Traceability and Recall in Manufacturing: Lot Tracking

What is traceability in manufacturing

Traceability is the ability to rebuild a product's history: which raw material lot it came from, how and when it was processed, and which customer received it. In a recall it decides whether you call back 5,700 parts or a whole month of 20,000.

There are two directions. Backward traceability goes from the finished product to the material lots, the process and the supplier. Forward traceability starts from a material or process lot and reaches the customers who received it. A good system runs both in hours, not days.

Traceability visual map

The map shows the lot, raw material, process and customer chain, and the recall example with 5,700 parts.

Lot or serial number: how fine should tracing be

The finer the level, the smaller the recall scope and the higher the cost of marking and recording. Let risk decide: a safety part calls for serial numbers or small lots; a low-risk item can take larger lots. The risk analysis in APQP and the control plan helps you choose.

What to record at each step

Traceability works like a chain: each link records what came in and what went out, and the lot identifier passes from one record to the next.

1. Receiving: supplier, delivery note, supplier lot, quantity and date. Create an internal lot and label the material.
2. Storage: location and balance by lot, rotated first in, first out (see WMS).
3. Production: on the production order, the lot of each component consumed, machine, shift, operator and relevant parameters. On computerized lines the MES captures this.
4. Inspection: result, sample and disposition (accepted, on hold, scrapped), tied to the lot.
5. Packing and shipping: finished goods lot, shipping document and destination customer.

The weak point is lot mixing: topping up a bin with new material over old, or combining leftovers from two lots, breaks the chain. Set an end-of-lot rule, line clearance and a single label per container.

Worked example: the scope of a recall

Fictional example: the supplier warns that lot L-2411 of O-rings (10,000 parts received) has hardness out of specification. The plant checks its records and builds the mass balance of the lot:

At customers: 2,000 + 2,200 + 1,500 = 5,700 parts. Inside the plant, to be held: 800 + 1,000 + 2,500 = 4,300 parts. The sum closes at 10,000, so the mass balance is 100%.

Without lot tracing, the plant would have to call back everything shipped in the period, say 20,000 parts. At a recall cost of $35 per part (logistics, sorting and rework, illustrative value):

The scope falls from 20,000 to 5,700 parts, 28.5% of the total, and the cost falls by $500,500. That is the financial case for investing in labels and data capture.

How to implement in 6 steps

1. Define the scope by risk: which products and components need lot or serial tracing.
2. Standardize the lot: a numbering rule (for example date + machine + shift) and a maximum size.
3. Identify everything: labels on materials, containers and product. Barcodes or 2D codes reduce keying errors; the GS1 standard uses application identifier (10) for the lot.
4. Record at the steps in the previous section, on well-controlled paper or in a system.
5. Require suppliers to identify the lot on every delivery, as part of supplier evaluation.
6. Test it with a mock recall at least once a year.

Recall and the mock traceability exercise

The exercise picks one material lot and one finished goods lot and times you: how long to list every affected lot, how many parts are with customers, how many are on hold, and whether the mass balance closes. A common internal target is a few hours and a 100% balance; set your own. Log failures as nonconformities and handle them with corrective action.

In a real recall the sequence is: contain (hold stock and stop shipments), define the scope through tracing, notify customers and, where safety is at risk, the authorities, and analyze the cause (an 8D report fits). Notification duties vary by country and sector, so check the rules of each market you ship to.

What the standards ask for

• ISO 9001: where traceability is a requirement, the organization must control the unique identification of outputs and keep documented information that allows tracing (clause 8.5.2). See ISO 9001.
• IATF 16949: extends the demand for the automotive sector, with traceability plans proportional to risk and record retention times. See IATF 16949.
• Regulated sectors (food, pharmaceuticals, medical devices) have their own lot and recall rules. Check the regulations for your product.

Common mistakes

• Lots that are too big: one defect hits days of production.
• Breaking the chain by mixing lots or reusing leftovers without a record.
• Tracing only the finished product and forgetting component lots.
• Loose paper records, illegible or filed with no index.
• Never testing: the first time tracing is used is the real recall.
• No mass balance: if the numbers do not close, parts are missing.

Frequently asked questions

What is traceability in manufacturing?

It is the ability to rebuild a product's origin, processing and destination through records tied to the lot or serial number.

What is the difference between forward and backward traceability?

Backward goes from the finished product to material lots and process. Forward goes from a material or process lot to the customers who received it.

What is a mass balance in a recall?

A check that everything received in a lot is accounted for: at the customer, in stock, in process or scrapped. In the example, 5,700 + 4,300 = 10,000.

How do you set the lot size?

By risk and the cost of a recall. A smaller lot narrows the scope but needs more identification and recording.

Does ISO 9001 require traceability?

Where it is a requirement, yes: the organization must control unique identification of outputs and keep documented information (clause 8.5.2).

How often should you run a mock recall?

At least once a year and after changes in system, supplier or process, measuring time and whether the mass balance closes.

About the author

Vagner Soares

Lean Manufacturing & Behavioral Management Specialist

Over 20 years in the automotive and metalworking industries (GM and Dana), Lean Manufacturing practitioner since 2006. SENAI instructor and mentor in Brazil’s Brasil Mais Produtivo program, delivering consulting, training and audits for 50+ companies, combining quality, productivity and people development.

LinkedInInstagramFull profile →