Variables sampling plans using composite samples for food quality assurance

Testing composite samples is a useful strategy to achieve sampling economy. Several studies have shown the effectiveness of this technique under the assumption of perfect mixing of primary samples. This paper investigates the effect of imperfect composite sample preparation on the performance of two and three-class variables sampling inspection plans, and identifies scenarios in which testing composite samples is not advantageous. The design of sampling plans using composite samples is discussed and an implementation guide based on two points of the OC curve for perfect and imperfect mixing is provided.