Retrieving high quality and ready-to-use information from databases
Study the relationships between instrumental and sensory analyses
Objective: The Sicarex Beaujolais (Wine experimental research cooperative company) wanted to explain the relationships between the phenolic composition of red wines and their sensory analysis. Thus, the objective was first to conduct an exploratory analysis and then to predict sensory attributes based on informative analytical parameters.
Our solutions: INRA (Ag. French public research institute) analyzed the polyphenolic composition using 2 HPLC-MS methods (2 instrumental tables) and carried out sensory evaluation with a trained panel (1 sensory table). First, an exploratory multi-way analysis (ACCPS) was performed on a wide range of wines. Then, multivariate models were built to predict certain sensory properties.
Customer benefits: Thus, the Sicarex Beaujolaisdemonstrated that the 2 instrumental analyses were complementory when characterizing the chemical composition of red wines. The Sicarex Beaujolais also gained knowledge on the relationships between the polyphenolic composition and organoleptic perception of wines. The study also determined the sensory attributes that can be predicted using chemical (instrumental) data.
Innovation and pragmatism: Multi-block methods are innovative tools for multivariate data analysis. They explain the relationships between various data sets, such as complementarities or similarities between sensors. Thus, their results help when comparing and selecting of the most relevant analytical methods for product characterization or for the optimization of a measurement protocol. Moreover, the relationships between instrumental and sensory data explain how to improve product quality assessment with a few selected analytical parameters.