On the Design of Healthy Menus through Evolutionary Computation
Coromoto León (cleon@ull.edu.es)
Departamento de Ingeniería Informática y de Sistemas, Universidad de La Laguna, Spain.
Gara Miranda (gmiranda@ull.edu.es)
Departamento de Ingeniería Informática y de Sistemas, Universidad de La Laguna, Spain.
Eduardo Segredo (e.segredo@napier.ac.uk / esegredo@ull.edu.es)
School of Computing, Edinburgh Napier University, United Kingdom.
Departamento de Ingeniería Informática y de Sistemas, Universidad de La Laguna, Spain.
Casiano Rodríguez (crguezl@ull.edu.es)
Departamento de Ingeniería Informática y de Sistemas, Universidad de La Laguna, Spain.
Juan Manuel Ramos (jramospe@ull.edu.es)
Departamento de Ingeniería Informática y de Sistemas, Universidad de La Laguna, Spain.
Abstract
The menu planning problem has great interest in the field of health due to the importance of correct nutrition in today's society. In this work we consider two objectives. The first is the cost of the menu, and the second, we want to minimise the degree of repetition of the meals that shape the final menu. The main goal is to offer a varied, economic, and balanced plan - from the nutritional point of view - considering a prefixed number of days and ensuring a proper nutrition for children at school canteens. When generating meal plans, we also take into account other special circumstances, such as allergies, food intolerances, and specific constraints regarding religion or lifestyle habits (vegetarians or vegans, for example).
To solve the problem we have applied standard multi-objective evolutionary algorithm which have shown very good results in the related literature: NSGA-II, SPEA2 and Adaptive IBEA. As a result, the algorithm provides a set of non-dominated solutions from which the final user can choose the one which better suits to the children needs and the canteen features.