1
25-054
Leveraging AI for Improved Contracting and
Procurement in Mining
Paul Culvenor
Hevi Pty Ltd, Brisbane, Australia
Brad Gyngell
Hevi Pty Ltd, Sydney, Australia
ABSTRACT
Ensuring effective management of contractual processes is
essential for maintaining profitability and operational effi-
ciency in mining. However, due to limited capacity and
resources, sites often struggle to manage these processes
effectively, leading to oversights and financial losses.
Recent advancements in Artificial Intelligence (AI) offer
promising solutions to these challenges. Large Language
Models (LLMs), like those powering ChatGPT, now enable
machines to understand unstructured data like contracts.
These can therefore be applied to address these challenges
by taking some of the heavy lifting off of humans.
In this paper, we analyze a real-world case study of
the application of LLM tools for preventing a $400k con-
tractual issue on an underground gold mine in Western
Australia.
We provide a gap analysis of the failure modes in exist-
ing processes that led to the issue. We then describe the
application of LLM tools to flag it early and facilitate a col-
laborative solution between the operator and the supplier.
This includes a comparative analysis of base case vs LLM
case to evaluate the impact on project cost and time.
Finally, we discuss how the integration of AI can fos-
ter improved relationships between clients and contractors
across a broad range of potential contractual applications,
allowing the time required for collaborative and proactive
problem solving.
INTRODUCTION
In a Western Australian underground mining project, the
contractor had been performing well under a development
contract, consistently meeting key performance indicators
set by the client. Both parties were satisfied with the overall
performance and profitability of the project, with the con-
tractor maintaining a steady 15% margin.
However, a shift in ground conditions within a particu-
lar development drive severely impacted productivity over
a two-month period which led to significant contractual
issues.
PROBLEM IDENTIFICATION
The poor ground conditions included:
• Increased jointing and fracturing
• Decreased compressive strength
• Inclusion of a major slip plane dipping in the oppo-
site direction to development
In response to the deteriorating conditions, the project
team implemented the following strategies:
• Increase roof and wall support
• Decrease depth of cut
• Increased number drill holes
These actions led to reduced development advance rates
and increased cost of consumables like bolts and mesh.
The contract was structured such that the contractor
performing the work was responsible for their own costs
and was paid per meter of development. Therefore, they
were fully exposed to the increase in costs, while simultane-
ously experiencing a decrease in revenue. This led to the
contractor’s profitability swinging from a 15% margin to
a 10% loss.
MISSED CONTRACTUAL OPPORTUNITY
Despite early signals of reduced productivity due to dete-
riorating ground conditions, the site’s project management
team failed to identify that these conditions qualified as a
“latent condition” under the terms of the contract.
The latent condition clause stipulated that the contrac-
tor would have been entitled to receive extra payment from
25-054
Leveraging AI for Improved Contracting and
Procurement in Mining
Paul Culvenor
Hevi Pty Ltd, Brisbane, Australia
Brad Gyngell
Hevi Pty Ltd, Sydney, Australia
ABSTRACT
Ensuring effective management of contractual processes is
essential for maintaining profitability and operational effi-
ciency in mining. However, due to limited capacity and
resources, sites often struggle to manage these processes
effectively, leading to oversights and financial losses.
Recent advancements in Artificial Intelligence (AI) offer
promising solutions to these challenges. Large Language
Models (LLMs), like those powering ChatGPT, now enable
machines to understand unstructured data like contracts.
These can therefore be applied to address these challenges
by taking some of the heavy lifting off of humans.
In this paper, we analyze a real-world case study of
the application of LLM tools for preventing a $400k con-
tractual issue on an underground gold mine in Western
Australia.
We provide a gap analysis of the failure modes in exist-
ing processes that led to the issue. We then describe the
application of LLM tools to flag it early and facilitate a col-
laborative solution between the operator and the supplier.
This includes a comparative analysis of base case vs LLM
case to evaluate the impact on project cost and time.
Finally, we discuss how the integration of AI can fos-
ter improved relationships between clients and contractors
across a broad range of potential contractual applications,
allowing the time required for collaborative and proactive
problem solving.
INTRODUCTION
In a Western Australian underground mining project, the
contractor had been performing well under a development
contract, consistently meeting key performance indicators
set by the client. Both parties were satisfied with the overall
performance and profitability of the project, with the con-
tractor maintaining a steady 15% margin.
However, a shift in ground conditions within a particu-
lar development drive severely impacted productivity over
a two-month period which led to significant contractual
issues.
PROBLEM IDENTIFICATION
The poor ground conditions included:
• Increased jointing and fracturing
• Decreased compressive strength
• Inclusion of a major slip plane dipping in the oppo-
site direction to development
In response to the deteriorating conditions, the project
team implemented the following strategies:
• Increase roof and wall support
• Decrease depth of cut
• Increased number drill holes
These actions led to reduced development advance rates
and increased cost of consumables like bolts and mesh.
The contract was structured such that the contractor
performing the work was responsible for their own costs
and was paid per meter of development. Therefore, they
were fully exposed to the increase in costs, while simultane-
ously experiencing a decrease in revenue. This led to the
contractor’s profitability swinging from a 15% margin to
a 10% loss.
MISSED CONTRACTUAL OPPORTUNITY
Despite early signals of reduced productivity due to dete-
riorating ground conditions, the site’s project management
team failed to identify that these conditions qualified as a
“latent condition” under the terms of the contract.
The latent condition clause stipulated that the contrac-
tor would have been entitled to receive extra payment from