Developing a prediction model for consumer acceptance

The international research project combines our expertise in the field of product performance, from the human sensory analysis of food to the analytical investigation of aromas and flavors to the use of algorithms and machine learning.

Due to their wide variety of sensory properties and their clearly defined composition, fruit gums are ideal model foods. In the project “CandyCrunch” our scientists are using them to test the combination of physico-chemical analysis data with sensory perception to predict consumer acceptance. 

Digital support for sensory product optimization

Fruit gums are extremely popular with consumers due to their diverse shapes and flavors. However, as with other sweets as well as savory snacks, the growing nutritional awareness of consumers is also changing the demand for healthier alternatives to conventional fruit gums. For this reason, the industry is focusing on developing a wide range of new fruit gum formulations that not only have a different nutrient composition but also take sustainability aspects into account.

The challenge in development is to achieve the desired sensory properties of the new fruit gums. Deviations from familiar products could be rejected by consumers, which would jeopardize the acceptance and thus the market success of the new products. This would pose a high risk for small and medium-sized companies in particular and hinder the development of healthier alternatives.

In order to minimize this risk, the Cornet research project “CandyCrunch” aims to develop a model for predicting consumer acceptance of new healthy fruit gum formulations based on the chemical, physical and sensory properties of a product. With its holistic approach, the project maps the entire purchasing process: from the initial consideration of the packaging - including label declarations - to consumption and the final assessment by the consumer. The targeted combination of instrumental and sensory data with the hedonic evaluation by the consumer is used to determine factors influencing acceptance. Research partners from Germany (Fraunhofer IVV, IVLV) and Brazil (School of Food Engineering (FEA)/ University of Campinas, Instituto de Tecnologia de Alimentos (ITAL)) as well as small and medium-sized companies from both countries are participating in the research project.   

Methods for sensory and analytical evaluation of fruit gums

The focus is on the comprehensive characterization of both fruit gum products already established on the market and those newly developed in the project. The properties such as aroma, taste, color and texture of the products are initially characterized using instrumental analysis. A special focus is placed on the investigation of aroma-active compounds, whereby their composition and the release mechanisms from the food matrix are of particular interest. Using advanced techniques such as gas chromatography ion mobility spectrometry (GC-IMS), the variety of volatile compounds can be detected quickly and precisely.

The analytical data obtained in the first step helps to create a detailed sensory profile of the individual fruit gum products. These are expanded and supplemented by human sensory studies. In order to be able to reliably predict consumer acceptance of new products at a later stage, the popularity of the fruit gums will be evaluated in consumer studies. In addition, the multisensory interactions of packaging and product properties on perception will be investigated as part of the project. The influences of additional health and sustainability-related information provided on the evaluation of popularity and purchase intention will be investigated.

Based on the data obtained from the sensory profiles in conjunction with the results from the consumer studies, all the main areas of research will be brought together using chemometric methods and machine learning to enable targeted prediction of consumer acceptance for novel products.

Digitization of sensory perception for sweets

The results from this project can be adapted in many ways to other products and sectors. With the help of digital forecasting tools, consumer acceptance can be tested early on in the product development process, reducing development costs and increasing the chances of success of new products on the market. This is a significant relief for small and medium-sized companies in particular, as resources are limited and the financial risk is greater.

The tool is transferable to any industry that offers packaged consumer goods with advertising claims whose sensory perception (smell, taste, color, feel or texture) is decisive for the product experience. With the help of the digital image of the sensory product characteristics and the correlation with hedonics, product development can be digitized and the consumers’ reactions predicted, which frees up the necessary human resources and makes the development process more targeted and objective. In addition to the food and beverage industry, this is also of particular interest for cosmetics, household products and food supplements.