The extensions of ordinary fuzzy sets such as intuitionistic fuzzy sets (IFS), Pythagorean fuzzy sets (PFS), and neutrosophic sets (NS), whose membership functions are based on three dimensions, aim to describe expert's judgments more informatively and explicitly. Generalized three-dimensional spherical fuzzy sets are introduced with their arithmetic, aggregation, and defuzzification operations in the literature. Our aim is to extend classical CODAS (COmbine Distance-based Assessment) method to spherical fuzzy CODAS (SF-CODAS) method and to show its application with an illustrative example. The paper also defines spherical fuzzy distances based on the membership, nonmembership and hesitancy parameters. To calculate the desirability of an alternative, SF-CODAS method uses the Euclidean distance as the primary and the Spherical distance as the secondary measure. These distances are calculated based on the negative ideal solution (NIS) and the alternative that has the greatest distance to NIS is the best alternative. The paper also carries out comparative and sensitivity analyses between IF-TOPSIS, IF-CODAS and SF- CODAS.