shipments to stores

Big Box retailer saved $5000 the first week they had verifier on a sales demonstration—they had not even purchased yet

In all cases, the verifiers paid for themselves immediately and then contributed significantly to the bottom line in preventing future chargebacks.

These are not extreme or unusual scenarios. These are companies, large and small, who have learned by paying attention and not accepting the idea that mistakes are an inevitable and unavoidable cost of doing business.

By John Nachtrieb 23/07/2018

The verifier is reporting low grades on your barcodes, consistently in the D and F range, but your scanners read them without a problem. Furthermore, your customer of many months has never reported a problem. Which do you believe? What is the smart way to proceed?

Understand what the ISO barcode grading system means. For a deeper dive, have a look at a recent article we published about this. The abridged version: an ANSI A does not guarantee that your barcode will work everywhere. Some scanners may fail to decode them. Conversely, an ANSI F does not guarantee that your barcode will work nowhere. Some scanners will read them perfectly. The ISO specification provides a method for predicting barcode performance, based on a proven and measurable set of print quality parameters. It does not provide guarantees—but it does provide a compelling defense against barcode quality-related liability. All of which emphasizes the question—what do you believe, where is your defense and how do you manage the risk of bad barcodes?

An “F” Grade Barcode Is Not a Guarantee to Fail

Things change: scanners and vision systems, new printing technologies and advanced packaging materials are uncharted territory to the venerable ISO standard—that is also evolving, but at a slightly slower pace.  Direct printing on flexible materials and disinfectant-compatible laminates are but two examples of recent technological developments that sometimes challenge the established specifications and definitions of acceptability.

So, back to the original question: when do you ignore your barcode verifier? When non-negotiable requirements of your customer make it impossible to produce a barcode that achieves a passing ISO grade, proceed diligently:

Document the requirement—do not proceed with just a verbal mandate

Notify your customer of the ramifications

Diplomatically require them to absolve you of any liability–in writing

Only then, ignore your verifier.

By John Nachtrieb 18/06/2018

If you believe that barcode verification is important (not everyone does), your choices are simple:

a) Purchase a barcode verifier

b) Use a barcode test lab

Verifiers are expensive. There are legitimate product development costs and limited market size reasons for that. At the end of the day, buying a verifier is a significant investment justified by the ironic clash of a belief and a hope:

The belief the barcode verifier will pay for itself in the long run

The hope it will never be necessary

In actuality, this ironic clash only occurs if you have never paid a chargeback for bad barcodes or had a narrow escape. Those who have experienced a barcode problem understand the cost and value of risk management. In many cases the verifier pays for itself again and again.

One reason to use a barcode test lab is to avoid the high initial expense. Depending on the cost of testing services and how often you need them, it could take years for the cost of test services to equal the cost of a verifier.

Other reasons to use a barcode test lab include:

Outsourcing the cost of the personnel and training who uses the verifier

Access to equipment that is always calibrated and conforming to the latest specifications

Access to a wide range of equipment

A resource that can quickly identify and solve barcode problems

An authority/advocate who can vouch for barcode quality when scanning problems arise

An independent 3rd party whose quality assurance work is unbiased

A test lab may have a range of devices they could use to test your barcodes. Like test driving a car by renting it, you could familiarize yourself with a variety of barcode verifiers and their test reports as product research for ultimately buying a verifier.

An experienced barcode test lab can more quickly solve barcode mysteries that can vex users for hours and days, such as barcodes that simply do not scan. Experience can distinguish the simple and obvious to esoteric issues and get to the solution quickly.

There are some downsides to a test lab:

Time delay and expense of shipping samples to the test lab

Response time of the test lab

Test lab cost

It is possible to email high quality images, avoiding the cost and delay of shipping live samples of barcodes to the test lab. While this is not ideal, some critical attributes of the barcode can be legitimately tested and graded. These include symbology type, data structure, check digit and quiet zones.

Test lab costs vary by provider. Here are key questions to ask when looking for a barcode test lab:

Can you test my (name of barcode type)? If the lab has a very limited stable of equipment they are probably a very small and limited operation. This isn’t necessarily a bad sign but it could also mean their experience and limited. Point-blank ask them.

How many samples should I send? If the answer is vague, or “…as many as you want…” beware. A better answer is, “Send as many samples as accurately represent your print run. At very least we need to see samples from the beginning, the middle and the end.”

How do you charge? Is there a per-sample charge? A minimum or set-up charge? Are there any other charges? If I need to get my barcode samples back, is there an additional charge for that?

If I need a large quantity of barcodes tested, are there discounts? Yes, from Barcode test.

What is your response time? We respond with test data within about an hour of receiving incoming samples.

How do you report the test results? We attach a verification report for each symbol tested, and explain findings and recommendations in the covering email.

Is anything else besides the quality report included in your barcode testing services? From Barcode Test a telephone conference is always included.

By John Nachtrieb 08/06/2018

There is no shortage of myths and beliefs circulating about inline barcode verification. There are two ways to deconstruct this technology:

Regulatory: ISO and industry applications such as UDI

Practicalities from use cases

REGULATORY REQUIREMENTS

The ISO standards for 1D barcodes and 2D symbols (ISO15416 and 15415 respectively) specify the attributes of the barcode which are to be measured, how each parameter is graded, and how a final symbol grade is determine and reported. The same standard applies to inline and offline or spot- checking verifiers.

It is important to note that most of the ISO parameters address issues of reflectivity. This is because reflectivity—or more accurately, reflective differences—are now scanners capture and decode the data contained in the barcode. This is true for all types of scanners: wands, lasers, linear or area CCD arrays and digital camera scanners.

INDUSTRY APPLICATIONS

The UDI Final Rule is representative of many industry applications: it defines barcode types (symbologies) and unique data formatting (parsing) schemes. Industry applications also detail how where the barcode should appear on a product or package. For example, a 1D barcode on an airline bag tag is present in both ladder and picket fence orientation to ensure it will work even if a printer pixel is burned out.

PRACTICALITIES

Barcode verification is important because of the job barcodes perform, from maintaining retail inventories to tracking medical devices, securing pharmaceutical supply chains, assuring accurate drug dosing at bedside and logging maintenance history of critical aircraft engine parts. Because of these practicalities, it is essential that barcodes are verified in their final form—as they appear when they are scanned in the retail store, surgical theater or aircraft maintenance facility.

It is also important that barcode verification is done in similar conditions to where they are scanned. Because of the wide variety of user environments, it is important for verification to avoid adding an additional variable to the process. In other words, to normalize and control the verification environment. This includes elimination of excess ambient light. While this might not match the environment in which the barcode is ultimately used, it establishes a reliable benchmark to predict barcode performance. It cannot be reasonably anticipated everywhere a barcode might be used, so testing it in a uniform environment proves that the barcode is viable under normative conditions.

DEFENSIBLE PROCEDURES

The barcode on a medical device is not scanning right—the manufacture receives a chargeback that threatens not only the trading partner relationship but also the future of the company itself.

Scenario A: The medical device manufacturer “verifies” their barcodes with overhead, fixed mounted scanners running a verification software. They have archived millions of verification records showing passing grades on their barcodes. The installer “calibrated” the system several years ago using an offline verifier. Workspace lighting was replaced with high intensity LED’s. The inline system reads the barcodes so they must be compliant—but are they?

Customer has a problem with the barcodes on a lot or batch. Complains to the manufacturer who swears their barcodes are fine. Dissatisfied, the customer sues.

The case goes to court; the plaintiff’s expert witness points out that the inline verification system is not compliant because (a) there has been no regular recalibration and (b) variables such as ambient lighting influence the verification results.

The vendor doesn’t have a leg to stand on because “…but we’ve never had a problem…”is not a viable defense. They pay a 6 figure chargeback and lose the account.

Scenario B: The medical device manufacturer spot-checks barcodes with a certified, ISO conforming offline verifier. They have a verification report for the first and last, and additional reports from intervals during each print run.

Customer has a problem with the barcodes on a lot or batch. Complains to the manufacturer.

Changes to workspace lighting did not affect the verifier, which detected changes in the printing accuracy; continuous adjustments have maintained print accuracy. The offline verifier is recalibrated annually.

Customer is impressed and suspects problem could be with their scanners and has them checked against reference standard “golden” barcodes. Scanners are found to be out-of specification and are replaced.

By John Nachtrieb 31/05/2018

Q: What could be worse than a non-scanning barcode?

The answer might surprise you!

A: A barcode the scans the wrong information perfectly.

Think about it—it is a real train wreck. Here is one case history we personally witnessed.

A few years ago, we bought some vanilla yogurt along with about $150 of other stuff—a typical family grocery shopping trip. Nothing unusual happened at the register but I noticed that the receipt tape called out lemon yogurt. I thought nothing of it at the time. A few days later, back at home doing the dishes: the yogurt was eaten and the empty container was in my hand about to be discarded. Instead I brought it to the test lab. Sure enough, it scanned perfectly but the encoded information was different that the numbers beneath the barcode. Voila!

Next trip to the grocery store I sought out the store manager. Yes, there had been an inventory glitch. They kept ordering too much lemon yogurt but were chronically out of vanilla. It took them a while to figure it out; meanwhile, customers were not happy.

There are several ways this could happen. Sometimes it happens in the graphics design department. Software does not always prevent such errors when a highly creative person insets a barcode image.

A super-aggressive scanner can misread a marginal quality barcode, transposing or substituting a character or two. Barcode verification prevents marginal barcodes from causing transposition and misread problems.

A database entry error can cause an erroneous look-up at the scanner. Usually this causes a no-read but sometimes the product barcode will point to a different product in the item lookup.

This is a nightmare. That comforting “beep” at the checkout is really no comfort at all: the scanner recognized something—was it the right thing? It would be far better if the barcode failed altogether. That mesmerizing beep lulls us into a warm, comfortable acceptance that all is well, when actually a train wreck is unfolding before our very eyes.

Barcodes that misbehave in this way are difficult to detect. How can these problems be identified? Here are some ideas:

Guard databases diligently. Making central databases accessible to many users makes them vulnerable to mistakes. If this is necessary, designate a responsible party for maintaining data integrity and checking it regularly.

Minimize the use and distribution of multiple database copies. If this is necessary, sync them regularly to a known “golden” copy to avoid inaccuracies

Verify and validate product barcodes to the ”golden” database. Some barcode verifiers have product lookup capability to perform this confirmation.

Pay attention to apparent inventory anomalies such as spikes, drop-offs or unexplained supply chain errors. This could signal a barcode mis-identification

Be aware that the comforting “beep” only means that something happened. It does not necessarily signal what you think (or hope) happened.

Does this indict barcode technology and predict its demise in favor of some new, better identification strategy such as RFID or invisible barcodes? Actually not. Alternative auto ID technologies have the same reliance on databases, which is where these errors originate.

Vision-based product recognition could avoid these problems, but that solution is outweighed by the new problems it cannot solve, such as batch or lot based recalls, best-use-by and expiration dating. Barcodes are not perfect but they are still the best, most accurate and least expensive solution for what they do.

By John Nachtrieb

DPM or Direct Part Mark barcodes have been around for a long time, but this method for marking parts and sub-assemblies is evolving, as the use cases expand and advance. The military has long been a leader in using DPM to identify anything from engine blocks and major engine system such as exhaust, valve trains and intake manifolds. Aerospace was also an early adopter, marking jet engine components such as fan blades. More recently, medical device manufacturers have started marking implants and surgical instruments. As with conventional barcodes, marking parts provides a quick, simple and error –free identification method. Marking a part directly, on the part itself rather than on a tag or label, ensures that the identifying information cannot be lost or misapplied to a different part.

Linear DPM Barcodes

As the uses of DPM have evolved, so have marking methods. Early marking a was limited to photo-chemical etching, dot peen and engraving. Laser technology has made it possible to direct mark previously difficult to mark substrates including plastics and a wide variety of metals. And as in earlier use cases involving conventional barcode marking, standardizing DPM quality lags behind actual uses. In other words, some industries and applications are “pushing the envelope” into unknown and unregulated territory.  An example from the automotive industry is the use of 1D or linear barcodes directly marked onto parts and subassemblies. The current DPM specification applies only to 2D symbologies and most (if not all) currently available verifiers cannot test and grade 1D DPM barcodes.

Verification of DPM symbols is a very important aspect of using them due to the vital roles they perform:

Logging the maintenance history, failure rates and replacement scheduling of mission-critical machine parts in aircraft, military hardware and life-support equipment

Tracking pharmaceuticals in the supply chain to assure security prevent counterfeit infiltration

Recording implants in support of EMR (electronic medical records) accuracy and recalls

…and a broad spectrum of other applications.

Because of its wide range of uses, DPM presents unique challenges to verification. Verification depends on standards, which provide the basis for predicting ability to successfully scan and decode the barcode symbol. Unlike a barcode printed in black ink on a white label where the reflective differences are specified and grade, a DPM symbol has an unpredictable (and usually very low) amount of reflective difference between the symbol and the substrate. What is the minimum acceptable amount of reflective difference? How is the reflectance value of a roughcast or stainless steel or aluminum substrate measured?

DPM barcodes on a flat surface are the simplest use case. Parts to be marked can also be cylindrical, and sometimes very small. As with conventional barcoding, DPM scanning is line of sight. The entire symbol must be visible to the scanner even on a tiny laparoscopic device or orthopedic fastener.

By John Nachtrieb - 20/03/2018

Print quality used to be the most important consideration in barcode quality. Data formatting was something of a minor issue since UPC was almost the only show in town and the structure of a UPC symbol was simple:

The six digits on the left represented the brand owner. These number were assigned by GS1

The five digits on the right represent the item, assigned by the brand owner.

Add to these 11 digits a 12th digit, called a check digit, and there you have it—a Version A UPC.

A few other symbologies could have specific number sequencing rules. The US military was employing a numbering scheme for Code 39.  The automotive industry was also using Code 39 in the early 1980’s. In these and other early use-cases, data formatting was still quite basic. The advantages of bar coding made them attractive in new use-cases, but more data capacity was required. The development of 2D symbologies such as Data Matrix Code solved the data capacity problem. Expanded data capacity meant additional packets of different types of data, not just brand and item.

The UDI Use Case

UDI is not the only use case in which data format is important, but it is one of the most interesting because the medical device industry requires such a broad scope of data. The data formatting of the UDI symbol may seem illogical at first but the rationale becomes familiar and makes sense. The USI data formatting rules apply identically to 1D symbologies such as GS1-128 and 2D symbologies such as GS1-Data Matrix.

Each packet of data has its own unique prefix, called an Application Identifier or AI. GS1 refers to them as ‘flags’. Each prefix identifies a type of data, including whether the data is a fixed or variable length, and whether it includes numerical-only data or alphanumerical data. Fixed length data must follow a specified format.

Application Identifiers

AI’s are numerical-only characters and are 2, 3 or 4 digits long. In the human readable interpretation, parentheses bracket the AI’s but the parentheses are not encoded in the symbol data. This could cause confusion when a variable length field such as a lot, batch or serial number happens to terminate in a string of numerical characters that are the same as an AI. To prevent confusion the GS1 standard requires insertion of a FNC1 Group Separator at the end of each variable length AI.

It makes sense for all the variable length data to be consistently located toward the end of the data string and not inserted between packets of fixed length data. This seems illogical to new UDI users because the AI’s do not appear in numerical sequence. For example AI (17) Expiration Date will appear earlier in the data string than AI (10) Batch or Lot Number.

International Uncertainty and Verification Challenges

By John Nachtrieb 18/01/2018

If you are already familiar with 1D or linear barcodes such as UPC and Code 128, you will find many familiar things in 2D or matrix symbologies, but some of those familiar things apply in unfamiliar ways. Generally, 2D symbols work the same as 1D barcodes: the scanning technology is based on reflectivity or more precisely, the reflective differences between the dark and light reflectance elements in the symbol. Here is an explanation of some other similarities. 

The stop/start patterns in 1D barcodes signal the scanner to the presence of a barcode. Because different 1D symbols have different start/stop patterns, they also enable the scanner to identify kind of symbol it is—UPC, Code 39, or whatever. Start/stop patterns provide basic, format information that is always the same for every UPC, every Code 39 or code 128 or ITF. The “intelligent” or meaningful encoded information will vary but the start/stop pattern remains the same for each symbol type.

In 2D symbols these are called Fixed Patterns—they perform the same function as start/stop patterns in 1D codes with nuanced differences:

they orient the scanner to the symbol directionally—what is up and down, what is left and right so the data can be decoded in its proper sequence

they tell the scanner what type of symbol is present

they help the scanner determine X dimension so it can select the correct aperture

1D barcodes must have a blank area of a specified, minimum size leading and trailing the barcode. Quiet Zone dimension is expressed as a multiple of the X dimension; for example in UPC the minimum left and right quiet zones are each 9 times X.

2D symbols also have a quiet zone that completely surrounds the symbol on all four sides, but it is relatively small, and differs for different types of 2D symbols. The Quiet Zone surrounding a Data Matrix code must be equal at least 1X. The Quiet Zone for QR Codes intended for mobile scanning (smart phone) is >2X; for all other scanning is >

Quiet Zones as expressed above and in the ISO15415 Specification designate the required Quiet Zone as a > Larger Quiet Zones are always better. This is equally true of ISO15416 regarding 1D barcodes. 

Bar Width Reduction Bar width reduction is applied in a linear barcode to compensate for a predictable  amount of press or dot caused by the printing process. Pigment applied to a substrate tends to spread. Bar width reduction compensates for that spread so that bars gain into their nominal width rather than exceed the allowable bar width tolerance and degrade scanning performance.

Bar width reduction for 1D barcodes affects only bar width and not bar height since bar height is not critical to scanning success.

Press or dot gain also occurs in printing 2D symbols so bar width reduction is an important consideration here too, but since 2D symbols are two dimensional, BWR must also occur in both the X and Y axis.

Since printing (with some exceptions) is an in-motion process, transport speed can affect image quality in the direction of travel. 1D barcodes printed with the bars parallel to the direction of travel are somewhat less sensitive to gain than when oriented perpendicular to direction of travel. Similarly, 2D symbols are more sensitive to gain in the perpendicular axis. Bar width gain is also related to print speed. Rather than attempt to compensate for high transport speeds when printing 1D or 2D symbols, bar width control improves at constant, lower transport speed.

1D and 2D symbols with smaller X dimensions have smaller plus or minus width tolerances than symbols with larger X dimensions. Thus, accurate bar width reduction is more critical when printing smaller symbols.Error Detection vs. Error Correction   1D barcode check digits are like the spell checker in a word processing application: they detect an error but do not correct it. Check digits prevent misreads. Check digits are optional in some 1D barcodes, and when present, add only one or two characters to the encoded data string.

2D symbols have error correction capability, which is like the auto-correct function in word processing: it finds and corrects errors. Most 2D symbols use the Reed-Solomon error correction methodology.

Reed Solomon error correction originated in 1960, many years before barcodes. We unknowingly encounter it in mass storage systems including CD’s and DVD’s to correct for tracking errors by the reader (player).

QR Code has four levels of error correction, which is determined at the design stage; a QR Code can survive as little as 7% damage or as much as 40% damage.

Higher levels of error correction increases the data content of the QR Code, which can require it to occupy more space, or to reduce the X dimension. Data Matrix code error correction is user-scalable depending upon the symbol size and total number of codewords. 

X Dimension

The X dimension of a 1D barcode is the width of the narrow bar. In a 2D symbol the X dimension is called an element or module. In both symbol types, the X dimension is the basic building block of the symbol. In a 1D barcode the X dimension, in conjunction with the narrow-to-wide ratio of bars and spaces, and the amount of encoded data, determines the physical size of the entire barcode and its quiet zones. In UPC, X dimension is expressed as “magnification” but otherwise means the same thing—physical size of the symbol.

In 2D symbols, the element or module size and the amount of encoded data determines the physical size of the 2D symbol. The level of error correction factors into the amount of encoded data: more error correction increases the amount of encoded data.

Symbol Contrast    

In 1D barcodes, there are three factors to consider in symbol contrast.

First, there must be a minimum reflectance difference between the barcode and the background. The minimum reflectance difference for 1D barcodes is 50%.

Second, the background is always the Rmax maximum reflectance or “light” value; the barcode is the Rmin minimum reflectance or “dark” value. The Print Contrast Signal or PCS system specifies this reflectance requirement. Presently, reverse-imaged 1D barcodes with light reflectance bars and dark reflectance background (spaces and quiet zones) violate.

Third, 1D barcodes are always* scanned in 660nm light. Originally the light source was a laser but even with modern camera imagers, the light source emulates the 660nm laser. The red spectrum light source prohibits 1D barcodes printed in reddish colors, which would render the barcode invisible in red light. Neither can barcodes be printed over a green background, which would appear black in red light.

QR Code symbols are not held to the PCS system, and since the “scanners” are often smart phone cameras that do not use 660nm light. These symbols can be printed in a light color against a dark background and in a wide range of colors. Ironically this makes it more difficult to be sure the color combinations are acceptable, but the key issue is contrast: there must a minimum amount of contrast difference. However, the two reflectance values—light and dark—must be uniform. There must be no significant differences in the light reflectance value, and no significant differences in the dark reflectance value. The QR Code above violates that rule due to an ill-advised design decision.

Data Matrix symbols printed on labels and packaging are subject to the PCS system with Symbol Contrast grading based on the 50% minimum Symbol Contrast threshold.

Direct Part Mark (DPM) symbols are imaged directly onto substrates such as metal, plastic or glass, and are not held to the Symbol Contrast standard but are still verified (and scanned

) with 660nm light. The basic requirement is to achieve minimal or better contrast difference between the symbol and its substrate to be able to decode it, but no contrast threshold is expressed.

DPM verifiers are required to provide lighting at 90°,45° and 30° to optimize the likelihood of capturing a successful decode, from which it is then possible to evaluate and grade the symbol on its other properties.

By John Nachtrieb 18/12/2017

When a barcode verifier cannot scan a barcode, it can be a puzzling, frustrating situation. Without a verification report, how do you correct an unknown problem?

What about when the verifier reads the barcode perfectly but the scanner doesn’t? Now you are really in the dark! How do you deconstruct this situation and figure out what’s going on? Here are some points that can help you navigate that path of discovery.

First of all, what grade is the verifier reporting for the barcode? If it’s an ANSI F, all bets are off about the scanability of that barcode. While an F grade does not “guarantee” that the barcode will fail to scan in all circumstances and all possible scanners, it certainly indicates the likelihood of scanning problems. Likewise a D grade. The ISO standards for barcode verification are a method of predicting likely scanner performance based on several key print parameters. Some scanners are more forgiving than others, some are less forgiving for some of those parameters. Scanner performance can degrade over time. Rough handling or challenging environments can influence scanner performance. Just like life, there are a lot of moving parts and nothing is really nailed down permanently.

It can be confusing when a verifier fails a barcode—obviously it successfully scanned the barcode to fail it. If the verifier can scan the barcode, why would a scanner be unable to scan it? For very technical reasons, the

verifier is able to decode and calculate a grade even a very poor quality barcode that a scanner would likely be unable to decode. It is this ability that makes a verifier so expensive—and so important.

The ISO verification standards are a predictor of barcode performance in a system with myriad variables. At present, there is no better way of predicting whether or not a barcode will scan.

Consider an even worse situation, when a verifier grades a barcode a C or better, but the scanner cannot read it. What can cause that?

Consider this: barcode scanning is all about detecting the reflective differences between the barcode itself and the background. Think about how different scanner technologies detect the

Digital Camera Scanner

reflective differences:

A laser scanner draws a very thin, bright line across the barcode, and the reflective differences between the bars and spaces of the barcode are reflected back into the receptor in a very directional way.

CCD array and digital camera-based scanner are different than laser scanners. They both flood the barcode with a blast of relatively diffuse light. The reflective differences between the bars and spaces in a barcode can be less discernable, especially of the scanner is held point-blank range and/or at 90 degrees to the barcode.

This can be problematic if the RMax (light reflectance) substrate is very glossy, and even worse if the RMin (dark reflectance) barcode is printed in a glossy black. The diffuse, non-directional light can make it difficult for the scanner receptors to discern the reflective differences in the barcode and its substrate. Positioning the scanner at a more oblique angle may help.

Finally, it is possible the scanning difficulty isn’t caused by the scanner optics or the reflectivity of the barcode. Scanner programming could be the culprit. When setting up a scanning system, limiting the number of digits that can be accepted in the barcode data field can cause a perfectly legitimate, A grade barcode to not scan.

By John Nachtrieb 17/10/2017

Multiple times—and in multiple ways—we’ve addressed the issue of barcode verifying from a cost perspective. We have acknowledged how expensive barcode verifiers are, relative to barcode scanners. We have discussed the total cost of ownership of a barcode verifier which, like a car, has an initial outlay (price) followed by ongoing maintenance, which includes re-calibrations, re-certifications, the occasional factory repair and the eventual price of replacement.

Like many (I hope all) of you, my personal world is larger than barcode technology, although I love this industry and love the work I am privileged to do in it. My larger world is nourished in a number of ways, including Seth Godin’s excellent blog. His feed today speaks eloquently and honestly to our industry and the truly-believed but often confused objections sometimes expressed about verifying barcode quality.

Here is what Seth says in the October 16, 217 feed entitled Price vs. Cost:

Price is a simple number. How much money do I need to hand you to get this thing?

Cost is more relevant, more real and more complicated.

Cost is what I had to give up to get this. Cost is how much to feed it, take care of it, maintain it and troubleshoot it. Cost is my lack of focus and my cost of storage. Cost is the externalities, the effluent, the side effects.

Just about every time, cost matters more than price, and shopping for price is a trap.

In barcode technology generally and barcode quality in particular, cost is more relevant than price, although cost of ownership must be taken into consideration. For example there are devices available that are represented to be verifiers but are not certified, compliant test devices. What do the test reports produced by such devices mean? Are they authoritative?  To an ISO auditor, the answer to both questions is a conclusive “No”. But are they less expensive?

The question begs another question. If the reports from a non-compliant device would not be acceptable to an ISO auditor, and would not reliably signal a problem or the lack of a problem with a barcode, why is price even a consideration? That is akin to someone who has been diagnosed with a serious ailment shopping for a resource (a therapy, a medication, or a surgeon) that is not specifically qualified to address the issue—including claimed or cleverly misrepresented “solutions” that may be utterly unqualified. This is where cost and price get confused.

Even if the price of an un-certified test device is substantially less than the price of a certified device, the test results from which one would you rather rely upon in a contended chargeback situation with a valued, up-until-now loyal customer?

Then there is the extreme—the “verifier” isn’t even masquerading as a qualified device: it’s a scanner or, worst of all, a smart phone. This is the extreme not only of the mis-directed, confused price vs. cost point of view, but also the extreme of adding to the cost with time wasted pretending (or actually believing) to be doing something meaningful.

By John Nachtrieb 03/10/2017

Opposing sides of the inline versus offline verification debate sometimes try to make it sound like one or the other is unquestionably superior. It’s a bit like claiming that air is better than water when in fact both have their place—and neither is intrinsically better than the other.

The advantages of inline verification are self-evident. You get 100% automatic verification of every barcode exiting a printer or moving down a conveyor. If the barcodes are, for some reason, low or deteriorating quality, the automatic verification system can become a very demanding master if human intervention is required to restart the system or deal with an overflowing shunt area. Likewise it can become a major source of label and ribbon waste if it automatically (and with increasing frequency) backfeeds and reprints bad labels. The key to intelligent use of inline verification isn’t “set it and forget it” as is sometimes implied by the purveyors of these systems. Set it, forget it–and regret it. Verification is most effective in situations where it is employed to monitor and confirm the effectiveness of in-place quality systems—not as last line of defense in a system devoid of quality procedures and practices. This is equally true of inline as well as offline barcode verification.

Offline barcode verification can do some things inline verification usually cannot—or does not. Product lookup, for example, is more typically an offline barcode verification option.  In addition to reading and grading the barcode based on print quality and data structure, an offline verifier can also refer to a lookup table or simple database to match the barcode data to a known product, thus confirming that the right barcode is on the product or item. Of course this confirmation does not occur without human intervention—a person looking at the lookup item description on the verification report.

This brings up an important point shared by both inline and offline verification. Both types of barcode verification produce data in the form of a verification report—data that must be looked at and interpreted by a human in order to act on it. Things change as a process proceeds. In a thermal transfer printing process, for example, print head temperatures can increase over time. Increasing print head temperatures cause X dimension growth and print gain, which is reported as a degradation over time of the parameter Modulation. Neither inline nor offline verification systems do anything more than report this. The printing system does not detect it and compensate for it. An operator may detect that the inline system is starting to reject more and more labels as the print run proceeds, but without looking at the data, it is impossible to discern what adjustments are necessary. But if the inline system was put in place as part of a strategy to reduce or eliminate operator involvement, payroll savings may be offset by cost increases in label and pigment waste.