Mix del producto óptimo usando algoritmos genéticos

Abstract

Optimal product-mix means to determine the quantity of products to produce to maximize the gain. To determine the optimal product-mix of a Cooperativa Industrial, it has been built a hybrid model combining the discrete-event simulation and the genetic algorithms. The discrete-event simulation is used to infer the indirect unitary cost of each product due to the use of a system activity-based costing. To apply a system activity-based costing the posteriori information is required to, but one can know it (to approach it) a priori applying the discrete-event simulation. The genetic algorithms determine the optimal product-mix that maximizes the utility. These genetic algorithms use the code of value for the chromosomes and they include techniques for the solution of problems with lineal restrictions. The optimal product-mix obtained with the model decrease the losses of the product-mix used in the first semester, of those products whose unitary cost is greater than its price in 43% and they increase the utility in 123%.