Correct parts can still make the wrong packing kit

## Examine the combination the customer receives

Nine of thirty-six reviewed dispatches contained packing components from different revisions. The uncomfortable feature was that an individual component could be made correctly to its own drawing and still be wrong beside the other items in the kit. A report of completed quantities did not reveal that relationship. The customer needed the intended configuration for a particular application, not a collection of independently plausible parts. I asked our team to examine the kit as a physical combination rather than assume that separate production records could establish the result at dispatch.

Revision changes are often discussed as document changes. The drawing is updated, the relevant colleagues are informed and the new requirement enters the planning system. Meanwhile, material already made remains on racks, part-built kits occupy preparation areas and earlier orders may still require an older configuration. Those conditions do not disappear when the new document is issued. The operating question is how the different states remain recognisable to the people handling them. I wanted the discussion to include stock already present and work already under way, not only the next clean order beginning after the revision.

The problem did not mean every older revision was unusable. Its suitability depended on the intended order and the authorised configuration. Discarding everything old would have been a simple way to avoid some mixing, but it could also waste material or interfere with legitimate commitments. The technical and customer owners needed to determine what remained appropriate. Operations needed to preserve the identity of those decisions through storage and preparation. That distinction prevented us from treating a physical segregation proposal as permission to redefine the customer's requirement or to make a technical substitution for convenience.

The dispatch manager's proposed response was another final checklist at loading. It had a practical attraction: it could be introduced quickly without reorganising the preparation area. The final check is important, and I did not argue for removing it. My concern was the amount of ambiguity arriving there. If several earlier steps allowed components to lose their relationship with the intended kit, the last person would be asked to reconstruct that relationship under dispatch pressure. A checklist can prompt a necessary question, but it cannot always supply the missing evidence needed to answer it.

## Reduce the ambiguity before the last check

We funded forty-seven lakh rupees for physical segregation, kit identity and supporting dispatch preparation. The arrangement used space and time and delayed a planned convenience improvement elsewhere. I supported that sacrifice because the problem had developed before loading. We needed the preparation sequence to carry the intended combination forward, so that the final check could confirm a coherent account rather than repair a confused one. The proposal retained technical acceptance with qualified colleagues and customer-configuration decisions with their proper owners. Operations was responsible for making those decisions survive into the physical shipment.

A segregation arrangement can itself become complicated if it is designed without the people who pick and prepare the work. I wanted the team to test whether the distinction was visible where the decision was made, including when several orders were being handled close together. A tidy plan on paper is not enough if the physical flow encourages material to cross back into an indistinguishable group. The work needed to fit the actual preparation sequence. That is why I treated the expenditure as an operating programme with follow-up, rather than a purchase of labels that could be declared complete on delivery.

The programme also required a clearer account of responsibility when a revision changed. Someone had to know which stock and kits were affected and which decisions remained with technical or customer colleagues. Without that account, each department could complete its own update while assuming another department had addressed the physical material. I wanted the handover to identify the unresolved items, not merely confirm that a notification had been sent. The useful evidence was whether the next person could determine what belonged in the intended kit. A message delivered successfully was only one part of that evidence.

The cheaper checklist-only route remained a legitimate comparison. In another process, with simpler configurations or stronger existing identity controls, it might be sufficient. The reason for our choice was the defined pattern we had found in the reviewed dispatches. I do not present segregation as a universal answer to every revision problem. It can consume space and create more handling, which must be justified. A board paper should make those costs visible rather than describe the preferred control as though it improved every dimension of the operation at once.

## Treat the remaining case as evidence

The subsequent thirty-six-dispatch review found one mixed-revision case. We reported it and investigated how it had occurred. The reduction from nine did not make the remaining case unimportant, particularly for the customer whose kit was affected. It also did not establish perfect traceability across all products or the whole company. The result concerned the defined review. I wanted the operating team to retain that boundary because a dramatic improvement can encourage broader claims precisely when management should be learning from the exception that remains.

The investigation of a remaining case is different from repeating the programme's success narrative. It asks whether the same route failed again, whether a new condition was involved or whether the intended arrangement was not being used as expected. The answer may require a narrow correction rather than a complete redesign. I would not assume the programme was worthless because one case remained, just as I would not assume the case was irrelevant because the average result improved. Both reactions avoid the practical question of what the evidence now says about the way the kit is prepared.

Directors should understand the distinction between component completion and customer readiness when reviewing an industrial operating report. A company can report high output and still create costly work at the point where parts are combined. The board can ask whether management's measures reflect the intended application, how revisions affect material already in the system and what the final check is expected to prove. It should not choose individual storage locations or instruct dispatch staff. The oversight value lies in examining whether the operating account follows the product far enough to describe what the customer actually receives.

The case changed my own language about completion. I now ask which combination is complete and for which intended requirement, rather than accept a total of finished components as the end of the discussion. Our forty-seven-lakh-rupee programme reduced the mixed-revision count in the reviewed group from nine to one, with additional space and preparation work. It preserved the final check while reducing the uncertainty reaching it. That is a more useful account than saying we improved dispatch discipline. It identifies the relationship that had been lost, the cost of preserving it and the evidence that still required attention after the improvement.