Determination of consumer acceptance limits to sensory defects using survival analysis

Survival analysis concepts and calculations were applied to consumers' acceptance/rejection data of samples with different levels of sensory defects. The following defects in UHT milk were studied: acid, caramel, cooked, dark color, lipolytic and oxidized. For each defect a series of nine concentrations were prepared and tested by 50 member consumer panels. The lognormal parametric model was found adequate for most defects and allowed prediction of concentration values corresponding to 10% probability of consumer rejection. For cooked flavor the model was not adequate due to a large number of right censored data, consumers did not find this defect objectionable. For oxidized flavor there were also many right censored values, but this was due to an inadequate concentration range. (C) 2004 Elsevier Ltd. All rights reserved.