Run-of-mine (ROM) coal sampling is one of the most important factors in determining the disposition of ROM coal for an overall emission control strategy. Determination of the amount of sample, or still better, the frequency of ROM coal sampling is thus essential to the analysis of overall emission control strategies. A simulation model of a portion of the Upper Freeport coal seam in western Pennsylvania was developed employing conditional simulation. On the simulated deposit, different mining methods were simulated to generate ROM coal data. ROM coal data was statistically analyzed to determine the sampling frequency. Two schemes were suggested: (1) the use of geostatistical techniques if there is spatial correlation in ROM coal quality, and (2) the use of classical statistics if the spatial correlation in ROM coal quality is not present. Conditions under which spatial correlation in ROM coal quality can be expected are also examined. To link the ROM coal and coals from other sources to coal stockpiles and subsequently to solve coal blending problems, where varying qualities of stockpiled coals are normally used, an interactive computer program was developed. Simple file-handling, for stockpiling problems, and multi-objective goal programming technique, for blending problems, provided their solutions. The computer program was made suitable for use on both minicomputer and microcomputer. Menu-driven and interactive capabilities give this program a high level of flexibility that is needed to analyze and solve stockpiling and blending problems at the power plant.