Optimization of Healthy Diet Menu Variation using PSO-SA
Abstract: Optimal healthy diet
in accordance with the allocation of cost needed so that the level of
nutritional adequacy of the family is maintained. The problem of optimal
healthy diet (based on family budget) can be solved with genetic algorithm. The
algorithm particle swarm optimization (PSO) has the same effectiveness with
genetic algorithm but PSO is superior in terms of efficiency, PSO algorithm has
a lower complexity than genetic algorithm. However, genetic algorithms and PSO
have a problem of local optimum because these algorithm associated with random
numbers. To overcome this problem, PSO algorithm will be improved by combining
it with simulated annealing algorithm (SA). Simulated annealing algorithm is a
numerical optimization algorithms that can avoid local optimal. From our
results, optimal parameter for PSO-SA are popsize 280, crossover rate 0.6,
mutation rate 0.4, first temperature 1, last temperature 0.2, alpha 0.9, and
generation size 100.
Author: Imam Cholissodin,
Ratih Kartika Dewi
Journal Code: jptinformatikagg170005
