Non-destructive Evaluation of Powder Mixing for Food Processing Industries

Abstract

Food and beverage (F&B) fast-moving consumer goods (FMCG) contribute a major portion of global food loss and waste in the food supply chain in the form of food loss and package defect-induced waste. Food products pass through several processing lines including, mixing, filling, dispensing, sealing, and packaging before dispatch. To save on the cost and footprint to build additional machinery, producers resort to using the same lines for processing food powders with similar or limited ingredients. The storage, mixing and filling conditions of processing lines affect the food powder quality. Hoppers are large stainless-steel containers to store and mix contents before filling into jars/containers for subsequent sealing/capping at high speeds. Such high-speed mixing and filling lead to poor powder homogeneity. Currently, such content-filled containers are subjected to invasive sample extraction for offline spectroscopic analysis in a laboratory on randomly picked samples, which is time-consuming, uses expensive machinery, requires user-dependent data analysis, and results in food loss and waste in the form of poor-quality powders and/or powder/package waste from batch rejection. In this work, we focus on applying a non-invasive sensing technology to prevent food loss and waste generation through early detection and screening of poorly mixed powders in filled containers. A capacitive system is proposed to detect dielectric differences between varying levels of binary blending of beverage powders to showcase its application for powder homogeneity analysis. The detection scheme showed a fast response time (50 ms) and a low detection limit of 15% for detecting powder fraction in a binary blended mixture non-invasively.