heuristic algorithm

Loading and haulage operation in open pit mines is the last stage of the mining process. truck- shovel system, due to its many advantages  including high flexibility, is preferred for this operation. Due to high operating costs, proper fleet management and optimization can significantly affect the project economics. Truck allocation and dispatching issue is a very complex problem, especially in large mines with numerous loading and dumping points. Because of the problem size and complexity, employing mathematical methods is not justified due to very high solution time which leads to employing super computers. To overcome the aforesaid shortcoming, heuristic algorithms can be applied. In this paper, in MATLAB environment, a heuristic algorithm was developed to solve allocation and dispatching problem of transportation fleet of a real mine. According to the obtained results, a running time of 39 seconds was computed for the heuristic algorithm. Finally, the same problem was solved with an available mathematical model with a running time of 24 hours which shows the superiority of the proposed algorithm over the mathematical modeling.

Comminution is the most energy intensive operation which constitutes the major portion of operating and capital costs of the mineral processing plants.  Working at the maximum operating capacity of comminution equipment plays a significant role in the efficiency of the circuit.  Also, due to the effect of crusher efficiency on the downstream circuit performance, optimization of the crushing circuits has received considerable attention.  In this research, the effect of feed chute design on tertiary cone crusher performance at the Sarcheshmeh copper complex was studied.  A close monitoring of the performance crusher revealed that main problems were high fluctuations of power draw and uneven and high-rate wear of crusher liners.  Such pitfalls were clear evidences of an improper feeding arrangement into the crusher.  Accordingly, various feed chute designs were employed in the simulations by an in-house developed DEM software called KMPCDEM© to find more uniform feed distribution on the distribution plate of the crusher.  Results showed that by changing the shape of feed chute from cubic to cylindrical, decreasing its surface area from 0.34 to 0.24 m2 and increasing the cylinder length above and below the feed chute plate from 0 to 45 cm and from 53 to 95 cm, respectively, uniform feed distribution was obtained.  After installing the new feed chute design in the plant, a detail monitoring over a period of 15 months showed a reduction of the standard deviation of crusher power draw from 13 to 3 kW.  A better crusher control caused choke feeding.  Therefore, 36% increase in the crusher throughput and finer and narrower product size distribution occurred.  Furthermore, the life of crusher liners increased from 8 months to 15 months on account of more uniform and lower rate of wear on mantle and liners.

SummaryThe aim of this paper is to present a new algorithm to determine ultimate pit outline and mining sequence simultaneously based on the maximization of the net present value (NPV). For this purpose, a nonlinear binary mathematical model was established and then a heuristic algorithm was developed to solve this NP-Hard problem. IntroductionThe ultimate pit limit is an important problem which is determined by maximization of undiscounted profit or NPV. The floating cone algorithm and its modified versions, Korobov algorithm, Lerchs-Grossman method and maximal flow algorithm were developed to generate ultimate pit limit based on the maximization of the undiscounted profit. Nevertheless, it is better to determine the pit outline based on the maximization of NPV. To achieve this goal some algorithms like Wang-Sevim, Latorre-Golosinski and Roman were established.Methodology and ApproachesThe binary and nonlinear mathematical model to determine the ultimate pit limit on the basis of maximizing NPV and a few suggestions for its linearization were presented. Afterwards, by defining the concepts of downward cone, positional weight and nearest ore index, a heuristic algorithm was developed to determine the ultimate pit limit and mining sequence all together.Results and ConclusionsThe algorithm was applied for 2D and 3D block models and the results showed that it is able to produce optimum outcome. Complexity of the algorithm is low and easy to use and as well as for education purpose. It is also able to consider variable slopes and grade-based constraints for production planning in the algorithm.

Determination of the ultimate pit limit is one of the most important design in open pit mining. In the present methods, the final pit is determined based on the maximization of undiscounted profit. For this purpose, various mathematical, heuristic and meta- heuristic algorithms have been developed. Among these methods the most used one is the Lerchs-Grossman (LG) algorithm. It has been mentioned that it is better to design the ultimate pit based on the maximization of the net present value. In other words, the optimal arrangement of blocks and the final pit outline must be determined simultaneously. For that reason, in this paper the binary and nonlinear mathematical model of this problem and some suggestions for its linearization have been presented. In addition, in order to reduce the number of decision variables the concepts of the earliest and the latest possible time for extracting a block have been defined. In the following, with the use of downward cone and positional weight and the earliest extraction time of ore blocks, two heuristic algorithms have been developed for simultaneous determination of the final pit and blocks extraction sequence. At the end the results of these algorithms have been compared with the LG algorithm. The results show that these algorithms are capable to produce good result.