Search

International audience; as open-pit mines consist in the accumulation of ore layers of different characteristics, their extraction process results from the iteration of a sequence of elementary operations performed on parcels by specialized or multipurpose machines. Medium-term extraction programming consists in deciding the assignment of the available machines over time. This problem is addressed through Discrete Event Simulation implemented in a phosphate mine, taking into account all the extraction constraints to drive the extraction process. This combines an effectiveness criterion (degree of urgency of obtaining certain ore qualities) and an efficiency criterion (machine displacements). While the parameterization used to control simulation quickly produces many technically feasible scenarios, making the final optimal choice is extremely difficult, particularly as one needs to take into account time considerations in extracted ore progressive provisioning and machine use. Our paper describes the bases of a visual decision-support system (DSS), currently under development, which enables the processing of results to compare multiple scenarios and adjust parameterization in order to come up with a good solution. This web-based DSS is coupled to the simulator via a relational database.

This paper presents the architecture of an interactive scheduling decision support system (ISDSS) allowing users to find the optimal solution for fertilizer production on parallel heterogeneous processors. The proposed approach takes into account different production process constraints such as launch time, delivery date, preventive maintenance and the impact of scheduling on supply chain management. The ISDSS implemented is run by a relational database used to customize the structural data and the problem parameters. A user interface is available for ISDSS users to define the scheduling problem and design the solution based on tables and graphs for detecting possible issues. The ISDSS architecture was implemented on java using independent modules.

Open-pit deposits are often characterised by a stack of layers of different geological nature. Some layers are worthless while the ore of the others is of a varying economic value depending on grade. To reach a layer, it is necessary to have first removed the upper layers above the extraction zone. This action results in uncovering the layer in this particular place and in facilitating access to the layers below. This extraction process involves a series of 2–7 basic operations; each one is performed by a machine, some of which are able to perform up to three different operations. Ensuring the consistency of mining extraction scheduling over a few months, in order to meet known or forecast demand, is a challenging task. A mining extraction model based on mathematical programming is proposed but it is hardly usable, due to its size. Therefore, a model based on a Discrete Event Simulation, is created to test how ore supplies are affected by the tactical and operational decisions relating to the choice of parcels to be processed and to the allocation of machines to the different basic operations