Resolving quality issues in food manufacturing with advanced data analysis

Multivariate data analysis tools help manufacturers get deeper insights and make sense of their complex data, from understanding raw material properties to identifying critical process variables and the relationships between them which affect quality.

This case study explains how a chocolate manufacturer used best practice to resolve a quality issue and optimise their production processes using multivariate data analysis and experimental design.

Executive Summary

• The challenge: Quality issue resulting in a large amount of end product being scrapped
• The solution: Used multivariate data analysis and Designed Experiments to identify and manage the variables causing the quality problem
• Methods used: Principal Component Analysis (PCA), various regression methods, Fractional factorial designs
• The result: Resolved the quality problem, enabling Nidar to reduce waste and process failures, saving approximately $1M per year. The newfound knowledge was transferred to other production lines.