Twenty four semi-hard cheeses produced during autumn (n = 12) and summer (n = 12) periods were manufactured and ripened at an industrial scale. Tryptophan and vitamin A fluorescence spectra were scanned on the 24 cheeses at 2, 30 and 60 days of ripening. Principal component analysis (PCA) and factorial discriminant analysis (FDA) were applied on the spectral data sets. The first five principal components (PCs) of the PCA extracted from each data set (tryptophan or vitamin A) of cheeses produced during autumn or summer period were pooled into a single matrix and analysed by FDA. Regarding cheeses produced during the autumn period, the percentage of samples correctly classified was 95.8% and 86.1% for the calibration and validation samples, respectively. Similar results were obtained from cheeses produced during the summer period. Finally, concatenation technique was applied to the tryptophan and vitamin A spectra recorded on cheeses independently of their production seasons. Correct classification was observed for 87.5% and 80.6% for the calibration and validation samples, respectively. Although this statistical technique did not allow 100% correct classification for all groups, the results obtained were promising considering the significant effect of the season on the cheese characteristics. (C) 2006 Elsevier Ltd. All rights reserved.