Prediction of chiral separations using a combination of experimental design and artificial neural networks

In this work the advantages of using artificial neural networks (ANNs) comb ined with experimental design (ED) to optimize the separation of amino acid s enantiomers, with a-cyclodextrin as chiral selector, were demonstrated. T he results obtained with the ED-ANN approach were compared with those of ei ther the partial least-squares (PLS) method or the response surface methodo logy where experimental design and the regression equation were used. The A NN approach is quite general, no explicit model is needed, and the amount o f experimental work. can be decreased considerably. (C) 1999 Wiley-Liss, In c.