Farm machinery selection using simulation and genetic algorithms

Computer simulation and genetic algorithms were used to optimize peanut farm machinery selection. The objective of optimization was to maximize net returns above machinery costs. A computer simulation model was used to determine net returns above machinery costs. The simulation model determined net returns above machinery costs for a given machinery set, but did not find an optimum machinery set. The optimum machinery set was determined using two search schemes-an exhaustive search and an artificially intelligent search. The exhaustive search scheme involved running the simulation model with all possible machinery sets, and then selecting the machinery set that produced the highest returns. Alternatively, genetic algorithms were used as an intelligent search scheme to generate machinery sets for the simulation model. A genetic algorithm found a near-optimal solution in 10% of the total time required by the exhaustive search. Modifications in the generic algorithm not only reduced the search time by half; but also improved the quality of the solutions.