Identification of the significant factors in food quality using global sensitivity analysis and the accept-and-reject algorithm. Part III: Application to the apple cold chain

• Monte Carlo method was used to model time temperature variability of apples.
• Apple variability in background colour and firmness was modelled.
• Duration in CA storage was identified as important factor for final apple quality.
• 7.24% of apples had a final firmness below the quality criteria after wholesale.

The aim of this manuscript is to apply the methodology proposed in an accompanying part I describing the different sources of variability in the apple cold chain allowing the identification of significant factors on the evolution of apple quality. The methodology uses a deterministic model to predict the evolution of quality (firmness, colour) and stochastic models to account for the variability of product time–temperature history, biological properties and initial quality. Hereto, a dedicated field test to obtain a realistic set of time–temperature data for the long term controlled atmosphere (CA) storage of apple was conducted. The apple cold chain under consideration in this study comprises the following links: pre-cooling, CA storage, non-refrigerated transport and wholesale. Then, the simulation results, using kinetic models for describing the evolution of firmness and colour as influenced by the time–temperature profiles, and following the method presented in part I (global sensitivity analysis and an accept-and-reject algorithm) are analysed. Our research indicated that the product time temperature profile along the cold chain was the most influencing factor on the final apple quality. A refined analysis by the accept-and-reject algorithm revealed the impact of CA storage and pre-cooling.