Meaning
Statistical selection is a quality control methodology where a second set of samples is tested if the first set yields ambiguous results. Suppliers apply a double sampling plan to reduce testing costs while maintaining a high level of confidence in the batch quality. This approach allows for an early decision to accept or reject the batch based on the first small sample, while reserving the larger second sample for borderline cases.
Sampling Stage
The initial step requires the quality inspector to draw a random sample of a predefined size from the shipment. If the number of defective units in this sample is below the acceptance threshold, the inspector accepts the entire batch. Conversely, if the defect count exceeds the rejection threshold, the inspector rejects the batch immediately.
When the count falls between these two numbers, the double sampling plan dictates that a second, often larger, sample must be drawn and inspected.
Decision Boundary
The cumulative results from both the first and second rounds of testing determine the final disposition of the batch. In this stage, the total number of defects from both samples is compared against a combined acceptance number. This rigorous methodology prevents marginal shipments from entering the supply chain while protecting the factory from false rejection of compliant goods.
Risk Apportionment
Standard purchase agreements specify the exact parameters of these inspection protocols to distribute commercial risk between the buyer and the manufacturer. Implementing a double sampling plan protects both parties from the financial consequences of incorrect batch decisions. It ensures that any rejected batch has undergone two independent rounds of analysis before a formal return-to-vendor claim is filed.
This procedure is recognized by Chinese maritime courts as standard evidence in commercial supply contract disputes, reducing the likelihood of drawn-out litigation over material quality.