Neural networks have been widely used for both prediction and classificatio
n. Back-propagation is commonly used for training neural networks, although
the limitations associated with this technique are well documented. Global
search techniques such as simulated annealing, genetic algorithms and tabu
search have also been used for this purpose. The developers of these train
ing methods, however, have focused on accuracy rather than training speed i
n order to assess the merit of new proposals. While speed is not important
in settings where training can be done off-line, the situation changes when
the neural network must be trained and used on-line. This is the situation
when a neural network is used in the context of optimizing a simulation. I
n this paper, we describe a training procedure capable of achieving a suffi
cient accuracy level within a limited training time. The procedure is first
compared with results from the literature. We then use data from the simul
ation of a jobshop to compare the performance of the proposed method with s
everal training variants from a commercial package.