Color management is the method of communicating color information between devices with predictable and consistent results. A traditional color management workflow communicates color based on tristimulus colorimetry. This method of communication reduces the accuracy and uniqueness of the color information.Convexity is used in this work to find the most extreme set of spectral reflectances that all result in a single CIE XYZ value. Once the entire convex set is determined analytically, a method is needed to deduce the most optimal reflectance. A number of different techniques are implemented, primarily involving quadratic programming.The analytical method used to find all spectral reflectances for a given CIE XYZ set is evaluated using a set of training spectral curves that are an average of several measurements of a test chart printed according to a standard printing condition. The CIELAB values are compared to the training set under different viewing conditions. For example, the maximum CIE XYZ color difference within the convex set under the second viewing condition, D50 simulator, was ΔE_ab 20. All CIE XYZ values derived from the convex set of reflectances under the original viewing conditions resulted in the same CIE XYZ value as the starting data.Two different approaches to finding an optimal curve are evaluated. The first technique uses two different smoothness constraints to estimate an optimal reflectance curve. The next approach uses an aim curve to estimate reflectances. The curves generated from this technique were very close to the aim and gave no difference to the original CIE XYZ values. A media adjusted scaled aim curve was then used to further improve the results.After the reflectance curves are generated, a workflow needs to be in place to utilize the additional color information. A spectral CMS extends the functionality of a colorimetric approach to include both spectral data of the color data and the viewing conditions, as well as a dynamic CMM.