• Written by Sonali Subhadarsini, Product Manager

In theory, AI for production in metals and mining should be able to provide immediate answers to critical questions about output, maintenance, and quality, but in practice, many production teams are still waiting on reports to get the information they need, and even then, the data may be incomplete or outdated. The shift supervisor, for example, may need to know the current production levels against plan, but this information is often only available after a delay, by which time the opportunity to make adjustments has been missed. The category manager, on the other hand, may be trying to compare the performance of different suppliers, but the data on quality and reliability is scattered across different systems and not easily accessible.

The production questions that currently wait on a report

Production teams in metals and mining have a wide range of questions that they need to answer on a daily basis, from what is the current production level, to what is the status of maintenance on critical equipment, to what is the quality of the output. However, many of these questions are currently dependent on reports that are generated on a periodic basis, such as daily, weekly, or monthly. This means that the information is often outdated by the time it is available, and the team has to make decisions based on incomplete or inaccurate data. The maintenance manager, for instance, may need to know the history of maintenance on a particular piece of equipment, but this information is only available in a report that is generated at the end of each month.

A typical example is the question of what is the current level of scrap, and how does it compare to the same period last year. This information is critical for the production team to understand the effectiveness of their processes and to identify areas for improvement. However, the data on scrap is often only available in a report that is generated on a weekly basis, which means that the team has to wait until the end of the week to get the information they need.

The data behind them: output against plan, cycle times, scrap and quality records, maintenance history

The data that is required to answer these questions is typically scattered across different systems, including production planning, quality control, and maintenance management. The production planning system, for example, will have information on the production schedule and the output against plan, while the quality control system will have information on the quality of the output and the level of scrap. The maintenance management system, on the other hand, will have information on the history of maintenance on each piece of equipment, including the date and time of each maintenance activity, and the parts and materials that were used.

By bringing all of this data together, the production team can get a complete picture of their operations and make informed decisions about how to improve their processes. The category manager, for instance, can use the data on quality and reliability to compare the performance of different suppliers and make informed decisions about which suppliers to use. The shift supervisor, on the other hand, can use the data on output against plan to identify areas where the production process can be improved, and to make adjustments to the production schedule as needed.

Sector constraints: grade variability and maintenance backlog

One of the key challenges facing production teams in metals and mining is the variability of the grade of the ore or other materials that they are processing. This can have a significant impact on the production process, as different grades of material may require different processing conditions, such as temperature, pressure, and flow rate. The production team needs to be able to adjust their processes quickly in response to changes in the grade of the material, which can be a challenge, especially if the data on the grade of the material is not readily available.

Another challenge facing production teams is the maintenance backlog, which can have a significant impact on the availability of equipment and the overall efficiency of the production process. The maintenance manager needs to be able to prioritize maintenance activities based on the criticality of the equipment and the impact on the production process, which can be a challenge, especially if the data on the maintenance history of each piece of equipment is not readily available. By using AI for production, the maintenance manager can get the information they need to prioritize maintenance activities and minimize the impact on the production process.

What improvement looks like: OEE, unplanned downtime hours, scrap rate

Improvement in the production process can be measured in a variety of ways, including overall equipment effectiveness (OEE), unplanned downtime hours, and scrap rate. OEE is a measure of the efficiency of the production process, and it takes into account the availability of equipment, the performance of the equipment, and the quality of the output. Unplanned downtime hours, on the other hand, are a measure of the amount of time that equipment is unavailable due to unexpected maintenance or other issues. The scrap rate, which is the amount of material that is rejected due to quality issues, is also an important measure of the effectiveness of the production process.

By using AI for production, the production team can get the information they need to identify areas for improvement and to track the effectiveness of their processes over time. The shift supervisor, for example, can use the data on OEE to identify areas where the production process can be improved, and to make adjustments to the production schedule as needed. The maintenance manager, on the other hand, can use the data on unplanned downtime hours to prioritize maintenance activities and minimize the impact on the production process.

What this does not do, or when it is the wrong choice

While AI for production can be a powerful tool for improving the efficiency and effectiveness of the production process, it is not a solution for every problem. In some cases, the data that is required to answer critical questions may not be available, or it may be incomplete or inaccurate. In these cases, AI for production may not be the best solution, and other approaches, such as manual data collection or analysis, may be more effective. Additionally, AI for production is not a replacement for human judgment and expertise, and it should be used in conjunction with the knowledge and experience of the production team.

If you are considering using AI for production to improve the efficiency and effectiveness of your production process, we recommend that you contact us to discuss your specific needs and requirements. We can help you to determine whether AI for production is the right solution for your organization, and to implement it in a way that meets your needs and achieves your goals.