Barcode Verifier: Print Quality Grades and RFP for Thai Plants

“Our customer told us at receiving that our shipping labels could not be read. We were charged for relabeling and manual data entry. We scan every label with a handheld scanner before shipment and they all read, so we do not know what is wrong.” We often hear this from quality assurance and shipping managers at Japanese-owned factories in Thailand. Here is the short answer: “it read on our handheld” is not the same as “it meets the grade required by the standard.” Barcode verifier implementation comes down to 3 decisions: (1) which standard and which conditions you measure against; (2) whether you measure by sampling or measure 100% of labels; and (3) whether you can run the verifier as a trustworthy measuring instrument. And “correct data” and “label mix-ups,” which a verifier does not check, are stopped by a separate mechanism.

All amounts, label counts, headcounts, complaint counts, and payback years in this article are original estimates and assumptions (placeholder values) created for this article, based on the model factory described later. They are neither industry averages nor survey results. In particular, the prices of verifiers and inline verification are placeholder values, because no publicly available primary information could be found. Please read them as a “calculation template” and replace them with your own measured figures and quotations.

Why Barcode Verifier Implementation Is Now an Issue for Thai Factories

Investment in Thai manufacturing continues in electrical and electronics and in food processing. According to a press release published by the Thailand Board of Investment (BOI) on July 23, 2026, investment applications in the first half of 2026 numbered 1,299 projects worth about THB 1.47 trillion, up 37% year on year. By sector, electrical and electronics accounted for 179 projects worth about THB 120.2 billion, and agriculture and food processing for 131 projects worth about THB 61.4 billion. Every time a new line starts up, there are more printers issuing shipping labels and more customer reading equipment receiving those labels.

Automotive parts present a mixed picture. According to an announcement by the Federation of Thai Industries (FTI) reported by Capital.com (September 28, 2026), automobile production in August 2026 was 124,646 units, up 10.93% year on year, while exports fell 2.04%. Meanwhile, according to The Nation (August 1, 2026), the FTI lowered its 2026 annual production target from 1.5 million to 1.45 million units. When production volumes are sluggish, relabeling work and chargebacks caused by labels that cannot be read at customer receiving directly cut into profit.

For medical devices, labeling regulations are changing. According to commentary by the law firm Tilleke & Gibbins (June 17, 2026), the Thai Ministry of Public Health’s notification on medical device labeling (2025) was published in the Royal Gazette on December 22, 2025, and took effect on June 20, 2026. UDI labeling has been made mandatory for the first time for software as a medical device (SaMD) in risk classes 2-4. Please confirm the details of scope and requirements individually with the competent authority (Thai FDA) or experts.

In retail, the move to 2D codes is progressing. GS1 US describes “Sunrise 2027” as an initiative to help retail POS systems read and process 2D codes. However, GS1 US itself writes that this is not a legal mandate and that it is not necessary to implement every 2D capability by that date. Nor is it about abolishing 1D barcodes. As of 2024, GS1 is reported to have said that 2D code pilots were under way in 48 countries. As labels carrying both 1D and 2D codes increase, so does the number of symbols that need to be measured.

On the equipment side, a series of products is moving verification out of “the quality assurance room” and onto “the line.” According to a March 27, 2026 article in the trade magazine Food Processing, Omron announced the VHV5 inline verifier, which it says can verify 1D, 2D, and DPM codes at up to 1,000 codes per minute (manufacturer’s stated value). Separately, according to a June 17, 2026 article in Motion & Mechatronics, the company’s VHV5-SRV handles up to 1,200 items per minute and is promoted as a move from offline sampling inspection at the start and end of production to calibrated real-time verification (manufacturer’s stated value). On the offline verifier side, Axicon released version 9.1 of its verification software in June 2026, moving ahead with support for the new editions of the standards (see below).

What Is a Barcode Verifier: How It Differs from a Scanner, and What It Does Not Check

A scanner checks “whether it reads”; a verifier measures “how well it reads”

The GS1 General Specifications (Release 26.0, January 2026) state that “it is important to note the difference between scanners and verifiers.” A scanner is a machine that reads data. A verifier, on the other hand, is defined as a “measuring instrument” for determining whether a symbol can fulfil its role of carrying data and delivering it when required.

This difference matters a great deal on the shop floor. According to an explanation by Barcode Graphics, which provides verification services, the fact that a barcode read on 1 particular scanner does not mean that it can be read by all scanners and imagers. Reader capability differs by model and settings, so a label that reads on your own handheld may not read on the fixed scanner at the customer’s receiving gate or on the reading equipment at a logistics warehouse. The company explains that the only way to confirm that a barcode works across a variety of readers is print quality analysis (verification) based on the ANSI/ISO methods.

Note that GS1 recommends that verification results not be used “only as a tool for rejection at receiving.” The ISO/IEC 15416 and 15415 methodology is positioned as a foundation for improving reading performance. For the issuing factory too, a verifier is, before being “evidence to keep the customer from complaining,” a “tool to find out early where printing has started to deteriorate.”

What a verifier does not check

Installing a verifier does not make every shipping label problem go away. GS1 lists the following points to keep in mind when reading verification results:

• Many verifiers do not measure symbol height
• Without software that checks against a database, they cannot confirm whether the encoded content is correct
• They do not check whether the human readable interpretation (HRI) matches the barcode data
• Sampling verification cannot assure the quality of the whole lot beyond the statistical confidence limits that correspond to the sampling rate
• Even a symbol that is perfect at the time of manufacture can deteriorate in the supply chain through scratches, freezing, or moisture
• Operator error causes results to vary

The 2nd and 3rd points are especially important. A label with the “wrong part number” printed cleanly passes the verifier with the highest grade. The USDA Food Safety and Inspection Service (USDA-FSIS) summary for July 2026 describes a case in which a public health alert was issued because burrito cartons contained a different product, a misbranding that meant they contained an undeclared egg allergen (Powder & Bulk Solids, August 7, 2026). This was a mix-up between the label and the contents, not a barcode print quality problem. It is the kind of incident that verifier grading cannot prevent.

Stopping label mix-ups and data errors requires a mechanism that checks part number, lot, and quantity against master data at the time of issue, and a mechanism that reads the label after application and matches it against the shipping instruction. Master data checks at label issue are explained in detail in “Label printing systems,” and checking gates before shipment in “Defect outflow prevention and shipping gates.” Think of the roles as divided: the verifier protects “readability,” and the checking mechanism protects “correctness.”

Print Quality Standards: ISO/IEC 15416, 15415, and 29158, and Why Application Standards Take Precedence

The 3 verification method standards and their editions

The methods for measuring barcode print quality are set out in international standards. Because the text of ISO/IEC standards is paid content, the explanation here is based on public GS1 documents and on explanations and manuals from verifier manufacturers.

The ISO/IEC 15416 method for 1D symbols became an international standard in 2000, after the US ANSI standard (1990) and the European standard (1995). According to an explanation by Axicon, ISO/IEC 15415 for 2D symbols changed substantially in the 2024 edition: from integer grades to grades with 1 decimal place (from 5 levels to 41 levels), a change in how symbol contrast is calculated, the merging of modulation and reflectance margin, and the addition of a new parameter that looks at module growth and shrinkage. The company explains that some 2D codes that previously failed will now pass, and that many codes will receive the same or a higher grade. The 2025 edition of ISO/IEC 15416 for 1D symbols has also updated thresholds and the way grades are calculated.

When a standard is revised, the change is reflected through verifier software updates. In its release notes for linear software (June 3, 2025), Axicon writes that it updated the threshold algorithm and grade handling to follow ISO 15416:2025. It also states that its software version 9.1 in June 2026 includes changes to support the December 2024 edition of ISO/IEC 15415 and the 2025 revision of ISO/IEC 15416, and that the main change is an update of the GS1 plug-in to GS1 General Specifications v26.0. Older manufacturer documents may still show older edition years, so in the RFP, always confirm “which edition of the standard the software supports” and “the terms on which software updates are provided.”

The grade scale: 4-0 and A-F use the same thresholds

ISO/IEC verifiers do not report a simple pass or fail; they report results on a scale of 4 passing levels from 4 (best) down to 1, plus 0 (fail). The ANSI standard uses the letters A-D (pass) and F (fail), but the thresholds are the same: A = 4, B = 3, C = 2, D = 1, F = 0. GS1 states that 1.5 or above indicates acceptable reading performance.

For 1D symbols, print quality varies along the height of the symbol. For this reason, GS1 guidelines have a grade calculated for each of at least 10 scan lines at positions that divide the symbol height into 10 equal parts, and the arithmetic mean of those (to 1 decimal place) becomes the overall grade. Individual parameters, however, are not averaged; each is reported as “the lowest grade” among the 10 scan lines. As a result, a verification report may show individual parameters lower than the overall grade, but only the overall grade is used for the pass/fail decision. This easily confuses people looking at a verification report for the first time, so cover it in internal training.

Here is one example of parameter thresholds. According to an explanation in a Cognex manual, symbol contrast (the difference in reflectance between the lightest and darkest parts) is graded A = 70% or more, B = 55-70%, C = 40-55%, D = 20-40%, and F = less than 20%. However, this is an example based on an explanation of the earlier edition of the standard, and the 2024 and 2025 revisions may change thresholds and calculation methods. When creating internal criteria, please confirm with the edition supported by the verifier you use and the manufacturer’s documentation.

Application standards take precedence

Another important point is the “hierarchy” of standards. According to an explanation in an Omron verifier manual, there are 4 types of standards involved in verifying barcode print quality: (1) verification method standards (ISO/IEC 15416 and 15415); (2) verifier conformance standards (ISO/IEC 15426-1 and 15426-2); (3) application standards (GS1 General Specifications, MIL-STD-130, ATA Spec 2000, ISO 15394, etc.); and (4) symbology specifications. It explains that application standards always take precedence over ISO standards, and ISO/IEC 15415 and 15416 are used as they are only when no application standard specifies otherwise.

In other words, what you decide before choosing a verifier is “which application standard, and what minimum grade, your customers or industry require.” Depending on the factory, several standards may coexist: Japanese OEMs’ own label specifications, GS1, medical device UDI, and the requirements of export markets. For reference, regarding MIL-STD-130, Omron’s 2018 FAQ describes 3.0 or above under ISO/IEC 15415 as a guide for conformance. As this shows, the required grade differs widely depending on the application standard.

GS1 Barcode Quality Requirements: How to Read 1.5/06/660, and Conditions for 2D Codes and DPM

1.5/06/660 means “grade / aperture / wavelength”

GS1 writes minimum quality specifications in the form “g.g/aa/www.” g.g is the minimum overall grade (out of 4.0, to 1 decimal place), aa is the measuring aperture (the size of the spot of light used for measurement, in units of 1/1000 inch = mil), and www is the wavelength of the light source (nm).

For example, 1.5/06/660 means “an overall grade of 1.5 or above, measured with a 6 mil (about 0.15 mm) aperture at a wavelength of 660 nm.” Note that the aperture figure is in mil, not millimeters. Even for the same label, the grade is not necessarily the same if the measurement conditions differ. That is exactly why acceptance criteria should specify not just “the grade” but also “the conditions under which that grade was measured.”

Which figures are required depends on the application

In the quick reference table of GS1 General Specifications Release 26.0, the minimum quality for almost all symbols, including EAN/UPC, GS1-128, GS1 DataBar, GS1 DataMatrix, and GS1 QR Code, is 1.5 (C) at a wavelength of 660 nm. As an exception, thick ITF-14 symbols with an X-dimension (the width of the narrowest bar) of 0.635 mm or more are given 0.5 (D) and 20 mil. The aperture is determined by the symbol specification table for each application: for example, 1.5/06/660 for general retail POS, and 1.5/10/660 for general logistics such as GS1-128 and ITF-14. Write the acceptance criteria for shipping labels after confirming which symbol specification table the customer specifies. You are free to aim for a higher grade than the customer requires as an internal standard, but keep it clearly distinguished from what GS1 requires.

For 2D codes, use a larger X-dimension and an aperture of 80% of the X-dimension

2D codes have conditions that 1D codes do not. Considering the optical effects of image-based reading, GS1 requires that GS1 DataMatrix and GS1 QR Code symbols printed on labels be printed at about 1.5 times the X-dimension allowed for 1D symbols in the same application. It also states that quality should be measured with an aperture of 80% of the minimum X-dimension allowed for that application.

This tells you one important thing for the RFP: because the aperture is determined by the X-dimension, the correct measurement conditions cannot be set unless you give the verifier manufacturer “the minimum X-dimension of the target codes.” For shipping labels carrying both 1D and 2D codes, list the X-dimension and aperture for each. If you are considering moving to 2D codes, please also see “Migrating to GS1 Digital Link / QR.”

DPM (direct part marking) notation and lighting

DPM, which is marked directly onto the part surface by laser or dot peen, is measured differently from labels. According to information in a standards database, ISO/IEC 29158 is said to modify the ISO/IEC 15415 evaluation method for direct marking and to define alternative lighting conditions and new parameters. According to a Cognex manual, it uses parameters such as cell contrast (the difference in mean reflectance between light and dark elements, divided by the mean of the light elements), and in addition to the default 45° lighting from 4 directions, 30° low-angle lighting and 90° coaxial diffuse lighting can also be used. The difference from 15415 is that the image brightness is adjusted and the threshold is calculated from statistics of the brightness distribution.

In GS1’s symbol specification table for small medical and surgical instruments, the minimum DPM quality is written in a form such as “DPM1.5/04-12/650/(45Q|30Q|30T|30S|90).” The leading “DPM” indicates that the grade was obtained using the 29158 (AIM DPM) method rather than ISO/IEC 15415, and the final parentheses give the angle of incidence of the lighting. The wavelength for DPM is 650 nm, which differs from 660 nm for labels. Note that this table is an example for medical instruments, so do not apply it as is to DPM on automotive parts; confirm with the customer’s specifications.

A DPM verifier is basically one that can switch lighting. According to Omron’s 2018 FAQ, the company’s LVS-9585 DPM verifier can use red dome, white dome, and 30° low-angle white lighting in 1 unit, and in in-house tests, 30° low-angle lighting often gave better results for dot peen than the other lighting. The same FAQ also states that because no reference DPM parts exist, GS1 does not certify DPM verifiers (as of 2018). Selection of equipment on the marking side is explained in “How to choose an industrial laser marking machine.”

Offline Verification vs Inline Verification

Verifiers fall broadly into 2 types: the “offline (sampling) type,” which measures labels 1 at a time on a desktop, and the “inline type,” which is built into a printer or line and measures 100% of labels.

As GS1 points out, sampling verification cannot assure the whole lot beyond the statistical limits that correspond to the sampling rate. If printing deteriorates between one sample and the next, the labels printed until the next sample are shipped in an “unmeasured” state.

Here are some examples of inline products, presented as options available on the market. Omron’s LVS-7510 is a printer-integrated unit for the Zebra ZT600 series that, according to the company, performs 100% print quality inspection and barcode verification without slowing print speed, judging against pass thresholds set by the user under ISO/IEC 15415 and 15416 (manufacturer’s claim). The company says it also provides OCR/OCV for reading lot and expiry text, serial number matching, CSV logging of results, and a relay to stop the printer, and that it can be used in combination with an offline verifier. TSC Printronix Auto ID announced in January 2021 that its printer-integrated ODV-2D verifier prints a pattern on labels that do not meet the required quality to identify them, and produces a grading report for each print job. The company also explains that quality records made at the time of printing have been used to dispute barcode quality chargebacks. In 2023, Cognex announced the DataMan 475V, which verifies 1D, 2D, and DPM codes at production line speed, as a product that “takes quality assurance out of the lab and into production.”

As with the LVS-7510 mentioned above, some inline products are said to be usable in combination with an offline verifier. In the model estimate below, too, the inline 100% configuration adds 1 offline verifier as a reference. This is to confirm that the inline judgments have not drifted and to produce verification reports to present in customer audits. If you are also considering automating label application, “How to choose an automatic label applicator” may also be useful.

Operating the Verifier as a Trustworthy Measuring Instrument: Calibration Test Cards, Calibration Frequency, and Measuring in Final Form

A verifier is a “measuring instrument.” The values of a measuring instrument are trusted only when calibration and procedures are in place. Build an operation that lets you answer when a customer asks, “Can we trust the results from that verifier?”

Confirm with a calibration test card

Test methods and minimum accuracy for verifiers are set out in ISO/IEC 15426 (15426-1 for 1D and 15426-2 for 2D). GS1 states that every verifier should have its calibration conformance periodically confirmed against a traceable reference, and names the “Calibrated Conformance Standard Test Card” for this purpose. Cards exist for apertures of 6, 8, 10, and 20 mil, and cards for EAN/UPC, ITF, GS1-128, GS1 DataBar, and GS1 DataMatrix are available from GS1 member organizations and are traceable to the US NIST (Composite and GS1 QR Code are not covered).

GS1 cautions that a verifier merely displaying “calibration complete” with a reflectance patch can report success even if the verifier is defective or the patch is wrong. Its position is that using the test card correctly is the only way for trading partners to make measurements they can trust. In acceptance testing at installation as well, make it an acceptance condition that measuring the test card gives results within its nominal values.

Calibration frequency and records

The guideline that GS1 issued for the verification services of its member organizations (2015) calls for operation such as the following. It is not a legal obligation on factories, but it is a useful reference when deciding your internal operation.

• Recalibrate at the frequency recommended by the manufacturer. If there is no recommendation, at least once a month
• Also recalibrate after a long period of non-use, or when the environment, such as lighting, changes
• Always recalibrate when changing the scan head, the measuring aperture, or the scan width
• Test calibration conformance with a test card at least once a year
• Keep records of testing, calibration, and maintenance for at least 2 years

The order is to first check the manufacturer’s recommendation and, if there is none, decide your internal procedure using the GS1 approach as a guide.

Measure in final form, and measure translucent substrates twice

GS1 asks that symbols be verified in their final form wherever possible (including overlamination, packaging, and contents). Where that is not possible and the substrate lets light through, measure once over a dark surface and once over a light surface, and take the worse result. For transparent or translucent labels, glossy film, and labels applied to curved surfaces, the result measured on the liner may differ from the result after application. Write “in what state to measure” into the procedures for both acceptance testing and routine sampling.

Verification report format

The format of verification reports is easier to decide if you use the GS1 format as a template. The verification report template updated by GS1’s change notification GSCN 24-011, effective September 2024, has fields such as verifier model, verification software version, date of last calibration, the number of the symbol specification table applied, print quality grade, decoded data string, and conformance to placement rules. There is also a field to check whether, when there are multiple retail barcodes, the 2D code is within 50 mm of the center of the 1D POS barcode, but this is for reference. GS1 states that the template itself does not create requirements and does not guarantee reading performance.

Cost and ROI: Estimating Barcode Verifier Implementation at a Model Factory

All figures from here on are placeholder values set independently by this article. They are neither industry averages nor survey results. Because there is no publicly available primary information on the prices of verifiers and inline verification, the amounts are placeholder values. For actual decisions, please recalculate using quotations from multiple companies and your own track record.

Common assumptions (Model Factory F)

Configuration A: Sampling verification with an offline verifier (desktop, supporting ISO/IEC 15416 and 15415)

Initial investment is THB 450,000 for the verifier, THB 50,000 for calibration test cards and fixtures, and THB 100,000 for sampling procedures and training, for a total of THB 600,000. Annual operating cost for calibration card renewal and maintenance is set at THB 60,000 per year.

The benefits (assumed values) are as follows:

• Reading failure complaints: 24 → 12. Sampling catches deterioration early from dirty print heads, ribbon wrinkles, and changes of label stock. However, lots produced between samples remain at risk. The reduction of 12 complaints × THB 40,000 = THB 480,000
• Staff: the 2 people change roles to sampling verification and record keeping, so the reduction is 0

Annual net benefit is 480,000 − 60,000 = THB 420,000 per year, and the simple payback period is 600,000 ÷ 420,000 = about 1.4 years.

Configuration B: Inline (printer-integrated) 100% verification on 2 lines + 1 offline verifier as a reference + integration with the label printing system

Initial investment is 2 lines × THB 900,000 = THB 1,800,000 for inline verification, THB 450,000 for an offline verifier (as the calibration reference and for audit support), THB 50,000 for calibration test cards, THB 400,000 for label printing system and MES integration (storing verification results linked to lots), and THB 300,000 for installation, FAT/SAT, and training, for a total of THB 3,000,000. Annual operating cost for maintenance and calibration is set at THB 180,000 per year.

The benefits (assumed values) are calculated as replacing the benefits of Configuration A (they are not added to Configuration A’s benefits). Configuration B’s reduction of 21 complaints includes Configuration A’s 12.

• Reading failure complaints: 24 → 3. The remaining 3 are things that verification at the time of printing cannot stop, such as damage in transit and data errors. The reduction of 21 complaints × THB 40,000 = THB 840,000
• Staff: reading checks 2 → 1 person (1 person is kept to handle failed labels and exceptions). 1 person × THB 240,000 = THB 240,000

Total benefits are THB 1,080,000 per year, annual net benefit is 1,080,000 − 180,000 = THB 900,000 per year, and the simple payback period is 3,000,000 ÷ 900,000 = about 3.3 years.

Key point 1: A wins on payback years. The 5-year cumulative figures are equal; the difference appears from year 6 and in “records for every label”

The 5-year cumulative figures are A: 420,000 × 5 − 600,000 = 1,500,000 and B: 900,000 × 5 − 3,000,000 = 1,500,000, so they are equal.

Taking only Configuration B’s increment, the additional initial investment is 3,000,000 − 600,000 = THB 2,400,000, the additional annual net benefit is 900,000 − 420,000 = THB 480,000 per year, and the payback on the increment is 2,400,000 ÷ 480,000 = 5.0 years. That is why the cumulative figures are level at exactly year 5, and B pulls ahead only from year 6.

Looking at investment payback alone, starting with Configuration A is the rational choice. If there are reasons to choose Configuration B anyway, they are not payback years but one of the following:

1. The customer requires verification records for every label (lot-by-lot records are needed for audits or to dispute chargebacks)
2. The loss per complaint is large (key point 2 below)
3. There are many label types or lines, and sampling cannot keep up

Key point 2: What if the loss per complaint were THB 100,000 instead of 40,000?

The factor that most affects the conclusion of this estimate is the loss per complaint. Some customers only require the labor of relabeling, while others impose chargebacks involving line stoppages or require heavy labor for investigation reports and corrective actions. Let us recalculate at THB 100,000 per complaint.

For Configuration A, 12 complaints × 100,000 = 1,200,000, minus operating cost of 60,000, gives THB 1,140,000 per year, and payback is 600,000 ÷ 1,140,000 = about 0.5 years. For Configuration B, 21 complaints × 100,000 = 2,100,000, plus staff savings of 240,000, gives 2,340,000; minus operating cost of 180,000, this is THB 2,160,000 per year, and payback is 3,000,000 ÷ 2,160,000 = about 1.4 years. The increment pays back in 2,400,000 ÷ 1,020,000 = about 2.4 years, against an additional net benefit of 2,160,000 − 1,140,000 = THB 1,020,000 per year.

In other words, what decides between A and B is the measured value of “the loss per complaint.” In the 90-day plan below, the first 30 days are used to tally the number of reading failure complaints over the past 1 year and, per complaint, the relabeling labor, chargebacks, and investigation labor.

Note that, of the 3 complaints that remain even in Configuration B, “data errors” are not stopped by a verifier. Master data checks at issue and reading checks after application are treated as a separate mechanism and are not included in this estimate.

12 Items to Write in a Barcode Verifier RFP

These are the items you should write, at a minimum, in the RFP (request for proposal) when asking manufacturers or distributors for quotations.

1. Target symbols, standards, and editions: types of 1D and 2D codes, and which of ISO/IEC 15416 and 15415 (29158 for DPM) they are measured under
2. Customer application standards and acceptance criteria: the standards for each customer and industry, and acceptance criteria in g.g/aa/www form (grade, aperture, wavelength)
3. X-dimension, labels, and substrates: the minimum X-dimension of each code, label stock and ribbon brands, and whether labels are transparent, glossy, or on curved surfaces
4. Whether DPM is used, and lighting: marking method (laser, dot peen, etc.), part material and shape, and the types of lighting needed
5. Sampling or 100%, and line speed: for sampling, the frequency and number of labels; for 100%, the maximum print speed and label spacing
6. Handling of failures: which of marking failed labels, reprinting, stopping the printer, and isolating the lot is done automatically
7. Calibration test cards, calibration procedures, and certificates of conformance: the types of test cards supplied (aperture and symbol), calibration procedures, and certificates of conformance
8. Verification software version and support for new standards: support status for 15416:2025 and 15415:2024, and the terms and cost of providing software updates
9. Output of result data, linking to lots, and retention period: the verification report format (including the date of last calibration and software version), output such as CSV, and the retention period
10. Integration with data checks (master data checks): connection to the label printing system or MES, and whether OCR/OCV or serial number matching is needed
11. Local support and calibration contact: the calibration and repair contact within Thailand, the route for repurchasing test cards, and support in Thai
12. FAT/SAT criteria: test items, samples used, acceptance criteria, and the scope of witnessing

Items 2 and 8 are especially important. If item 2 is left at “it reads,” the proposals from different companies cannot be compared on the same basis. If item 8 is not confirmed, you will not be able to explain the cause when results measured by the customer under a new edition of the standard disagree with your own. If you are also thinking about tracking at the serial number level, please see “Serial number traceability.”

What to Check at FAT/SAT

In the factory acceptance test before shipment (FAT) and the on-site test after installation (SAT), confirm with your own labels, not catalog values, that the system “measures correctly, stops failures, and keeps records.”

• Calibration check with a calibration test card: measure the test card and confirm that the results fall within its nominal values
• Repeatability with samples of known grade: prepare good, borderline, and bad samples and see whether the expected grades are produced
• Repeated measurements and variation between operators: measure the same label multiple times with multiple staff, and record the variation in results
• Detection and handling of failures with deliberately degraded labels: using labels that reproduce dirty heads or ribbon wrinkles, confirm that failures are judged and that handling such as marking or stopping works
• Missed labels at line speed (for 100% verification): run at the maximum print speed and count whether any labels go unverified
• For DPM, use real parts and real lighting: measure on actual parts, not test pieces, while changing lighting conditions
• Comparison with third-party results: compare results from services such as GS1 Thailand’s verification service with the results from your own verifier
• Record output and linking to lots: check whether results are saved linked to the correct lot and time and can be output as reports

The overall process of ordering and accepting inspection equipment is explained in detail in “Ordering and acceptance (FAT/SAT) of inspection equipment in Thailand.”

Issues Specific to Thailand and ASEAN

Use GS1 Thailand’s verification service

In Thailand, GS1 Thailand (สถาบันรหัสสากล) provides a barcode quality verification service. According to the service page updated in March 2026 (at the time of research), it covers 1D (GTIN-8/12/13/14, GS1-128) and 2D (Data Matrix, QR codes for GS1 Digital Link, GS1 QR Code) symbols, and accepts samples brought in person or sent by post, either as actual items printed at actual size and in actual color or as artwork. Standard turnaround is 5 business days, and express is 1 business day. For non-members, the fee is THB 700 per report for 1D and THB 2,000 for 2D, and members have a free allowance depending on their annual membership fee. These are service fees, not verifier prices.

You can use it to learn how your current labels actually perform before installing your own verifier, or to compare your own results with third-party results after installation. However, the page does not state the verifier models used or the editions of the standards followed, so please confirm the measurement conditions when making comparisons.

List the requirements of each customer and industry

In automotive parts, Japanese OEMs and Tier 1 suppliers often have their own label specifications, and for North America there are also AIAG standards. According to reports in trade media, AIAG B-17, a guideline for 2D DPM, is said to incorporate the DPM quality guideline developed by AIM. Please confirm minimum grade values in the customer’s specifications, not in articles explaining the standards. Attaching a list of the requirements of each customer, industry, and export market to the RFP prevents mismatches in manufacturers’ proposals. The overall approach to automotive parts traceability is organized in “Automotive parts traceability (IATF 16949).”

Heat, humidity, and dust

Thai factories are hot and humid, and some processes are dusty. In general, the storage conditions of thermal transfer ribbons and label stock may affect print quality, so include storage area management in your procedures. Also write into the RFP and procedures that, when the brand of label stock or ribbon is changed, the labels are re-verified with the verifier before and after the change. Even when the purchasing department switches brands to cut costs, set up a flow that passes through re-verification by the quality assurance department.

Medical device UDI

As mentioned above, under the Thai Ministry of Public Health notification (2025), UDI labeling has been required for SaMD in risk classes 2-4 since June 20, 2026 (labeling of registered SaMD that conformed to the 2020 notification is said to be usable for up to 2 years from the effective date). The commentary does not state requirements for UDI barcode formats or print quality grades. Please confirm the devices covered and the specific requirements individually with the competent authority (Thai FDA) or experts. Omron’s FAQ (2018) explains that in the United States, the FDA’s UDI requirements increased direct marking on devices that are reused and cleaned or sterilized, and thus increased the use of DPM. For an overview of UDI compliance for medical devices, please see “Medical device UDI traceability.”

Maintenance support

Check whether calibration test cards have an expiry date, and the route and timing for repurchasing them. Before ordering, also confirm whether calibration and repair of the verifier are available within Thailand, and whether the distributor can support on-site staff in Thai. Also write into the procedures whether, when the verifier breaks down, you fall back to handheld checks or supplement with external verification such as GS1 Thailand’s.

Equipment procurement in BOI projects

For projects receiving BOI investment promotion, please confirm the treatment of procuring or changing verifiers or inline verification individually with the competent authority (BOI) or experts.

Examples of options on the market

Manufacturers handling verifiers and inline verification include, for example, Axicon (the 6000 series linear verifiers and their successor for retail POS, the 6100-S, plus camera-based verifiers and verification software), Omron (the LVS-95XX series desktop verifiers and DPM models, the LVS-7510 built into Zebra printers, and the VHV5 and VHV5-SRV for inline verification), Cognex (the DataMan 475V for inline verification), TSC Printronix Auto ID (the printer-integrated ODV-2D), and Keyence (verification functions in code readers). There are other products as well. They are listed here only as examples and are not a recommendation or an evaluation of relative merit.

90-Day Plan

Days 0-30: Tally the number of reading failure complaints and the loss per complaint, and list customer requirements

Tally the past 1 year of reading failure complaints by customer, label type, and cause (fading, smearing, data errors, application position, damage in transit, etc.). At the same time, gather the relabeling labor, chargeback amounts, and investigation and reporting labor per complaint, and replace the model estimate’s “loss per complaint” with your own figure. In parallel, list each customer’s application standard and acceptance criteria (g.g/aa/www), together with label types and X-dimensions.

Days 31-60: Sample verification of current labels and isolating causes

Measure your current labels using GS1 Thailand’s verification service or a verifier manufacturer’s demo unit. Looking at which parameters pull the grade down gives you a lead on the cause, such as printer settings, dirty heads, the combination of ribbon and label stock, or an insufficient X-dimension. At this stage it also becomes clear how many complaints are caused by data errors or mix-ups. Those should be stopped by a checking mechanism, not by a verifier.

Days 61-90: RFP, FAT/SAT criteria, and order decision (A or B)

Based on the tally and the sample verification results, finalize the 12 RFP items and the FAT/SAT test items and acceptance criteria, and obtain quotations from multiple companies under the same conditions. Use your own version of the estimate to decide whether to start with Configuration A (sampling) or go as far as Configuration B (100% + reference unit).

The 4 deliverables you want to have in hand at the end of the 90 days are: (1) a tally of reading failure complaints classified by cause, and your own version of the investment estimate reflecting the loss per complaint; (2) a list of acceptance criteria for each customer and industry; (3) verification results for current labels and the isolation of causes; and (4) the RFP and FAT/SAT criteria.

Frequently Asked Questions

What is the difference between a barcode verifier and a barcode reader (handheld scanner)?

A barcode reader is a machine that checks “whether it can read” the data, while a verifier is a measuring instrument that measures print quality as “how well it reads” using the methods in the standards. Even if a barcode reads on 1 reader, it will not necessarily read on the customer’s reading equipment. A verifier produces a grade using methods such as ISO/IEC 15416 and 15415, and judges pass or fail against the conditions required by the customer or GS1. However, a verifier does not check whether the data is correct or whether it matches the HRI, so a separate checking mechanism is needed.

What is the difference between ISO/IEC 15416 and 15415?

15416 is the standard defining how to measure the print quality of 1D (linear) barcodes, and 15415 does the same for 2D codes. 1D symbols are evaluated from changes in reflectance along scan lines, and 2D symbols from images. A 2025 edition of 15416 and a 2024 edition of 15415 have been issued, updating matters such as how grades are calculated. For DPM marked directly onto parts, ISO/IEC 29158 (said to be the 2020 edition), which modifies 15415 for direct marking, is used.

What are the GS1 barcode quality requirements (what does 1.5/06/660 mean)?

It means “an overall grade of 1.5 or above, measured with a 6 mil (about 0.15 mm) aperture at a wavelength of 660 nm.” For many GS1 symbols the minimum grade is 1.5 (C), but the aperture is determined by the symbol specification table for each application. For 2D codes, it is said that the aperture should be 80% of the minimum X-dimension for that application. Please confirm with the customer which table applies.

How should DPM (direct part marking) be verified?

Measure with a DPM-capable verifier that can switch lighting, using the ISO/IEC 29158 (AIM DPM) method. In GS1’s table for medical instruments, “DPM” is placed before the grade, and the lighting angle is also stated. The wavelength is 650 nm. The suitable lighting changes with the marking method, such as dot peen, so carry out acceptance tests with real parts and real lighting. Please confirm acceptance criteria for automotive parts in the customer’s specifications. For requirements involving regulation, such as medical device UDI, please confirm individually with the competent authority (Thai FDA) or experts.

How often does a verifier need to be calibrated?

First follow the manufacturer’s recommendation. The guideline GS1 issued for the verification services of its member organizations calls for recalibration at least once a month if there is no recommendation, recalibration when changing the aperture and similar settings, testing with the calibrated conformance test card at least once a year, and keeping records for at least 2 years. It is not a legal obligation on factories, but it serves as a guide for internal procedures.

What does a barcode verifier cost, and what is the payback period?

Because there is no publicly available primary information on verifier prices, the model estimate in this article was calculated entirely with placeholder values. Configuration A, offline sampling, had an initial investment of THB 600,000 and a simple payback of about 1.4 years; Configuration B, inline 100% plus a reference unit, had THB 3,000,000 and about 3.3 years; and the 5-year cumulative figure was THB 1,500,000 for both. If the loss per complaint is THB 100,000, A pays back in about 0.5 years and B in about 1.4 years. Please recalculate with your own complaint counts and losses and with quotations from multiple companies. For the treatment of equipment procurement in BOI projects, please confirm individually with the competent authority or experts.

Summary

• “It read on our handheld” does not mean “it meets the grade required by the standard.” A verifier is a measuring instrument that measures grades using the ISO/IEC 15416, 15415, and 29158 methods, and pass or fail is decided by the conditions of the customer’s or GS1’s application standard.
• Write acceptance criteria as a set of “grade, aperture, and wavelength.” The 06 in 1.5/06/660 is 6 mil (about 0.15 mm). For 2D codes, use a larger X-dimension and measure with an aperture of 80% of the X-dimension.
• A verifier does not check whether the data is correct, whether it matches the HRI, or whether labels have been mixed up. These are stopped by master data checks and reading checks after application.
• Only when calibration test cards, calibration frequency, measurement in final form, and the verification report format are all in place does it become “trustworthy measurement.” Also confirm in the RFP whether the verification software supports the new editions of the standards.
• In the model estimate (placeholder values), offline sampling had the advantage in payback years, and the 5-year cumulative figures were equal. The reasons to choose inline 100% verification are a requirement for records of every label, a large loss per complaint, and a large number of label types. The key to the decision is your own measured “loss per complaint.”

TOMAS TECH supports Japanese-owned factories in Thailand from the preparation stage, such as tallying reading failure complaints, organizing customer requirements, and arranging sample verification of current labels, through to RFP drafting, attendance at FAT/SAT, and integration with label printing systems. Even if you are at the stage of “first wanting to understand how much loss our current complaints are causing,” please feel free to contact us through our contact form.

References

• GS1, “GS1 General Specifications Standard, Release 26.0” (January 2026): https://www.gs1.org/docs/barcodes/GS1_General_Specifications.pdf
• GS1, “1D Barcode Verification Process Implementation Guideline, Release 24.1” (July 2015): https://www.gs1.org/docs/barcodes/GS1_Bar_Code_Verification.pdf
• GS1, “GSCN 24-011 Verification template update” (effective September 2024): https://www.gs1.org/docs/barcodes/GSCN-24-011-2DVerification.pdf
• Barcode Graphics, “Barcode Scanners vs Verifiers”: https://www.barcode.graphics/barcode-scanners-vs-verifiers/
• Cognex In-Sight manual, “ISO/IEC 15416”: https://docs.cognex.com/is_630/web/EN/ise/Content/Reference/iso-iec-15416.htm
• Cognex DataMan 475V manual, “ISO/IEC 15415”: https://docs.cognex.com/dmst_619/web/EN/DM475V_Manual/Content/Topics/DM475V/TruCheck/ISO_IEC15415.htm
• Cognex DataMan 475V manual, “AIM DPM”: https://docs.cognex.com/dmst_617/web/EN/DM475V_Manual/Content/Topics/DM475V/TruCheck/AIM_DPM.htm
• Axicon, “Release notes (Linear)”: https://axicon.com/release-notes-linear
• Axicon, “News”: https://axicon.com/news
• Axicon, “ISO/IEC 15415 2D”: https://axicon.com/iso-2d
• Standards Council of Canada standards database (ISO/IEC 29158): https://scc-ccn.ca/standardsdb/standards/2047253
• Omron, “Appendix B: LVS-95XX Understanding the Standards”: https://files.omron.eu/downloads/latest/manual/en/appendix_b_lvs%C2%AE_95xx_understanding_the_standards_users_manual_en.pdf
• Omron, “LVS-9580 and LVS-9585 DPM FAQs” (July 2018): https://assets.omron.com/m/3695b3632a123eae/original/LVS-9580-and-LVS-9585-DPM-FAQs.pdf
• Omron, “LVS-7510 Series Brochure”: https://assets.omron.com/m/8ec783a6e094ca9/original/LVS-7510-Series-Brochure.pdf
• TSC Printronix Auto ID (ODV-2D, January 19, 2021): https://apac.tscprinters.com/en/news/tsc-printronix-auto-id-strengthens-odv-2d-inline-barcode-verifier-portfolio-adding-support
• Automate UK (Cognex DataMan 475V, October 16, 2023): https://www.automate-uk.com/news-publications/100-accurate-barcode-verification-from-cognex/
• Food Processing (Omron VHV5, March 27, 2026): https://www.fponthenet.net/article/221212/Inline-verifier-for-real-time-barcode-quality-control.aspx
• Motion & Mechatronics (Omron VHV5-SRV, June 17, 2026): https://www.motionandmechatronics.ca/product-news/omron-vhv5-srv-barcode/
• GS1 US, “Sunrise 2027”: https://www.gs1us.org/industries-and-insights/by-topic/sunrise-2027
• Keyence, “GS1 Sunrise 2027 Explained”: https://www.keyence.com/blog/gs1-sunrise-2027-explained-requirements-timeline-and-preparation.jsp
• Dynamsoft, “GS1 Sunrise 2027: Transition to 2D Barcodes”: https://www.dynamsoft.com/blog/insights/gs1-sunrise-2027-transition-to-2d-barcodes-for-global-commerce/
• Quality Magazine (AIAG B-17): https://qualitymag.com/articles/86738-cognex-verifiers-ensure-compliance-with-aiag-quality-guideline
• GS1 Thailand, “Barcode Verification”: https://gs1th.org/service-aidc/
• BOI press release No.117/2569 (July 23, 2026): https://www.boi.go.th/upload/content/PR117_2569EN.pdf
• Capital.com (Thailand’s August 2026 car production, September 28, 2026): https://capital.com/en-gb/news/thailand-s-august-car-production-rises-10-9-on-year
• The Nation Thailand (FTI’s 2026 production target, August 1, 2026): https://www.nationthailand.com/business/manufacturing/40069285
• Tilleke & Gibbins, “Thailand Introduces UDI Labeling Requirements for Software as a Medical Device” (June 17, 2026): https://www.tilleke.com/insights/thailand-introduces-udi-labeling-requirements-for-software-as-a-medical-device/
• Powder & Bulk Solids, “USDA recall update for July 2026” (August 7, 2026): https://www.powderbulksolids.com/product-recalls/usda-recall-update-for-july-2026