Not enough data yet? What about expertise
Presented at the 8th Annual BayesiaLab Conference on October 29, 2020.
About the Presenters
Mathieu Le Tertre, Charles Gomes
L'Oréal Research and Innovation, Paris
Abstract
Global warming is accompanied by environmental changes that threaten the potentially permanent degradation of human and natural habitats. This is the reason why L’Oréal has decided to take its efforts even further through a program called “L’Oréal for the Future” and is fully committed to a radical transformation of its activities by 2030. Replacing ingredients that can have a harmful impact on the environment and consumers is now an obsession to ensure clean but still high-performance products.
L'Oréal, like all the major players in the world of cosmetics, must deal with this complex subject and find new natural ingredients or new ways of combining ingredients. What solutions can be put in place when we don't have enough data to address these issues?
L'Oréal has therefore decided to draw on the knowledge of its experts to imagine new combinations of ingredients that could replace an ingredient that no longer meets the challenges of tomorrow through elicitation. The objective of this session is to share this modus operandi, which has materialized, through an enriching discussion between experts, by the proposal of virtual formulas. The experiments then carried out are re-injected into the Bayesian network allowing an update of the model and iterative learning.