Statistical multivariate analysis of the dosing process results for predictive production and quality management – a case study from the food industry

Purpose: The aim of this article is to perform a multivariate statistical analysis of package filling process results for predictive production and quality management. The article presents a case study from the food industry that demonstrates the feasibility of using an appropriate set of control charts for ongoing and predictive production and quality management. Design/methodology/approach: The objectives of the article were achieved through the use of Statistical Process Control (SPC) tools, in particular control charts. The control charts used include both traditional numerical chart such as Xbar and S and special charts such as MA, EWMA, CUSUM and GCC. Findings: SPC tools such as control charts have proven to be extremely useful in monitoring the filling process and predicting future performance. By carefully monitoring the process using traditional and special control charts, it is possible to quickly identify small, gradual or sudden changes that may occur in the production process before the process gets out of control. Research limitations/implications: The research will continue by identifying additional factors that affect the quality of the product, particularly as regards precision and accuracy of dosing, and by evaluating the process studied in terms of its ability to meet customer requirements. Other statistical techniques will also be used to identify patterns and relationships between the various parameters of the process under study. This approach will provide more comprehensive information about the quality and ability of the dosing process to meet customer requirements.. Practical implications: By implementing the right SPC toolkit and using dedicated software that significantly speeds up data analysis, companies can effectively control the quality of the production process. By monitoring the behaviour of the process over time and detecting small changes and trends, it is possible to respond to potential problems in advance. Originality/value: This article is intended for production process managers who want to learn how to use the right SPC toolkit to obtain information about the process behaviour and the moments when intervention actions should be taken.