Analytics

Failure Mode Analysis is an important method to understand the processes of system failure in detail and identify where and how a system might fail. Emphasizing failure prevention may help minimize the risk of harm to system components as well as the staff working with the system.

Failure modes (More than 150 standard modes are available)

The failure mode process is considered a bottom-up approach; the analysis starts with specific data that builds up to form a more general plan of action. How we do it – each component of the observed system is thoroughly examined for likely breakdown causes. For every identified breakdown scenario, corresponding effects should then be pointed out. This allows the organization to have an extensive map of failure modes and effects, organized according to their level of impact on the business.

What analytics is right for you? Choose the right analytics for the appropriate category of machines

1. Condition based monitoring
2. Preventive maintenance
3. Predictive maintenance
4. Prescriptive maintenance
5. Run-hour based maintenance

1. Condition based Monitoring

Condition based monitoring uses sensors to provide meaningful insights into the current health of various devices or items of equipment in buildings. These sensors collect data to monitor crucial operating parameters such as vibrations, sound anomalies, airflow, and current. The various types of predictive maintenance take condition-based maintenance to the next level. Building on this sensor data, predictive maintenance then uses advanced analytics and artificial intelligence to predict machine failures before they happen or to anticipate maintenance needs before they become urgent.

Condition based monitoring in maintenance is focused on preventing asset failures, downtime, and unnecessary practices by monitoring, asset health to determine what maintenance needs to be completed and when. It can be considered essential to any predictive maintenance strategy. Condition based monitoring dictates that maintenance should only be performed when certain thresholds are reached or indicators show signs of decreasing performance or upcoming failure.

Traditionally, monitoring a machine for these thresholds and indicators included non-invasive measurements, visual inspections, performance data and scheduled tests and is completed at pre-set or predetermined intervals. Condition based monitoring, however, looks at potential failure modes and their indicators, and then monitors for those.

Common types of condition based monitoring include vibration analysis and vibration monitoring, oil analysis, and temperature tracking. Real-time data is gathered through sensors, providing an ongoing method of testing and tracking asset health. This data can be extremely useful not just for the maintenance department, but for operations staff and fleet managers as well.

The goal of condition based maintenance is to spot upcoming equipment failure so maintenance can be proactively scheduled when it is needed – and not before. This is part of what separates predictive maintenance from preventive maintenance.

2. Preventive maintenance

Preventive maintenance is the act of performing regularly scheduled maintenance activities to help prevent unexpected failures in the future. Put simply, it’s about fixing things before they break.

The schedule itself can be based either on usage or time, or a combination of both. For example, the schedule may call for an oil change every 10,000 cycles, or for a component to be replaced every six months. This schedule is rigorously followed, regardless of how much life is left in the lubricants or parts.

What does preventive maintenance entail?

Through machine learning, operational data analytics and predictive asset health monitoring, engineers can optimize maintenance and reduce reliability risks to plant or business operations. Software designed to support preventive maintenance (which is sometimes called preventative maintenance) helps produce stable operations, ensure compliance with warranties and resolve issues impacting production — before they happen.

Benefits of preventive maintenance

Start getting maximum utility from your assets and achieve cost savings by pursuing a preventive maintenance strategy. Added benefits: greater organization and always-on operations.

Extends the asset life – Systematically schedule maintenance and inspections to ensure assets, achieve their full lifecycle and warranties are kept up to date.

Reduces maintenance – Manage planned and unplanned maintenance, inventory and spare parts costs. Better insight into your operations and assets helps you make a significant reduction in maintenance costs.

Boosts productivity – A well organized labor force is a more productive one. It improves scheduling, vendor management and both workflow and financial reporting — all without paper.

Reduces unplanned downtime – Identify repairs earlier in the asset lifecycle for always-on operations that reduce downtime and optimize production.

3. Predictive maintenance

It is a method that employs data analysis tools and techniques to find potential flaws in processes and operational abnormalities so that they can be corrected before they fail. Predictive maintenance ideally enables the frequency of maintenance to be as low as feasible to avoid unforeseen reactive maintenance while avoiding the costs associated with performing excessive amounts of preventative maintenance.

Predictive maintenance foresees issues before they arise by using previous and current data from multiple operations. Predictive maintenance requires consideration of three key areas: 1. Benchmarking inventory use; 2. Real-time monitoring of asset performance and condition; 3. Analysis of work order data.

The importance of predictive maintenance – It gathers a wealth of information that sheds light on crucial questions like:

• What is the likelihood of an asset failing within a given time frame?
• What is the likely root cause of a particular fault?
• Which assets are at the highest risk of failure?
• What necessary maintenance should be performed to resolve the problem effectively?

These data points help businesses strategically plan resource utilization to achieve maximum uptime and productivity across portfolios.

How to put a predictive maintenance program in place:

Determine essential resources or equipment: The strongest prospects to exhibit the highest return on investment from predictive maintenance programs are assets with high repair or replacement prices.

Assemble information from several sources: Take hard copies and digital copies of maintenance records and historical data from your management system for the selected assets.

Determine the failure modes: To identify failure modes for each asset, do a Failure Mode & Effects Analysis.

Implement sensors and technologies for condition monitoring: IoT sensors and condition monitoring techniques should be used to gather information on anticipated failure mechanisms.

The creation of predictive algorithms Using historical asset data and sensor-gathered operating condition information, create prediction models.

Implement a pilot and continuously monitor:  Implement predictive maintenance for your test assets, and then validate your predictive maintenance strategy based on results like improved asset reliability, reduced downtime, and fewer maintenance and equipment outages.

4. Prescriptive maintenance

Prescriptive maintenance leverages the approaches and capabilities of preventive and predictive maintenance to completely optimize system performance. With prescriptive maintenance, devices – in collaboration with operators – are proactive participants in their own maintenance. Several trends are merging to disrupt manufacturing, especially in regard to maintenance. These include the main forces of digitization (Social, Mobile, and Cloud), IoT, and Big Data analytics.

5. Run-hour based maintenance

Run-hour based maintenance is performed on a calendar schedule. A maintenance plan for a piece of equipment is put together that needs to be performed regularly. It is planned maintenance, as it must be scheduled in advance. This means that it can be used with both predictive maintenance and preventative maintenance.