Implementing a Predictive Maintenance Platform for an Iron Mining Company

LeverX reimagined the technical maintenance procedures for a large mining manufacturer by creating and integrating a full-scale predictive maintenance solution into their workflows.

Client and Challenge

Our client, one of the largest iron mining companies in their country, faced significant production losses due to frequent equipment failures and malfunctions. The company needed a solution that would allow them to detect early signs of equipment failures, prevent them, and thus ensure a high level of equipment reliability and performance. The main goals were:

• To reduce production downtime and financial losses associated with equipment breakdowns.
• To minimize technical maintenance costs.
• To ensure workplace safety.

At the discovery phase, we understood we couldn't build a new solution on top of the existing one due to incompatibility issues. The optimal solution was to migrate data from their legacy ERP to its modern variant, which we customized with predictive maintenance functionality. This presented two key challenges: data migration and synchronization, and slow performance. The legacy system couldn't handle the large amounts of condition data from multiple machines simultaneously, which made it impossible to implement the predictive maintenance functionality.

Methodology and Approach

We used a flexible Agile methodology, working in a highly dynamic environment and maintaining constant synchronization with the client. This approach allowed us to effectively manage the project, even during remote work in the pandemic.

Our solution is based on the Reliability-Centered Maintenance (RCM) approach, which helps enhance equipment uptime and reduce the need for asset replacements. We developed software that allows for creating maintenance plans based on RCM principles, including:

• Composing an equipment catalog and assigning criticality ranks.
• Developing maintenance strategies to reduce the likelihood of failure.
• Creating process charts with maintenance instructions.
• Drawing up maintenance plans, which were transformed into shift tasks for repair personnel.
• Collecting failure statistics and analyzing their root causes for continuous strategy improvement.

At the MVP stage, we used Microsoft Excel spreadsheets to draft repair and inspection plans. This simple method helped us identify and remove redundant operations that didn't reduce risks, which improved the quality of maintenance strategies and lowered the number of accidents and maintenance costs.

Solution

Ultimately, we delivered a full-scale predictive maintenance platform.

Key modules of the solution:

• Intelligent risk assessment tools. The client can conveniently monitor risks based on predetermined criteria. For example, we determined that a bearing failure on a wagon dumper's rotor leads to an emergency stop and high replacement costs.
• Real-time performance monitoring dashboards. Operators can estimate potential financial losses and the funds required to prevent risks.
• Strategy implementation. The system automatically creates and schedules tasks in a calendar to execute the developed maintenance strategies.
• Root cause analysis. For investigating failures, we implemented advanced visualization of causal connections in the form of "Miro-like" mind maps, allowing users to collaboratively work on identifying the root causes.
• Highly customizable identity access management (IAM). We implemented a customizable, role-based system using Microsoft Azure IAM Tool to ensure high data security.

Migration to SAP S/4HANA:

We chose SAP S/4HANA because the deprecated SAP ERP version (SAP ECC 6.0) was nearing the end of its official support, had functional incompatibilities, and lacked scalability. The migration centralized all maintenance activities into a single information system.

Project Results

The integration of the predictive maintenance functionality into the new platform brought significant, measurable results to the client:

• Reduced repair costs by 40% and increased equipment reliability by 3%.
• The introduction of an entirely new method of proactive maintenance, which allows for continuous monitoring of equipment condition, forecasting failures, and efficiently planning spare parts supply.
• The platform's high performance capabilities allow for data reading in less than a second.
• To date, the solution is used by more than 1,000 users in five of the client's companies. In total, 1.5 million items were added to the equipment catalog, 20 thousand maintenance strategies were developed, and 400 thousand actions were performed.