This chapter is developed in more depth than the others. It shows how a readiness question becomes a set of evidence requests, validation rules, and a defensible finding.
Why it matters
Before an automated system can reliably store, move, identify, or pick an item, it needs the information required to handle that item correctly.
In many operations, that information starts in product master data held across ERP, WMS, supplier, and other systems. A person compensates for a wrong carton dimension without noticing. A storage algorithm, a gripper, or a vision system uses the value it is given.
When physical characteristics are missing, inconsistent, or outdated, a well-designed project may need extra data cleanup, validation, or changes to its operating assumptions.
What can go wrong
- Slotting and storage design are calculated from dimensions that do not match the physical product.
- Items are assigned to totes, trays, or locations they do not fit, and are discovered at induction.
- Weight-based checks reject good picks because stored weights are wrong.
- A case-quantity mismatch between systems produces inventory differences after every automated transaction.
- Barcodes on inbound product cannot be read at the required rate, or do not match the item record.
- Handling constraints such as fragility or orientation exist only in the knowledge of experienced staff.
Warning signs
- Item dimensions are populated but have never been physically verified.
- The WMS and ERP use different units of measure.
- Packaging quantities vary by supplier without consistent records.
- Fragility, orientation, or handling restrictions are undocumented.
- New products are introduced without a defined data-validation process.
- Many records carry the same placeholder value, such as 1 × 1 × 1.
What evidence to examine
| Evidence | What it can establish |
|---|---|
| SKU master export | Attribute completeness and potential anomalies |
| Packaging hierarchy records | Consistency across each, inner-pack, case, and pallet levels |
| Dimensioning and weighing records | Whether physical measurements support stored values |
| WMS/ERP data mappings | Attribute synchronization and ownership |
| Product images and handling specifications | Suitability of available information for the intended vision and handling use cases |
| Master data change procedures | Whether data quality can be maintained after deployment |
How the evidence is validated
An export on its own proves little. It becomes evidence when it is tested against stated rules. The rules below are typical, and the required attributes depend on the handling workflow being proposed.
| Check | Example rule |
|---|---|
| Completeness | Every active SKU has length, width, height, and weight at each packaging level it is handled at |
| Plausibility | No zero, negative, or placeholder values, and calculated density falls in a credible range for the product type |
| Unit consistency | Each attribute uses one declared unit of measure per system, with documented conversions |
| Hierarchy arithmetic | Case weight is consistent with each weight multiplied by case quantity, within a stated tolerance |
| Cross-system agreement | The same SKU carries the same values in ERP and WMS |
| Identification | Barcode values are valid, unique to the item and packaging level, and match the item record |
| Physical agreement | A representative sample, measured on the floor, matches stored values within tolerance |
Measurement method matters as much as the numbers. The GS1 Package and Product Measurement Standard defines a consistent way to determine which face of a package is the front and how each dimension is taken. Adopting one documented method removes a common source of disagreement between suppliers and systems.
Illustrative finding
The finding names its evidence, states the rule it was tested against, and limits its claim to what the evidence supports. It does not predict project failure. It identifies work to be done and the order to do it in.
What to do next
- Define the required attributes for the intended workflow, by packaging level.
- Export the active SKU master and run completeness and plausibility checks.
- Rank exceptions by volume, so the fastest-moving items are corrected first.
- Physically measure a representative sample and record the method used.
- Assign an owner for each attribute and add validation to new-item setup.