Acceptance Sampling
What does acceptance sampling decide?
Acceptance sampling decides lot disposition—accept, reject, or rework/quarantine—based on sample results. It is a go/no-go release rule, not a full-population quality estimation method.
What parameters define an acceptance sampling plan?
Common plan inputs include lot size, sample size, acceptance number, inspection level, AQL (acceptable quality level), and sometimes LTPD or producer/consumer risk targets. These parameters determine the discriminatory power of the plan.
What happens when the sample size is too small?
A sample that is too small gives the plan insufficient discriminatory power, so a bad lot can be accepted. This can later cause line stoppage, scrap, customer escapes, or hidden raw-material contamination downstream.
Acceptance sampling is a statistical quality-control method used to decide whether to accept or reject a lot by inspecting a representative sample rather than every unit. It is applied to incoming raw materials, supplier shipments, WIP, and finished goods when 100% inspection is too costly, slow, or destructive. In Dynamics 365 Supply Chain Management, it is configured through acceptance sampling charts and item sampling under inventory quality control.
On a receiving dock, a lot arrives as a pallet or batch, and quality team identifies lot size before selecting the sampling plan. Instead of checking every carton, the operator pulls the required sample count randomly across the lot. Units are checked against acceptance number and AQL. If defect count stays within limit, the lot is released to stock or production; otherwise it is rejected, quarantined, or returned to the supplier. In high-volume manufacturing, this supports fasteners, castings, molded parts, PCB assemblies, and raw-material coils or pallets, where a statistically defensible go/no-go decision is needed without delaying dock-to-stock flow. It is also used when inspection is destructive or expensive, such as pull tests or contamination checks. Operationally, the method depends on lot integrity: the shipment must be one coherent lot, the sample must be random, and the procedure must be standardized in an SOP or ERP quality rule set.
Non-random sampling at receiving: Pulling only the easy-to-reach cartons from the top of a pallet biases the result and lets concentrated defects in the middle or bottom pass into inventory, undermining the entire sampling plan.
Wrong AQL or acceptance number: A plan that is too lenient accepts lots with meaningful defect rates, while a plan that is too strict rejects good material and triggers supplier disputes, dock congestion, and line starvation.
Misusing sampling as a quality estimate: Treating sample results as a precise measure of lot quality creates false confidence in supplier capability and hides process drift, because acceptance sampling is only a go/no-go disposition rule.