Abstarct: The main purpose of blasting operations in open-pit mines is rock fragmentation in order to obtain the required dimensions for the crushing system, which leads to optimum mining costs. In order to obtain the optimum dimensions, consideration of a precise relationship between blasting pattern and fragmentation process, which can be described through size distribution, is compulsory. There does not exist a global model to propose a capable fragmentation distribution model to establish in all open pit mines. The main objective of this thesis is the evaluation of fragmentation size distribution using Rosin-Ramler formula and image analysis technique. In order to attain this purpose, required data including the photographs of fragmented rocks and other geological data were collected after each blast operation in Jajarm bauxite mine. The image analysis process was carried out by Split-Desktop software in order to find fragmentation size distribution, uniformity index and the average size of the fragmented rocks. The results illustrated that there exists a significant difference between the experimental models and image analysis outputs. Hence, a new regression model was proposed in this regard and studied carefully. It was observed that the proposed model has been able to release reliable results for uniformity index and mean fragment size compared with image outputs and as a result, it was considered as a new predictive model for Jajram bauxite mine. Also, four different empirical blasting patterns were implemented in this mine in order to find an optimum pattern considering the final fragmentation size. By implementation of the fourth pattern It was concluded that not only 79 percent of fragmented rocks have been passed through jaw crusher but the mining cost of bauxite (per tonne) has been reduced compared with the average cost of mining costs in Jajarm bauxite mine.