The study presented in this paper addresses uncertainties that arise in spectral measurement and analysis. These uncertainties can be dealt by means of fuzzy classification, but on order to achieve high quality results, it is necessary to manage and compute them beforehand. Previous research has typically only addressed uncertainty per wavelength in spectra, without considering the total uncertainty across the spectrum and in the analysis. This paper presents a method for managing and calculating uncertainties throughout the entire process and integrating them into a fuzzy pattern classifier. The approach is based on the Guide to the Expression of Uncertainty in Measurement and extended by our own methodology to manage uncertainty in spectra including the construction of a fuzzy pattern classifier. An example from UV/Vis spectroscopy illustrates the practical applicability of the methods presented.