Machine learning assisted domain modelling
Machine learning assisted domain modelling
Maptek DomainMCF uses machine learning to generate domain boundaries direct from sample data for rapid creation of resource models. Geologists feed in drilling or other sampling data and obtain domain or grade models in dramatically less time than traditional resource modelling methods.
The Maptek Compute Framework moves away from complex software installed on desktop computers and shifts costs from capital to operating budgets. Faster, secure cloud processing is supported by flexible licensing.
Generate resource models in minutes or hours, depending on size, data density and complexity of the deposit.
Many operations only have the ability to update their resource models once or twice a year. Using DomainMCF, your project can be modelled in minutes with results comparable to classical techniques. You remain in control of the process without the onerous preparation work. New data can be added and models regenerated quickly to reflect the current data.
Streamline workflow - data errors are highlighted and the modelling job cannot be submitted until changes have been made to remedy these.
DomainMCF solves another challenge - having valid geological data to input into the resource model. Data validation is completed within DomainMCF before the modelling phase.
The new geological domain modelling process using machine learning is especially suited to large volumes of data such as later stage exploration projects and operating mines. Your operation will benefit from multiple solutions run in parallel using high performance cloud computing.
Produce resource models up to 2000 times faster and more cost effectively than other solutions.
Generate resource models with certainty and report investment options to stakeholders.
Cloud based processing and machine learning save time for resource modelling.
Repeatable and reliable results with the ability to easily update resource models.
Identify potential projects and accurately interpret the volume of geological data for targeting high value projects.
Provide mine planners and potential investors with a quantitative assessment of risk due to geological uncertainty.
Build on success
Apply professional expertise to interpretation and evaluation, supported by automated machine learning approach.
Cloud processing solution without onerous start-up or customisation. Start generating models within minutes.
Maptek DomainMCF provides a data-driven process to generate alternative models to satisfy JORC reporting standards.
Machine learning techniques use all data sources for geological modelling, enhancing the understanding of complex deposits and improving decision making.
Technological advances have matured machine learning to a point where it can be readily and practically applied by solutions such as Maptek DomainMCF.
Generating models that include a measure of uncertainty leads to better informed decision making and compliant resource statements.
Maptek has developed a new solution for evaluating projects, progressing from database to resource report in 30 minutes.
Machine learning techniques trialled alongside traditional resource modelling at an underground metals mine demonstrates future benefits.
DomainMCF can change the way an operating mine uses geological and geotechnical models to keep information up to date in a production environment.
The new Maptek geological domain modelling process employing machine learning has delivered accurate shapes and volumes in much shorter time.
Machine learning engine for domain modelling is able to identify faulted geology in record time.
Applying machine learning to model marble reserves resulted in faster results and more uniform quality classifications to guide extraction.
This 2019 paper introduces you to orebody modelling of the future. Learn how machine learning allows rapid generation of domained orebody models, including the estimation of multiple numeric variables and uncertainties, directly from drillhole data.
This 2021 paper outlines the DomainMCF modelling approach to the spatial distribution of marble quality categorisation parameters and compares it to marble product classifications generated by the conventional estimation method.
This paper discusses the factors accentuating complexity in deposit modelling: data diversity, structural controls, chemistry, data volumes, process workflows and external non-geological constraints. A case study illustrates the risk of ignoring complexity, which can result in an overly simplified geological model.
This paper outlines the results of the Nova-Bollinger modelling trial conducted by the mine geology team for mineral resource estimation. Advantages and disadvantages of implicit modelling and machine learning methods for preparing mineral resource estimation domains are covered.
September 30, 2022
See how using machine learning to generate blocks models improves productivity through faster decision making. A case history demonstrates the way DomainMCF quickly generates models, allowing you to interrogate and analyse alternative interpretations.
April 27, 2022
Technical Lead Steve Sullivan chats about how the breadth of DomainMCF applications has surprised even its creators, less than a year after the machine learning engine’s release.
March 31, 2022
Maptek Senior Technical Sales Specialist & Technical Lead – DomainMCF Steve Sullivan presents Harnessing data complexity by using machine learning for rapid geological modelling for the AusIMM International Mining Geology Conference 2022.
September 19, 2021
Technical Lead, DomainMCF Steve Sullivan presents Recognising the impact of uncertainty in resource models.
April 7, 2021
Hear directly from Gretchen Moore, Senior Rock Mechanics Engineer at Sibanye Stillwater, on her experience using DomainMCF to create geological models from Diamond Drill Hole data.
March 5, 2021
From a small office offering geological database and plotting services, Maptek has grown to a global technology business with more than 20,000 users spanning 90 countries.
March 4, 2021
Deposits are complex by nature, but a lack of time and resources prevents us from interpreting their richness and complexity in full. This open-forum webinar discusses the modern domain modelling challenges geologists are facing today, and how harnessing technology is key to overcoming them.
November 25, 2020
Machine learning is already providing breakthroughs in rapid, accurate resource modelling. Learn how being transparent about uncertainty actually improves confidence. The recording includes the Q&A from two sessions.
July 27, 2020
This webinar outlines the power and functionality of DomainMCF for resource modelling. A short video is followed by the Q&A session from the APAC-Americas session (from 16:05), and then the Europe, Middle East and African session questions (from 54:40).
July 6, 2020
DomainMCF is a new paradigm for domain modelling which applies machine learning to rapidly create resource models.
June 15, 2020
An expert panel featuring Penny Stewart, Hugh Sanderson and Christie Myburgh explores the issues that arise when we apply machine learning to mining applications. These include how we harness the vast volumes of data made available from mining processes, transform it into knowledge and apply that knowledge to continually improve. See more Maptek Forums 2020 here: www.maptek.com/forums/.
June 15, 2020
DomainMCF is an exciting new approach to resource modelling which takes advantage of machine learning to help geologists to quickly and accurately model geologic domains. Sites can incorporate new drilling and other exploration data into the operation faster than ever before.