Maximizing Efficiency: Best Practices for Configuring a Successful Downtime System in the Mining Industry
Written by Matt Lindsay, P.Eng
When it comes to mining and mineral processing operations, unscheduled downtime and delays can lead to significant production chain losses. Not only does it result in lost capacity and production opportunity it also drives increased maintenance cost. A survey found that 81 per cent of organizations believe that digital tools play a significant role in reducing unplanned downtime. Implementing a downtime system is the first step to gaining a better understanding of your operations and figuring out the root cause of lost capacity. This article examines the best practices for configuring a downtime system.
Best practices for Configuring a Successful Downtime System
Once the business case has been approved the next step is to implement the downtime tracking system. Implementing a downtime system is essential for gaining insight into operations and identifying the root causes of lost capacity in the mining and mineral processing plants. Now, let’s look at some best practices for configuring a downtime system to achieve quicker return on investment, greater adoption, and assist the analytical process in the mining industry.
Defining your Machine Centers
When defining your machine centers, it is important to keep in mind that a machine center is a single point of measurement for downtime and generating key performance indicators (KPIs). You will want to identify internal storage or buffers in your process which allow assets to run independently of each other. Assets which operate together, such as conveyor belts in series, would be better defined as a conveyor belt system instead of monitoring each belt individually for downtime. We can leverage the reason code tree to track which assets are causing downtime within a system or process.
Building a Useful Reason Code Tree
A well-thought-out reason code tree is the backbone of any useful downtime system. The reason code tree sets the foundation for downtime analysis and uncovering the root causes of capacity loss. It is important to drill down through areas, assets, and components to arrive at failure modes and to avoid creating a reason tree that is too basic or too detailed. A flat reason code tree is easy for operators to navigate but lacks the detail required to understand the root-cause of downtime and ideally should be 4-6 levels deep. Ensure the reason code tree is easy to navigate, captures the key equipment, components and failure modes in the system and provides the necessary granularity for root-cause analysis.
Deriving KPIs from your Time Usage Model
A standard time usage model is critical for accurately monitoring asset performance and comparing plants and assets throughout an enterprise in the mining industry. The time usage model will determine the formulas for your asset KPIs. These KPIs can be useful to determine the major types of production losses that are occurring. It is also important to obtain agreement across departments on the classifications and definitions so that the resulting KPIs are meaningful.
Triggers & Auto-classification
Accurate triggers and auto-classification are crucial for capturing downtime and slowdown events. That means aligning triggers with the actual productive run time of your equipment and using data from other systems to classify events and lighten the load on your operators. And don’t worry, there are tools like minimum-event duration, timers, and moving averages to help you figure out the perfect event capture level. Some smaller or lower valued assets may generate many events throughout a shift and would be too time consuming to capture all the downtime reasons. Leveraging auto-coding based on PLC faults, interlocks and programmable logic can generate a dataset for Reliability Engineers without putting too much workload on operations staff.
Custom Attributes for Greater Context
Custom attributes can provide invaluable context for downtime events by tracking information like shift, season, operator, and process conditions. Integrating data from other systems like CMMS or ERP can also give you a better understanding of the bigger picture and help with cost accounting. And let’s not forget about equipment data, such as real-time telemetry data, sensor data, and maintenance logs can provide even more detailed information about the operating conditions that led to the event.
In the mining industry it can be valuable to collect the raw material quality with the downtime events. Is there a correlation between size distribution and plugging events? Does hard grind ore lead to more wear on equipment components? Do certain pits/ore faces lead to more slowdown events? These are the types of questions that can be answered once custom attributes are added to downtime events.
Tie it All Together with Actionable Reports
Finally, it’s all about turning data into actionable reporting. Utilize visualization tools like Pareto graphs, translate time into lost production capacity or lost opportunity, and assign ownership to key performance indicators like availability, utilization, and operating efficiency. Review these KPIs on a time series graph to see how you are performing over time and drill down to the underlying events which make up these KPIs.
In conclusion, implementing a downtime system is vital for gaining insight into operations and identifying the root causes of lost capacity in the mining industry. By following these best practices for configuring a downtime system, engineers and technicians can achieve quicker return on investment, greater adoption, and assist the analytical process. It is important to remember that unplanned downtime is estimated to cost 10 times the cost of scheduled downtime, so taking the time to properly configure a downtime system can have a significant impact on a company’s bottom line. So don’t be afraid to embrace technology, it can be your best friend in reducing downtime.