Application Of Dynamic Programming To The Grade Control And Scheduling In A Cut-And-Fill Mining Method

The first application of dynamic pro¬gramming to the problem of grade control and scheduling was by Muge et al (1979) and Ribeiro (1982) who applied it to the simple, two-dimensional case of longitudinal open stoping using two sublevels. In this paper the formulation of the problem is extended to cover the general case applied to a cut-and-fill mining method. The criterion for optimization is the total deviation of the average, grades of mining blocks from the overall average panel grade (or the average grade of a set of mining panels or the mine processing working point). This means to schedule the operation by removing a certain number of mining blocks at a time, subject to the geometric (mining) constraints, and such that the total deviation of average block grades from overall average grade is minimum. The method can be applied to complex orebodies with several ore characteristics and the deviation of each one of these characteristics can be further constrained by specifying maximum values. The method can also be generalised in order to take into account any necessary geometric constraints used to define the mining method. In this paper a case study is presented concerning the mine planning of a complex sulphide orebody located in the South of Portugal. As input for the definition of the mining sequence, a geostatistical numerical model of the orebody was built based on a previous ore typology. The short term planning is used to assess specific mining sequences and the production scheduling is performed on the basis of an objective function (e.g. minimization of grade fluctuation).