Statistical quality control in the food industry: a risk-based approach

Abstract

PurposeQuantitative metrics should be used as a risk management option whenever possible. This work proposes a framework for the risk quantification and the resulting risk-based design of control charts to monitor quality control points.Design/methodology/approachTwo quality control models were considered for the risk quantification analysis. Estimated operating characteristic curves, expressing the defect rate (on a ppm basis) as a function of the sample size, process disturbance magnitude and process capacity, were devised to evaluate the maximum rate of defective product of the processes. The proposed framework applicability on monitoring critical control points in Hazard Analysis and Critical Control Point (HACCP) systems was further evaluated by Monte Carlo simulations.FindingsResults demonstrate that the proposed monitoring systems can be tuned to achieve an admissible failure risk, conveniently expressed as the number of non-conforming items produced per million products, and these risks can be properly communicated. This risk-based approach can be used to validate critical control point monitoring procedures in HACCP plans. The expected rates of non-conforming items sent out to clients estimated through stochastic simulation procedures agree well with theoretical predictions.Practical implicationsThe procedures outlined in this study may be used to establish the statistical validity of monitoring systems that uses control charts. The intrinsic risks of these control systems can be assessed and communicated properly in order to demonstrate the effectiveness of quality control procedures to auditing third parties.Originality/valueThis study provides advancements toward practical directives for the implementation of statistical process control in the food industry. The proposed framework allows the assessment and communication of intrinsic failure risks of quality monitoring systems. It may contribute to the establishment of risk-based thinking in the constitution of quality management systems.