Fractional nonconformance assessment : this dissertation is submitted for the degree of Doctor of Philosophy in Statistics, Massey University, New Zealand

Abstract

Food quality and safety are important due to the very nature of the product and the potentially severe consequences of a fault in the manufacturing process. Statistical tools are widely used in food quality assurance. However, measurement error, which is inevitable in food manufacturing due to the variation and inaccuracy of the measurement system, affects the performance of statistical quality control activities. The concept fractional nonconformance was recently proposed to assess the probability of nonconformance for error-prone individual measurements. This thesis presents five pieces of work for fractional nonconformance assessment mainly suitable for food quality assurance and other applications. The new statistical methods developed pertain to control charting, acceptance sampling, and conformity testing areas. The application of the proposed methods is illustrated with real data from a leading New Zealand dairy product manufacturer. Interactive web-based Shiny apps providing step-by-step guidance to implement fractional nonconformance analytic tools are developed for practitioners.