A hierarchical method for evaluating products with quantitative and sensory characteristics

In some industries, such as food processing, product performance is assessed with multiple criteria including both quantitative measurements and subjective characteristics evaluated by a sensory panel. A sensory panel consists of trained individuals who compare products yielding data often in the form of matrices of paired comparisons. This paper presents a hierarchical method, based on the Analytic Hierarchy Process, for computing weights, or composite performance measures, for products of this type. The paper also presents a new method for collecting and analyzing sensory panel data. The proposed method enables comparison of alternative products taking into account multiple quantitative and sensory responses, their relative importance, and the experience and reliability of panel members.