Calibration Interval Review: Prepare the Equipment Evidence

Define the review question and authority

State whether the organization is establishing an initial basis, reviewing an existing interval or responding to a change or unexpected result. These are different questions and may require different evidence. Identify the instrument or defined equipment group and the requirements that govern the decision before collecting data indiscriminately.

Name the person or role authorized to conduct the technical assessment and approve the result. The coordinator gathers and organizes evidence, but should not infer an extension from a sequence of pass labels. Document the review scope and any constraints imposed by applicable procedures, manufacturer information or contractual requirements.

ILAC G24 provides guidance on selecting and reviewing recalibration intervals, including consideration of equipment history and use. It does not supply a universal interval for every NDT instrument. The workflow here supports an authorized review; it is not a replacement for the guidance, applicable requirements or technical judgment. ILAC G24

Build a complete event population

List the relevant calibration and service events for the defined period or review scope. Include the source report number and issue, not just a date copied from the current register. The latest due date does not reconstruct the sequence of previous events, particularly when reports were amended or dates were corrected.

Identify missing records and unresolved associations. A history containing only easily retrieved successful reports can give a misleading picture. Preserve references to events that require recovery and ask the reviewer how those gaps affect the assessment. Do not invent an expected event simply because the equipment normally followed a particular schedule.

For grouped equipment, define the membership and changes over time. Instruments introduced later, retired after problems or used under different conditions may require separate consideration. The coordinator should make those differences visible rather than treating every item with the same model name as an automatically comparable population.

Preserve incoming and final conditions distinctly

Identify whether reported results describe the incoming condition, the condition after adjustment or another stated stage. Keep the supplier's terminology and explanation with the extracted reference. A final acceptable result can be important, but it does not necessarily describe the condition in which the instrument arrived for service.

Link repairs and adjustments to the relevant calibration event. Record the sequence when the supplier provides it and flag ambiguity when it does not. The reviewer needs to know whether a change occurred before the results being compared. A service invoice and certificate issued close together should not be assumed to have a particular order without evidence.

If incoming information was not requested or supplied, document that limitation. Do not infer it from the final result or create a synthetic pre-adjustment value. The evidence package should make the missing information obvious so the reviewer can choose an appropriate response rather than rely on a deceptively complete table.

Check whether the reported quantities are comparable

Ask the technical reviewer to define the functions, ranges or points relevant to the assessment. A report that covers a different scope may still be useful, but its values should not be placed into one trend series merely because the column headings look similar. Preserve the original scope reference for each event.

Keep units, conditions and reported uncertainty connected to the source values where data are extracted. Avoid copying only a bare number into a spreadsheet. The coordinator should not perform undocumented conversions or normalize results without the approved method. If a transformation is authorized, retain its basis and the original value.

Identify changes in report format, laboratory, method or decision rule that may affect interpretation. These do not automatically make records unusable, but they are questions for the reviewer. A clean chart should not conceal a discontinuity that changes what the plotted information means.

Gather actual use and handling evidence

Provide the available records of equipment use, assignments, service demands and relevant handling or transport events. Distinguish measured usage, logged activity and staff recollection. Each can contribute context, but they should not be represented as equally precise data. An absence of entries may indicate missing logging rather than absence of use.

Record known changes in duty or environment at the level supported by the evidence. A move from occasional shop use to frequent field assignments is a useful fact for review if documented. Do not invent hours, severity scores or exposure values to populate a worksheet when the organization never measured them.

The reviewer decides which contextual factors matter to the method and equipment. The coordinator should preserve relevant source references and uncertainty rather than making an unsupported risk calculation. This keeps the package useful without turning administrative observations into technical conclusions.

Include the history of checks without overstating it

Where the organization's approved process includes intermediate checks, retain the relevant records and their defined acceptance basis. Identify the actual check scope and any changes in method or reference. A statement that checks were performed is less informative than a retrievable record showing what the process required and what was recorded.

Do not treat a check of one function as evidence that every function remained satisfactory. The technical reviewer must assess how the check relates to the interval question. The coordinator's role is to maintain the scope connection and identify missing or ambiguous records, not to generalize beyond the evidence.

Separate recorded results from later summaries. If a monthly summary states all checks passed, preserve the underlying records or their controlled reference where required. A summary can help navigation, but it should not erase exceptions, repeated measurements or corrective actions that matter to interpretation.

Verify the evidence index before analysis

Check a defined set of extracted entries against the original reports before the reviewer uses the package. Compare item identity, event date, scope reference, units and condition stage. A misplaced decimal or a result copied from the wrong column can survive into an otherwise sophisticated analysis. Keep the check focused on faithful transcription; it does not replace the technical assessment of what the values mean.

Record who checked the extraction, which entries were checked and what corrections were made. If an error pattern appears, broaden the check where justified by that pattern instead of assuming that a corrected sample establishes the accuracy of every remaining row. Retain the original source and the corrected index together through the organization’s document process.

A hypothetical non-comparable history

A fictional instrument has four retained calibration reports. The first two cover a defined function with incoming and final results. The third covers a narrower scope after a repair, and the fourth uses a different reporting format that does not clearly distinguish the incoming condition. A coordinator initially places all four pass statements in one column.

Before the technical review, the coordinator replaces that simplified summary with an event index showing scope, service changes and the unresolved reporting question. The reviewer requests clarification for the fourth report and decides how the narrower third event can be used. No automatic extension is generated from the four pass labels.

This fictional example demonstrates evidence preparation, not a conclusion about any instrument's proper interval. The useful outcome is that the reviewer sees the actual strengths and limits of the history rather than a polished but misleading series of identical status words.

Document the method and decision through the authorized process

The technical owner should identify the method used and the rationale for the decision under the applicable requirements. The coordinator can retain that reference and ensure that the defined population and source evidence are connected. Do not select a statistical or calendar method merely because a spreadsheet can calculate an output.

Keep the result explicit: the interval or date basis may remain unchanged, change under approval or require additional evidence before a decision. A review does not have to produce a longer interval to be useful. Recording why the current basis remains appropriate can be a meaningful outcome.

If the decision applies to a group, identify the included equipment and any exclusions. If it applies only to one item or defined duty, preserve that limit. A broad note saying interval reviewed should not authorize unrelated instruments or future uses that were outside the assessment.

Implement the result without losing its scope

Translate the approved decision into current records through the controlled date-change process. Verify each item identity and the approved value before entry. A group decision can still require item-specific dates, and a copied spreadsheet formula should not be assumed to implement the authorization correctly for every row.

Review reminder lead times and service planning separately. An interval decision changes the equipment-control basis only as approved; it does not automatically establish when purchasing should book a supplier or who should receive an email. Those planning choices should remain aligned with the actual service process.

The free NDT Connect tracker stores a current due date and status but does not analyze calibration trends or determine intervals. Its email reminder uses the recorded date. Retain the review method, evidence and approval in your controlled records, and use the tool as the current summary rather than presenting it as an interval-optimization system.

Retain limitations and the next review trigger

Record unresolved evidence limitations even when the authorized owner can make a decision. A missing early report, uncertain use estimate or changed configuration should not disappear from the package after approval. Future reviewers need to understand what was and was not available when the basis was established.

Identify the review triggers defined by the authorized process. These may concern new evidence or changes relevant to the equipment, but this guide does not prescribe a universal review frequency. Keep an administrative follow-up distinct from the calibration due date and from any technical check schedule.

The downloadable CSV is an evidence index, not a calculator. It identifies source events, condition stages and comparability questions so the reviewer can retrieve the original material. Remove the fictional row and adapt the index to the actual review scope. A transparent evidence package is more useful than a numeric recommendation unsupported by the available history.

Build a question log for non-comparable records

When a report cannot be directly compared with the others, record the precise question and the person able to answer it. Examples include whether a result is incoming or final, whether the same function and range were assessed, whether an adjustment preceded the reported values or whether a changed method affects interpretation. A generic uncertain-data flag is less useful than a specific review question.

Keep the original reported information intact while clarification is pending. Do not convert, average or substitute values merely to create a complete series. The technical reviewer should decide which transformations or comparisons are appropriate under the chosen method, and the evidence package should preserve the original source alongside any authorized derived information.

For a fictional three-event history, the first report provides incoming results, the second reports only final results after an adjustment and the third identifies a changed configuration. The coordinator records those distinctions and asks the reviewer how each event can contribute. Three passing summaries would hide the very differences the review needs to consider.

Close each question with the clarification source, reviewer decision and effect on the evidence set. An event may remain useful for one purpose while unsuitable for a particular comparison. Do not reduce that nuanced outcome to either include everything or discard the entire report without explanation. The question log lets a later reviewer understand why the dataset was assembled as it was.

Define grouped equipment without hiding exclusions

If the review considers a group, record how membership was defined and which instruments entered or left during the relevant period. Manufacturer model, configuration, duty and available history may all be questions for the technical owner. The coordinator should preserve the actual population and differences, not assume that a common product name establishes complete comparability.

Include the reason for excluding an item from an analysis where an exclusion is authorized. An instrument retired after a problem should not disappear from the available history merely because it is no longer on the active register. The technical owner determines the significance, while the coordinator makes the evidence and population changes visible.

Use a fictional group containing six instruments, one introduced recently and one retired during the review period. Their histories have different lengths and outcomes. Reporting six identical units with the same observed record would be misleading. A clear population table identifies the periods and source events available for each item so the reviewer can apply the actual method appropriately.

Retain the link between the group decision and the individual records it affects. If the approval excludes one configuration or applies only to a defined duty, preserve that boundary when current dates are implemented. A broad group title should not silently authorize every future instrument purchased under the same model name.

Finally, record the limitations of the available population. Sparse or inconsistent evidence may constrain the decision, but a transparent package still helps the authorized owner choose a defensible next step. The coordinator's contribution is reliable evidence organization, not an unsupported claim that a larger spreadsheet necessarily produces a better interval recommendation.