Abstarct: Despite the fact that Iran is rich in construction rock resources, exports are minimum in this area as a result of the low quality and the high cost of the produced rocks. This problem can be solved to a great extent by employing advanced tools, research, and accurate experiments regarding the cutting process. In this regard, the evaluation of the abrasion of diamond disks in the rock-cutting process and specifying the cost-reducing and life-extending factors for the cutting tools is of utmost importance when designing a rock processing plant. Therefore, estimating the abrasion of diamond disks and specifying the parameters that are effective in the process of cutting hard rocks is necessary. The abrasion of diamond disks depends on the physical and mechanical properties of the rocks and the operational specifications of the cutting device. In the first part of the study, a new system of classification was proposed baxsed on the physical and mechanical properties of the rock, including four characteristics (Schmidt hardness, uniaxial compressive strength, the Los Angeles abrasion index, and the elasticity modulus). These characteristics were weighted using classical rock engineering systems and by the Expert Semi-Quantitative (ESQ) method. Considering that the experts' opinions are considered absolute and fixed and that the distribution of the opinions is not taken into account in the classical method, the proposed method models the experts' opinions as a correct scheme using fuzzy logic and the corresponding weight allocation. After selecting the fuzzy Gaussian membership function as the most fitting function, the optimum fuzzy membership function was investigated in order to relate the parameters that are effective on the abrasion of diamond disks. In the proposed system, abrasion is categorized using the four aforementioned variables by dividing them into fixed, comparative, and logarithmic intervals in combination with the genetic algorithm. baxsed on the results, the logarithmic comparative method is associated with a smaller error, thus selected as the optimal method. Using the proposed classification system, the abrasion of the diamond disks in the process of cutting hard construction rocks (output) was evaluated and classified into five qualitative intervals (very high, high, medium, low, and very low) for nine sample hard construction rocks. In order to validate the results of the proposed classification system another classification system was developed using the mextaheuristic artificial bee colony algorithm baxsed on experimental results and its output was compared with that of the proposed system. Moreover, in order to validate the results of the proposed classification system, multiple cutting tests were conducted under various operational conditions using a customized disk cutter with special capabilities for this reason. During the set of experiments, the macroscopic abrasion of the diamond segments was measured. Ultimately, baxsed on the proposed model, the nine studied rock specimens were categorized into three classes. Specimens 1 and 2 were categorized into class 2 with a small abrasion rate that shows a high cuttability. Similarly, the specimens 3, 4, 5, 6, 7, and 9 were categorized into class 3 with a medium abrasion rate (0.000005 to 0.000008 mm3) and favorable cuttability. Specimen 8 was categorized into class 4 with high abrasion rate (0.0000072 mm3) that shows the high abrasion and low cuttability of this specimen. The results suggest that, in most cases, the proposed classification system has been able to rank the abrasion of diamond saws in the process of cutting hard construction rocks with minimum difference between the estimated and measured abrasion.