A Simulation-based Optimization Approach for Material Dispatching in Continuous Mining Systems

摘要

This paper examines a problem related to dispatching materials to spreaders in coal (lignite) mines operated under the paradigm of continuously excavated material flow. In the real world, complexity in analyzing such systems emerges from numerous factors, including several random variables and frequent changes of extracted materials. Most of the mathematical programming approaches are limited by the amount of the decision variables. Indeed, simplifying assumptions should be made about these factors to develop a manageable mathematical model. In this study, a new simulation-based optimization approach is proposed that can accommodate most of these factors. This approach consists of running alternately a deterministic optimization model and a stochastic simulation model. It combines simulation, transportation problem, and job-shop scheduling problem. The transportation problem provides a mechanism to optimize dispatch decisions. In other words, it finds optimal connections between excavators and spreaders. Because of the nature of the transportation problem, it is possible to have multiple connections for an excavator. Therefore, the job-shop scheduling problem deals with the allocation of spreaders to different excavators over time. Its objective is to find the processing sequences and starting times of each operation on each spreader, in order to minimize the total weighted tardiness. Finally, the simulation uses the dispatch decisions generated by optimization and computes particular performance indicators. The calculated values are then introduced into a control module. The control module suggests refinements to parameters of the optimization model (e.g. transportation costs, jobs order, and jobs weight). The iterative process ends after a stopping criterion is met. The proposed approach is tested on a large continuous mine under given different dumping sequences, and results are reported. The merits and limitations of the proposed approach as pinpointed and farsighted operations management are discussed.