Application of Uncertain Data Handling on the Assessment of Tomato Quality

The handling of uncertain data is demonstrated on an empirical grading function used for the assessment of tomato quality. The grading function studied here is designed to measure the departure of the properties of a tomato or a population of tomatoes from an assumed optimal tomato. Uncertain data are considered using the Taylor Series expansion of the grading function, a function of random variables, which provides the ability to determine the variability of the outcome of the function. Once this variability is quantified, confidence intervals are determined and considered. The degree of confidence in a result has a wide array of ramifications, ranging from providing valuable decision support to assisting in guidance of research activity. Results of this analysis are useful for the Advanced Life Support (ALS) community, as the application of the Taylor Series expansion is not limited to grading functions alone, but can be applied to any model where information describing the variability of the model inputs is available.