A main factor motivating consumer choice is the packaging: in many cases, the consumer choices are prevalently based on it. Actually, in planning the packaging of a new product on the market, due to the many variables that can influence the result, it is necessary to conduct a high number of preliminary analyses. It is therefore desirable to develop an automated method that allows obtaining information and reduces the analysis time and cost. In this work, we propose the use of data science/machine learning methods to verify, but also to predict, the effectiveness of the packaging in the Food&Beverage sector. As proof of concept, after doing a public survey about some Food&Beverage packaging, we value the ability of a feedforward Multi-Layer Perceptron (MLP) Artificial Neural Network, in predicting the results, i.e. if and how much the consumer likes the packaging. Trained MLP shows a very high correlations coefficient (> 0.98) and low mean square error (7.97) and error percentage (5.76%) values in determining the consumer response.

Towards the automated evaluation of product packaging in the Food&Beverage sector through data science/machine learning methods

Ciampi C. M.;Guerriero V.;Rampone S.
2022

Abstract

A main factor motivating consumer choice is the packaging: in many cases, the consumer choices are prevalently based on it. Actually, in planning the packaging of a new product on the market, due to the many variables that can influence the result, it is necessary to conduct a high number of preliminary analyses. It is therefore desirable to develop an automated method that allows obtaining information and reduces the analysis time and cost. In this work, we propose the use of data science/machine learning methods to verify, but also to predict, the effectiveness of the packaging in the Food&Beverage sector. As proof of concept, after doing a public survey about some Food&Beverage packaging, we value the ability of a feedforward Multi-Layer Perceptron (MLP) Artificial Neural Network, in predicting the results, i.e. if and how much the consumer likes the packaging. Trained MLP shows a very high correlations coefficient (> 0.98) and low mean square error (7.97) and error percentage (5.76%) values in determining the consumer response.
Artificial neural networks
Consumer satisfaction
Data Science
Food&Beverage
Machine learning
Packaging
Survey
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/20.500.12070/54778
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
social impact