OpenText 首頁。
Customer stories

Industry Mining LogoA global mining company

A global mining company accelerates research project timelines with OpenText Content Aviator

Industry Mining Logo

About the global mining company

This leading mining company extracts ore such as iron, nickel, and copper for worldwide distribution. The company is also engaged in the logistics and energy industries.
Metal mining image
  • Headquarters:
    Brazil
  • Key export:
    Iron ore
  • Net income:
    $80 billion USD

Summary

Challenges

  • Mining research projects slowed down by low-level manual work.
  • Growth and investment limited by lagging feasibility assessments.
  • In-house AI assistant lacked scalability for global reach.

Solution

  • Deployed a proof-of-concept project.
  • Partnered with AI experts to support best practices.

Results

  • Enabled more effective AI use via expert training
  • Boosted AI response accuracy by 47%
  • Unlocked rapid decision-making

Challenges

  • The company’s specialists were bogged down by low-level manual work, slowing mining research projects
  • Growth and investment were limited by lagging research and feasibility assessments
  • The company’s in-house AI assistant lacked scalability for global reach

This global mining company, headquartered in Brazil, works to provide the materials essential to human progress—safely and sustainably. Producing iron, nickel, copper, manganese, and more, the company undertakes careful research to assess both the viability and the social and environmental impacts of its mining operations.

Between on-the-ground geological studies, collaboration with local communities, and market and legal analyses, these research projects can take up to ten years. Typically, cross-functional teams spend weeks or months manually collecting and consolidating information for review as new product opportunities, market fluctuations, or environmental standards emerge.

The global mining company has recently scaled up its commitment to opening new mining projects, with investment rising around 10% year-on-year. However, it became clear that slow research and feasibility assessments were leading to missed investment opportunities. With a history of technological innovation to support its operations, the company sought a solution that could help its specialists cut down repetitive manual tasks, speeding up the research stage for both new and existing mines.

A spokesperson from the global mining company said, “Information about each mining project is typically stored in different formats across siloed systems, and accessing it takes up a significant amount of valuable time as workers search through mountains of documentation. AI, with its ability to quickly ingest and analyze large volumes of data, presented the perfect opportunity for reducing this manual work.”

Initially, the company aimed to develop an AI assistant internally based on ChatGPT, but quickly ran into problems with reliability, security, and scalability. As a long-term user of the OpenText™ Documentum™ Content Management platform, the global mining company saw the opportunity to try out OpenText™ Content Aviator™ with the OpenText Earn Your Wings program to help it speed up its journey to AI.

“Having support from a trusted vendor with a ready-made solution reassured us that we could achieve our ambitions,” explained the spokesperson. “Our key requirements included an assistant that could parse complex queries from experts and function in both English and Portuguese to maintain accessibility for global stakeholders.”

Hands carry the sand image

Compared to the months we had spent on developing a solution ourselves, the OpenText PoC was ready to be tested within just two days.

Spokesperson
Global mining company

Solution

To explore the possibilities of AI-assisted research projects, the global mining company joined the OpenText Earn Your Wings program and ran a proof-of-concept for OpenText Content Management integrated with OpenText Content Aviator.

Products deployed

Services provided

Deploying a proof-of-concept project rapidly

As part of the Earn Your Wings program, the global mining company received access to a secure OpenText Private Cloud sandbox environment with OpenText Content Management and Content Aviator to run a proof-of-concept (PoC) project. Reinforced with a retrieval-augmented generation framework, the OpenText solution was able to augment its output using the mining company’s domain-specific knowledge base, including legacy documentation and institutional knowledge that had previously been difficult to access or operationalize. The company gathered documents related to an existing mine, which were used to create workspaces in OpenText Content Management and integrated into the AI solution.

To compare the effectiveness and accuracy of the OpenText solution with its in-house AI assistant, the global mining company selected 18 predefined question types for Content Aviator based on the documents and prepared the corresponding answers that it expected to see. The questions focused on five key specialist areas: geological analysis, drilling and exploration data, mineral processing, social and environmental considerations, and project workforce and legal aspects.

“Compared to the months we had spent on developing a solution ourselves, the OpenText PoC was ready to be tested within just two days,” added the spokesperson.

 

Partnering with experts to support best practices

During the PoC, the global mining company worked closely with the OpenText Professional Services team. OpenText AI experts led workshops for teams across the company, explaining the possibilities of AI and LLMs and exploring potential business use cases.

Once the testing environment was set up in the cloud, OpenText assessed data quality and guided the company on selecting files that work best with Content Aviator, such as Microsoft Word documents, Microsoft Excel spreadsheets, and PDF files. Working through an iterative process, prompt engineers from OpenText helped refine the company’s queries, improving the accuracy of the AI’s response.

Metal mining image

We are now much more confident in approaching innovative AI solutions. The patience and professionalism of the OpenText team was unparalleled throughout.

Spokesperson
Global mining company

Results

Through close collaboration with OpenText, the global mining company successfully completed the PoC. By applying AI best practices, it’s unlocked the possibility of rapid and informed decision-making to make the most of investment opportunities.

Enabled effective AI use through expert training

For the global mining company, a key benefit of the PoC has been the knowledge transfer from OpenText AI experts to its employees. The company’s specialists learned best practices for using Content Aviator, such as using consistent capitalization and spelling out abbreviations when querying the AI assistant.

“The training we have received from OpenText Professional Services is crucial to our success going forward,” confirmed the spokesperson. “We are now much more confident in approaching innovative AI solutions. The patience and professionalism of the OpenText team was unparalleled throughout.”

Boosted AI response accuracy by 47%

Through careful prompt engineering and skill-building exercises, the company achieved a 47% boost in Content Aviator’s response accuracy.

The spokesperson said, “We were impressed by the high degree of accuracy in Content Aviator’s answers to questions that are specific to our industry. The OpenText solution also moved seamlessly between English and Portuguese queries without losing any of the robustness in its response—something we have not been able to achieve with our proprietary solution.”

Unlocked rapid investment decision-making

With Content Aviator’s successful PoC, the global mining company is currently evaluating next steps. Significantly, with the support of an AI solution for researching mining projects, it could cut out months of low-level manual work and be able to grow like never before.

For example, today, if the company wanted to assess the feasibility of using an existing mine to produce a different ore, it would have to hire a geologist to do around two months of work reviewing existing data. With Content Aviator, the relevant information could be gathered and consolidated in just a few hours, helping the company rapidly close in on viable investment opportunities and stay ahead of market shifts.

How can we help?