Applications of Metaheuristic Algorithms in Nutritional Optimization and Personalized Diet Planning: A Review

Abstract

Nutritional optimization links food selection and meal scheduling to nutrient adequacy, affordability, preferences and environmental objectives. This narrative review examines where metaheuristic search can assist these decisions and where linear or mixed-integer programming offers a more informative baseline. It distinguishes the optimization of a dietary model from evidence that a recommended diet improves health. Genetic and multi-objective evolutionary methods are discussed alongside particle swarm and differential evolution as possible search architectures, with attention to encoding, feasibility repair and expensive preference models. Published diet-modeling work supports the importance of acceptability constraints, and a multi-level NSGA-II menu-planning study illustrates the joint treatment of meals, days and weekly menus. The review develops an evaluation framework covering constraint satisfaction, solution quality, computational budget, dietary diversity and independent validation. It does not estimate pooled algorithm effects or establish a universal optimizer ranking. Traceability limitations in the numerical compilations restrict quantitative synthesis. Priorities include transparent food databases, matched mathematical-programming comparisons, clinically appropriate constraints and prospective assessment of adoption and outcomes.