You have things to arrange and choices to make. But it all starts with: determine the value of tracing the (status of the) item (whether it is material, component, subassembly, assembly, packaged product, presence, processing). There is a cost involved, and while overhead might be the biggest factor and make you go 'what the heck, let's do them all', especially in the early phases this is not the case.
Now what is value that traceability enables? The simplest overview is:
- when you cannot easily/cheaply/risklessly redetermine the status (i.e. did this pass/fail the test; did it undergo the cleaning, strengthening process)
- when you must be able to identify that part of your product for planned future action (i.e. service, calibration, repair, upgrades; as a pivotal assembly of 'stuff went so bad we need to recall' or
- when you must be able to identify that part of your product for unplanned but probably action as a vulnerable part (i.e. you have had problems with similar items in the past and want to be able to really narrow down your actions based on a strong root cause analysis which often builds on separating matters through traceability))
Once you've identified each part (and process/location) that has such characteristics you need to make a decision:
- Do I rely on the supplier provided batch indication (if they're all providing it it's the cheapest, but they might make mistakes you don't find out until later, and not all of them always care as much)?
- Do I transpose provided items into my own internal supply batch numbers (usually more consistency and control on numbering, though more upfront cost)? You need to have one place, and one place only where you do this, and you must foolproof it so every crosslink is made. If you forget one you might go up bone-creek in your production planning and traceability both.
- Will there be a difference in when or to what purpose I take lots out of the batches, so will I further specify traceability in-house at the part level, will I rely on the DHR or don't I care? Could be the case when you're treating material that came in before then using that lot in separate products.
Now you have an idea of whatfor and what, the how depends on your operation:
- Few but high cost items? Maybe encode in QR code and introduce scanners to make sure no writing mistakes enter into the record, and allow for easy notation later in the field (which is usually the realm of the high-cost items anyway).
- Many low cost items in many batches? Maybe still encode in barcodes, because the longer the number is the easier it is fouled up.
- Many low cost item in few batches? Handwritten on the DHR probably best. Shift to scan if you are producing huge amounts which do not have batch but separate DHR's.
- You can stand some wider spread? First-In First-Out on a date level.
- Not at all and save/insure against when it goes wrong.
Noone says you can't mix the above, but due to habituation of personnel its best to keep it to as few systems to learn for any one person as possible. Could still mean you have different levels at different lines (i.e. equipment versus consumable).
Advantage of scanners is that you might be able to drop data directly into an ERP and have searchable traceability. Because if you've never had to sort through thousands of DHR's to identify the relevant lots you don't know how much that Want becomes a strong Must. It also helps with production planning, so there's more happy stakeholders.
