A Comparison Between ARMPS and the New ARMPS?LAM Programs

In previous research, the laminated overburden model from the LaModel program was effectively integrated with Analysis of Retreat Mining Pillar Stability (ARMPS) through ARMPS-like input to create a laboratory version of the new ?ARMPS-LAM? program. This program takes the basic ARMPS geometric input for defining the mining plan and loading condition, and then automatically develops, grids, runs, and analyzes a full LaModel analysis of the mining geometry to output the section stability factor (SF), all without further user input. The initial ARMPS-LAM results were encouraging with a case history classification accuracy of 55% to 71%; however, a few of the input variables that were nominally included in the SF calculation showed independent significance in the classification accuracy. Therefore, in order to further improve the accuracy of the ARMPS-LAM program, an investigation of the SFs calculated by the new ARMPS-LAM program and the ARMPS program is detailed in this paper. The initial results of a linear correlation between the ARMPS-LAM SF and the ARMPS SF showed a strong correlation (R2 = 0.88), with the ARMPS-LAM SF averaging about 8% higher. The difference in SFs between the programs were further investigated, and, ultimately, the results indicated that the laminated overburden model as implemented in ARMPS-LAM distributes relatively more load on the section pillars for depths less than about 1,000 ft and less load for depths more than 1,000 ft. The results of this research highlight the potential for improving the ARMPS-LAM program in the future by implementing a more accurate loading calculation on the section pillars.