Investigation of the Random Value Statistical Distributions Models Choice Influence on the Mining Operations Modeling Results

Abstract

The importance of production planning for improving the performance indicators of a mining enterprise is indicated. The possibility of simulation modeling using for this aim is shown. It is shown that the created model has a large number of stochastic parameters. It is investigated that there is a problem of research lack about the choice influence of the mining modeling results with different statistical distributions. It is known that with an increase in stochastic deviations from the initial parameters, the productivity of queuing systems decreases. Purpose of work is to study this influence with four statistical distributions of a random quantity (uniform, normal, negative bi-nomial and Poisson distribution) for individual operations and their combinations. In addition, it is necessary to determine how much a change in one particular parameter will affect the overall result of the modeling. Materials and methods. In the previously created simulation model, a stochastic delay is added to the time of individual operations. The addition of such a delay with different sta-tistical distributions and with the same mathematical expectation is investigated. The simulation re-sults are compared with each other, for each individual operation the absolute and relative devia-tion of the results is shown. Further, a similar simulation is performed when all the simultaneously selected parameters changing. Result. It is shown that the magnitude of the deviation significantly differs among all deviations. It is shown that for various single changes in operations, the largest and smal-lest deviations can be given by different statistical distributions. To study the joint change with all parameters, 3 modeling scenarios are implemented: all uniform distributions (this case is used now), the scenario with the smallest deviation and the scenario with the largest deviation. It is shown that switching to another scenario leads to a significant change in the simulation. Conclusion. It is con-cluded that the used significant influence of statistical distributions choice to the accuracy of model-ing the operation of the mining machine is shown, especially when they are taken into account to-gether. The results can be used to clarify the influence of individual factors in the simulation model and improve the planning of potash mining operations, for individual mining machines too.

Mining Machine Simulation Model Refinement According to the Engines Energy Consumption Data

“…Как известно, результаты имитационного моделирования могут в значительной степени зависеть от выбранной модели случайного распределения [14]. Из формы графика можно сделать предположение, что из исследованных нами распределений [15] Примем уровень значимости равным 0,05, найденное значение χ 2 -статистики для 8 степеней свободы (10 независимых интервалом минус 2 независимых переменных) равно 15,507. Вычисленное χ 2 = 13,936 <15,507, поэтому получившееся распределение может быть отрицательным биномиальным.…”