aalto1 untyped-item.component.html
Controlling short-term deviations from production targets by blending geological confidence classes of reporting standards
Loading...
URL
Journal Title
Journal ISSN
Volume Title
Insinööritieteiden korkeakoulu |
Master's thesis
Unless otherwise stated, all rights belong to the author. You may download, display and print this publication for Your own personal use. Commercial use is prohibited.
Authors
Date
Department
Major/Subject
Mcode
R3008
Language
en
Pages
71+2
Series
Abstract
Meeting short-term production targets is desired by many companies, since this would enable them to finetune
the processing operation,meet budget plans and obey contract requirements. Recently stochastic optimization
solutions have been developed requiring geostatistical simulations as input. The significant value
added has been demonstrated, however, an operational implementation of such approaches for day-to-day
use is complex and seems currently difficult as it requires expert knowledge and extensive computational capacity.
To control the short-term deviations, a new fast metaheuristic scheduler is developed that blends Geological
Confidence Classes (GCC’s) from resource reporting standards. For the scheduler, a new penalty function
is developed to schedule for a target blend of GCC’s and a new method is developed to enforce smooth
mining patterns in three dimensions. The metaheuristic solver uses a Genetic Algorithm and an Ant Colony
Optimization algorithm to efficiently converge towards the Pareto optimum. To establish an optimal blend
of GCC’s, a methodology is developed which creates a range of equally probable scenarios of deviations from
production targets for different blends of GCC’s. A least-squares estimate can be fitted to these scenarios at
the required level of confidence to determine the optimal blend for a maximum allowed deviation.
An historical world class gold deposit is used to show that the monthly and quarterly deviations can be
controlled by blending GCC’s. Furthermore, the case study shows the possibility to establish an optimal blend
of GCC’s by using the developed methodology. The scheduler proofs to be able to efficiently create and evaluate
schedules to blend the GCC’s for this case study. For a maximum quarterly deviation of 15% at a 90%
confidence level, the established optimal blend is 59% ore tonnage classified as measured resources. For the
monthly deviations, a maximum of 15% is too low and cannot be met at a 90% confidence level.