What Is Bar Code Verification and Why Should You Care?

The bar code, introduced in 1975, has become ubiquitous. It is hard to find a manufactured product today that does not have a bar code associated with it. The bar code is, arguably, the world's least expensive and most reliable method of entering data. It is a part of our lives, but sometimes it is a troublesome part.

Use of the bar code has gone through three distinct phases. In the first phase (1975-1990), bar codes were printed but not widely read. This was a necessary phase in the development of the industry — labels had to have bar codes before it made sense for the user to invest in bar code equipment and systems. Some applications (retail) adopted bar codes sooner than others. During the second phase (1990-1998), bar codes were a productivity enhancement tool — the back-up systems (people) were still in place. We are now entering into the third phase — the bar code is becoming enterprise critical. Highly automated operations depend on bar codes being correct and readable every time.

An example of the second phase is the UPC code usage in the local supermarket. We are all familiar with the checkout person manually entering the code when the bar code reader fails. The checker is the "back-up system." This is an obvious productivity bottleneck and requires checkers with a higher level of training. Human errors can happen, and overuse injuries are an issue. Fully automated checkout systems are being developed — the third phase. Obviously, the read rate has to be greatly improved from the current state for this to be feasible.

Third phase applications are found in automated material handling. Large automated warehouses are designed to handle one million packages per day. The United States Post Office is touting automated handling on primetime television commercials.

A clear and quantifiable example of the penalties associated with failure to make the transition to the third phase is familiar to supplier to large retail store chains (e.g., Wal-Mart, K-Mart). Shipments to these customers with unreadable bar codes are rejected, and the supplier is fined. At least $27M in fines have been collected since the beginning of these programs.

Third-phase industry segments require bar codes that are readable every time. Second phase industry segments suffer productivity degradation. Bad bar codes cost money, but they are not catastrophic to business.

Two fundamental approaches to 100% readable bar codes suggest themselves:

1. Print perfect bar codes, or
2. Check the bar codes after printing.

Printer manufacturer's protestations notwithstanding, there are no perfect printing technologies. Thermal printers experience ribbon wrinkles and burned elements in the print head. Improper heat and speed settings can result in wide or narrow bar code elements, or voids. Impact printers can have misaligned or failed hammers, or simply depleted ribbons. Ribbon wrinkles occur in impact printers, but are uncommon as compared to the thermal technologies.

While there are no perfect printing technologies, today's printers can be very good indeed. A quality unit, which is well maintained, prints good bar codes almost all of the time. To some extent, however, this makes the problem worse. Visual inspection of the printed output as a means of checking quality is unattractive, since the vast majority of the bar codes are good. As a result, the human inspector rapidly loses interest.

In today's business environment, ISO9000 certification is often a requirement. Obtaining and maintaining ISO9000 certification requires a competent corrective action system. Corrective action is defined as the action taken to assure that the problem being addressed never occurs again. Replacing a worn out ribbon or burned out print head is not an effective corrective action. Ribbons will wear out or wrinkle, and print heads will burn out again, resulting in the production of bad bar codes.

In phase-three applications, or effective ISO9000 corrective action systems, an automated read-after-write of the bar code is indicated. This can be done with a conventional bar code scanner, or a bar code verifier.

Scanners and verifiers, while superficially similar, differ fundamentally. The scanner is a mass production item. The measures of merit of a scanner are price and performance. Performance being measured by how good the scanner is at reading bad codes and digging the necessary data out of the noise.

While using a scanner to verify a bar code is certainly preferable to doing nothing, it has an inherent problem. The scanner is a go/go-go device that will only assure that the code has some region on it that is readable by that scanner. Codes don't get better in shipment. The better the scanner, the less "assurance" is obtained. The best scanner for this purpose is actually one that isn't very good at reading bad bar codes. Approaching the typical scanner salesman with a request for a predictably "bad scanner" is likely to result in at least a moment or two of surprised silence.

A bar code verifier's function is to predict how well the bar code will be read by the typical scanner. There are a number of verification methods and implementations. The two primary verification methods are Traditional and ANSI. The two primary implementations are handheld or portable units, and online units. Handheld units are typically quality department or laboratory instruments. Used for bar code design checking, trouble shooting or bar code quality sampling, they are not typically suited for read after write requirements. Results are displayed to the operator through LED's and displays, control of the unit is typically through a limited function keyboard. Serial interfaces to printers and for data collection are often provided.

Online verifiers are packaged to attach to a printer or onto a fixed-mount device in order to scan bar codes on a production line. The online verifier may be configured to stop either the printer or the production line if a bad bar code is detected. It does so via serial communications with the system controlling the printer.

The primary verification methods are Traditional and ANSI. Both methods are available in handheld units, ANSI verifiers are only now becoming available for online applications. Traditional verification traces its beginning to micrometer and optical comparator usage in the early days of the bar code industry. Typical parameters analyzed are Average Bar Deviation (how wide/narrow are the bar code elements as compared to the specification), Print Contrast, and Absolute Element Widths.

ANSI verification refers to the ANSI X3.182-1990 standard. The objective of ANSI verification is to predict how well a scanner incorporating a particular optical configuration (wavelength and aperture) will read the symbol. The ANSI method analyzes the analog reflectance profile of the bar code and calculates a number of parameters on the profile. Each parameter is graded (A, B, C, D, or F) and the overall symbol grade is the lowest grade of any parameter. A grade "C" or better should be readable by the typical scanner.

The ANSI method is the best documented and most traceable verification method. Multiple-scan averaging (recommended by ANSI) provides a topography check on the symbol, guarding against the symbol's having only a small readable region.

If your customers don't use the bar codes you are printing, not to worry. If they are phase-two customers, bad bar codes cost them money. But, if they are phase-three customers, then bad bar codes may cost you money. The most certain method of assuring good bar codes is verification.

Richard P. McClellan, Ph.D.
President and CEO
RJS Systems International