With shallow resource extraction diminishing, miners are increasingly required to tackle deeper and more geologically challenging ore bodies, meaning proper mine scheduling is becoming even more critical.
Encompassing resource assessment and guided by thorough due diligence, effective project planning is mandatory when overseeing the life of any given operation.
Additionally, it is a useful tool for de-risking any potential adverse technical or environmental issues that may occur.
Not surprisingly, engineering also plays its part in this process as it helps underpin the creation of detailed logistical designs for a mine’s layout – including the proposed ore extraction method, its fleet access routes and the integration of the processing circuits.
Meanwhile, scheduling, budgeting and regulatory compliance focus on planning the sequence of extraction allocating the various resources and ensuring that all relevant environmental and safety regulations are met.
In this regard, measures related to waste management, reclamation and emergency response strategies should be in place.
According to industry software leaders, the key elements of planning are establishing goals, breaking down resource allocation, creating a realistic timeline as well as taking risk management, communication and output into consideration.
It is a process that entails giving each task or activity, as determined during due diligence, a start and end date. To meet a project’s overall deadlines, a schedule indicating when each job has to be completed should be generated.
In this regard, task sequencing, duration estimation, resource allocation, critical path analysis and the ability to make adjustments are of the utmost importance.
Furthermore, miners must comply with applicable legislation, meet all relevant standards and adopt the appropriate policies and procedures.
Completing period reports (daily, weekly and monthly) detailing scheduled versus actual maintenance activities and resource usage is also essential.
Ultimately, the result of any planning process should be a comprehensive project guide that serves as a roadmap for finishing tasks on time and aiding with the effective tracking of the advancement, and allocation, of resources.
According to the academic community, there are still a number of improvements that can be made when it comes to mine scheduling.
Melbourne University’s Professor Adrian Pearce, for instance, believes current practice and software tools do not provide adequate support for miners to properly analyse the impact of decisions made for one planning horizon on others.
As a result, a new generation of scheduling technology is required to realise effective multi-level management.
Fast and automated tools, Prof Pearce said, provided a means of adapting plans in response to the new data as it arrived.
In turn, this led to “interesting research” questions which addressed issues pertaining to variations in predictions and their uncertainty.
To tackle this issue, the university’s Faculty of Engineering and Information Technology has been developing algorithms that can underpin a series of robust plans within reasonable time frames and draw upon the fields of robust optimisation, simulation boundaries and contingency analysis.
“Mine planning is complex with short, medium and long-term objectives that are often competing rather than complementary,” Prof Pearce noted.
“Increasing capability in data collection and management, moreover, is enabling faster and more frequent updating of orebody models.
“Fast, and predominantly automated, scheduling tools can take advantage of these frequent model updates by developing new, or adjusting existing plans, as required.
“This contrasts with current practice in which the time-consuming nature of planning prevents this kind of agility.”
While this required an across-the-board cutting-edge approach, it was only three years ago that Epiroc’s David Olsson Granlund, in his Computer Science and Engineering master’s thesis for Sweden’s Luleå University of Technology, pointed out that mine scheduling was typically carried out manually using some sort of planning tool.
This task, however, could quickly become difficult and time-consuming as a result of several complicated factors and constraints.
For example, Granlund said, most of the tasks required specialised equipment which may be shared across multiple headings (such as a passageway in an underground mine).
“The complexity of the problem will grow to test the capabilities of even an experienced human planner,” he noted.
“The underlying scheduling problem is notoriously hard to solve optimally for large instances.
“To mitigate the performance degradation for large scheduling instances, an iterative solver strategy involving the use of OR-Tools (an open source software that tackles combinatorial optimisation problems and includes specialised solvers for constraint programming, linear and mixed-integer programming, vehicle routing and graph algorithms) and automation could be implemented.
“With this strategy, the scheduler is able to solve much larger instances and initial tests resulted in the same objective values as the optimal strategy.”
Although rescheduling procedures supported scheduled maintenance due to unforeseen circumstances such as delays or machine breakdowns, it was concluded that while automatic scheduling and rescheduling were feasible, they first needed to be evaluated by experienced schedulers in the field before being applied in a production environment.
“There is no objective measurement for how ‘good’ a schedule is as long as it satisfies the goal tonnage which makes the implementation difficult to evaluate,” Granlund said.
“Therefore, it would need to be evaluated on a case-to-case basis together with a human scheduler, both to find the correct configuration and to see if the automatically generated schedules break any logical rules that may require new constraints on the model.
“In the performance department, the scheduler would benefit from more extensive and thorough tests to get a better understanding of what affects the running time.
“This could also provide information on what parts of the implementation need to be further optimised.”
Taking this on board, tests might include using different resource set-ups to determine the effect on resource-abundant versus resource-constrained schedules, increasing the number of workers on hardware with more virtual centralised processing units, varying the time granularity or testing the sub-goal deviation objectives.
