Online Program

Time Intensity Curves Modeling: Parametric and Conventional Approach Comparison (309877)

Keywords: Time intensity curves, Parametric curves, logistic curves, curve fitting, sensory profiling

Two different approaches for time intensity curves analysis are compared. In one side, the conventional approach that is already implemented in XLStat is performed for a fruit beverage dataset with flavor intensity attributes. The parametric approach is implemented in R and performed for the same dataset. At the end of the analysis the parametric approach works well under after a rigorous data cleaning, but with the advantage of interpretable parameters that allows statistical analysis. A novel R-based script was created to automate the analysis of Time Intensity Curves based on the work of Eilers and Dijksterhuis (2004). This enables the evaluation of Time Intensity curves with a parametric model with interpretable parameters. This method can be used to identify the unique dynamic experiences of our products and can be used to statistically evaluate individual panelist performance. This represents another import tool for Sensory evaluation.