Non-dominated Sorting Genetic Algorithm–II Based Optimization for Healthy Meals

There is a growing recognition of the fact that healthy food directly influences the quality of life, as it provides and promotes good health. However, due to the hectic pace of daily life, many people find it difficult to eat healthily, causing increasingly alarming diet-related diseases which have led to about two-thirds of all deaths due to chronic conditions such as diabetes, cardiovascular diseases, and kidney disease. Nevertheless, these conditions can be prevented and minimized if each individual takes meals suited to his body, thereby reducing mortality rate and improving well-being. Setting up healthy meals for different individuals could be tedious and complicated because of the different requirements and preferences that would require consideration of several objectives ranging from energy, cost of meal, food preference to seasonal availability, hence showcasing healthy meal generation as a Multi-Objective problem. This research presents a daily meal optimization that adopts Non-dominated Sorting Genetic Algorithm (NSGA-II), a Multi-Objective Evolutionary Algorithm (MOEA) for user-tailored and optimal meals, capable of simultaneously maximizing energy composition of food items and minimizing the cost of meals. The dataset used comprises two hundred Nigerian foods obtained from Nigerian food composition table. NSGA-II was able to find suitable solutions in reasonable time because of its outstanding diversity metric preventing it from converging to local optimum, and its ability to maintain a set of non-dominated solutions. The output of the algorithm indicates that our proposed method is suitable for optimizing and suggesting timely food combinations and can thus enhance healthy living.