A quality problem surfaces – a contaminated batch, a damaged shipment, a customer complaint – and the first question is always which containers were involved and where they have been. Whether that can be answered within hours or not at all depends entirely on records that were kept before anybody needed them. Traceability is a capability built in advance or it is not available.
The questions that get asked
Which containers carried this batch. Where those containers went next. What else was carried in them since. And when they were last washed or inspected. Four questions, and each one needs a different piece of the record – a container-to-batch link, a movement record, a contents history, and a maintenance record.
What is needed to answer them
Container identity, at whatever level – batch or unit. A record of what went into which container and when. A record of where containers went. And a record of wash and inspection events. Where any of those four is missing, the corresponding question cannot be answered and the investigation has to fall back on assumption.
Batch level is usually enough
For most quality problems, knowing which batch of containers was involved is sufficient, and batch-level records are far cheaper to keep than unit-level ones. Where a fleet has batch numbering and a dispatch record, most investigations can be narrowed quickly. Unit-level traceability is needed only where a single container’s history is genuinely at issue.
What a recall actually involves
Narrowing the scope to the smallest set that could be affected, which is what the records are for. Without them, the scope has to be set conservatively – everything that could possibly have been involved – which is expensive and damages confidence. The value of traceability is mostly in how much it lets you exclude.
How long to keep the records
As long as the product’s own traceability requirement, which for most industrial applications is a few years and for food or medical applications can be longer. Where a fleet serves a customer with a defined retention period, that period applies to the container records as well, and it is worth confirming rather than assuming.
What it costs to keep
A dispatch record with container batch and contents, which most operations already produce in some form. A wash and inspection log, which is cheap once the inspection exists. And a way to query them, which usually means a spreadsheet rather than a system. Where a fleet has all three, traceability is available at effectively no marginal cost.
The failure mode
Records exist but cannot be queried – a paper log in a folder, or a spreadsheet that is not structured. Where a traceability exercise has failed, it is usually here rather than at the point of capture. Structuring the record so that a query by batch or by date is possible is a small amount of work done once.
What a good answer looks like
A good traceability answer names the containers, the date, the contents and the destination, and it arrives the same day. Where the record supports that, an investigation narrows to a handful of units instead of a month of production, and the difference in cost is usually two orders of magnitude. That is the entire return on keeping container records, and it is realised on the one occasion in three years when something goes wrong.
Testing it
Once a year, run a mock trace: pick a dispatch from six months ago and answer the four questions from the records. Where it takes more than an hour, the records are not structured for the purpose and the exercise has found a real gap. The identification needed to support it is discussed in what a container number should carry.