Forge - September 2009

Recent PRIMO case study, showing DOT for the decline and PUNO for the level layouts

Access optimisation for mine planning

Good mine planning is fundamental to maximising the value of a mine. A network design team from 3 universities has been developing access optimisation software for more than 10 years, based on new mathematical algorithms. Vulcan is used to visualise complex orebodies and create barrier files.

Part of the AMIRA project, PRIMO seeks to integrate optimal stope design, infrastructure design and scheduling for underground mines. The DOT and PUNO tools have been further developed as part of this project.

The access optimisation software

• has been proven by case studies
• facilitates minimisation of both development and haulage costs
• enables cost savings of up to 10% over the life of the mine

The team developed DOT (Decline Optimisation Tool), allowing a mining engineer to design an optimal network of declines, satisfying mine constraints and operational requirements while minimising development and haulage costs.

A related software product, PUNO (Planar Underground Network Optimiser), designs an optimal layout of level development that minimises the haulage and development costs for ore extraction on each level. It is also able to manage the requirements of different mining methods.

Traditional decline design for mine access and haulage relied on the engineer’s experience and skill. Often only 1 or 2 options were tested in order to limit the detailed repetitive work. Moreover, the emphasis was on minimising development costs rather than the combination of development and haulage over the life of the mine.

The new software tools allow an engineer to quickly and effectively test multiple options to determine the optimal decline and level layout designs, thus minimising the final combination of development and haulage costs.

DECLINE OPTIMISATION TOOL

DOT aids mine planning engineers to design a near-optimal network of navigable declines in an underground mine, servicing a given set of access and haulage points associated with an orebody.

1. DOT finds a network of decline paths avoiding user-specified no-go regions.
2. DOT takes into account operational constraints such as gradient, turning circle, straight sections at junctions.
3. DOT will also determine near-optimal locations for breakouts of subsidiary declines in the network, given the network topology.
4. DOT can automatically select the best choice of breakout points, where multiple options are available.
5. DOT returns centreline strings of the optimal decline, and a summary report showing selected points with development and haulage costs.
   

PLANAR UG NETWORK OPTIMISER

PUNO connects access point(s) on a crosscut or decline to a network of ramps and tunnels for level development. Specified points on the stopes are connected to footwall or hanging wall drives and vent raises, with subsidiary ventilation networks, as required by the mining method.

As with DOT, PUNO constructs a least cost network, taking into account development and haulage costs. PUNO produces a convenient template for mine designers to add details such as loading and passing bays.

1. PUNO can be given a selection of access points from the cross-cuts or on declines. It will then generate a design for each such access point.
2. Running DOT and PUNO in conjunction, the total cost of level development and network of declines can be computed for a range of access points. Additional savings can be found by approaching the ore zone from different directions.

Designs can be easily updated to take into account upgraded cost or geological data. Evaluation of multiple designs from DOT and PUNO can aid in decisions on the mining method, cut-off grade selection, and optimisation of scheduling.

DOT and PUNO work particularly well in the Vulcan environment. Specifically, barrier files to avoid the orebody, faults and old workings, can be generated readily using Vulcan functionality.

The team has added special features enabling the operator to adjust part of the design, keeping the rest fixed. This is helpful when the search space is very large, e.g. if there are many variable junctions in the network, or if the barriers are especially complex. The Vulcan environment enables the operator to quickly assess which parts of the network need additional optimisation or guidance to move away from barriers.

DOT AND PUNO HAVE BEEN PROVEN BY FIELD-TESTING FOR LOCAL
AND INTERNATIONAL MINING COMPANIES....

Thanks to Dr Marcus Brazil, Dr Peter Grossman, Professor Doreen Thomas & Professor Hyam Rubinstein (Melbourne University, Australia); Emeritus Professor David Lee (University of South Australia); & Professor Nick Wormald (University of Waterloo, Canada), and to the PRIMO sponsors for support of the project.