Real-world situations frequently need more than a single solution to the optimization tasks within. For instance, clustering and classification problems must have all optimal prototypes/rules discovered in order to form a complete solution to the learning assignment. Conversely, one solution problems may require all possible configurations of the result or solutions space might favour the appearance of the blockage into local optima phenomenon. The detection of all multiple optima of a problem is consequently the aim for the mainstream of practical applications. Present paper describes a recently developed evolutionary technique for multimodal optimization, i.e. genetic chromodynamics. Applications of its standard algorithm and variations in the fields of clustering, classification and scheduling are presented. Cluster prototypes can be evolved and obtained together with the optimal number of groups. Subpopulations can contain rules for a class and at least one rule for each category of the classification problems results. Finally, multiple optimal schedules in job-shop scheduling problem can be reached. Results prove the suitability and ability of the technique to determine multiple optimal solutions and cope with local optima. Future work is concerned with some mechanism to avoid the explicit appointment of values for the radii within.