Spreadability will make food production more sustainable

Anouk Lie-Piang has developed a model to predict the influence of "enriched ingredients" on the structure of foods. Thinking from the functionality of ingredients opens new possibilities for more sustainable food production and offers innovative perspectives for product development in the food industry.

Food products consist of various ingredients to achieve desired properties such as shelf life, taste and texture. Lie-Piang's research group aims to make ingredient production more sustainable. Usually, ingredients are isolated from agricultural products, such as wheat and soybeans. However, producing these isolates requires significant amounts of water, chemicals and energy.

Enriched ingredients as sustainable alternatives

Lie-Piang focused on enriched ingredients, which are more sustainable because they have undergone fewer processing steps. However, their use is more challenging because of their complexity and lack of uniformity in the final product. Lie-Piang studied the functionality of these ingredients to determine how they can be successfully applied in food products.

An innovative method for product development

With her developed method, product developers can predict the functionality of ingredients. For example, by determining the "spreadability" or "gel structure" of an ingredient, enriched ingredients can be used in the same way as isolates without compromising the properties of the final product. This allows product developers to work backwards and determine the ingredients and concentrations needed for specific products.

While certain products may require higher purity ingredients and conventional production methods for maximum sustainability, less pure ingredients can often provide the same functionality in a more sustainable manner. As a result, the final product can be nearly identical but produced in a more environmentally friendly manner.

An expert-informed machine learning model

Lie-Piang's method uses machine learning, an innovative technique in the food industry. Although it focuses more on prediction than understanding, the final decision remains with the food technologist. Their expertise allows them to steer the models to develop an expert-informed machine learning model.

Resource-online.nl