A close-up of wrenches and screws

A Guide to Condition Monitoring Approaches to Maximise Return on Investment

The objective of any condition monitoring system is to improve operational efficiency by reducing maintenance costs, minimising downtime (increase availability), improving reliability and optimising performance. Condition Monitoring is not new to the railway industry, but the industry is still heavily reliant on preventative maintenance and schedule-based maintenance strategies. More advanced condition monitoring strategies can be applied to achieve more extensive gains in operational efficiency, but the gains come at a cost in terms of new technology and new resource/process requirements. The overall objective is to reach a point where the savings out way the costs, resulting in a genuine Return on Investment (ROI), as achieved in other similar applications.

To achieve a meaningful ROI from condition monitoring it is essential to explore all options for the maintenance strategy against all assets. Key considerations are usage, criticality, maintainability, redundancy (back-up assets) and value of the asset, alongside maintenance complexity and cost. Different strategies may be appropriate for different assets.

Over the years, various maintenance strategies have emerged and all still have a place somewhere in industry. They can be summarised as follows:

Condition Monitoring Approach 1Reactive: “It’s broken, I need to fix it now”Effectively a ‘run to failure’ strategy, allowing 24/7 operation with no interruptions for maintenance. This can be appropriate where asset reliability is high and failure impact is low – typically when replacement/repair cost is low or where back-up assets are readily available, so downtime may not cause any significant operational impact. The main risk to this approach is consequential impact if the part that fails causes unforeseen damage to other parts of the overall system. The approach can also be blind to the impact of asset performance degradation as the fault develops prior to failure.
Condition Monitoring Approach 2Planned / Schedule based: “If I maintain it regularly, it might last longer”?Planned or Schedule-based maintenance is a proactive approach of preventative maintenance, reliant on regular/periodic maintenance work at intervals based on the usage, criticality and value of the asset. The schedule is typically driven by a time-based schedule (e.g. running hours) or a usage-based schedule (e.g. miles driven). The schedule is usually created based on data from Reliability Centered Maintenance (RCM) policy, supported by a Failure Mode Effects and Criticality Analysis (FMECA). This is a traditional approach which has been used in many industries for several decades and is often used where the safety case is a significant driver, or where downtime results in severe or costly operational impact. The potential downside is that the scheduled maintenance approach can result in parts being replaced or retired long before they are actually worn out since the maintenance is not dependent on knowledge about asset/machine condition.
Condition Monitoring Approach 3Condition based: “I know there is a problem, I’m going to fix it or at least reduce the risk of failure?”Condition-based maintenance (CBM) is focused on detecting and diagnosing faults earlier to allow pre-emptive maintenance, based on evidence that the machine performance is deteriorating and is likely to fail in the future. Essentially the CBM approach results in maintenance when there is a need for it. Various machine parameters are monitored and used to derive condition indicators which are influenced by the health of the machine over time. Unusual behaviour patterns (anomaly detection) can be detected and used to instigate appropriate remedial action, before they result in absolute failure of the asset. In some cases, simple trend analytics can be applied and used to extend time between planned maintenance activities (e.g. extended service intervals), based on the known health of the machine at a given time. Knowledge of the condition can allow a risk-based approach to enable maintenance to be carried out a more convenient time, for example, by aligning repairs or replacements with other scheduled downtime activity.
Condition Monitoring Approach 4Predictive: “How long have I got?”Predictive Maintenance adds failure knowledge, risk assessment and time forecasting to the diagnostic function to estimate Remaining Useful Life (RUL) enabling a prognostic assessment to be made. Correlating machine performance, based on condition monitoring data, with known failure characteristics enables predictions of the time to the next failure to be estimated (using data driven machine learning techniques). The output is a prognostic forecast based on knowledge, risk and probabilities, enabling assets to be safely run for longer and maintenance activity to be optimised, both in terms of schedule and depth.
Condition Monitoring Approach 5Prescriptive: “What should I do?”Prescriptive Maintenance optimizes RUL by integrating the CM with the maintenance planning system to get the corrective action resources in place at the right time. Prescriptive Maintenance does not only identify a problem or predict a failure, but it also provides a focused solution that can be acted upon by generating advisory information. The goal is to have a maintenance system output which is effectively a work order defining resource requirements in terms of labour, tools and parts required to perform the maintenance activity. The work order is scheduled through an optimisation process to align resource availability with minimal downtime impact, whilst also minimising further failure or safety risk..

So… what approach should you take?

There is no simple answer since each customer, each business and each asset may have different requirements and/or be at a different point on the condition monitoring journey. The first step needs to be a comprehensive review of the current state, to determine those areas where the most value can be added in the shortest possible time and at the lowest costs. As with any business change initiative it will then be necessary to clearly map out the objectives, with intended benefits, before making any change to process. Our next blog will explore the journey in more detail.

SmartVision™ has the functionality required to achieve the full potential of the maintenance journey. Its modularity can be used to build system functionality in the stages required, from a simple data monitor through to a fully capable predictive and/or prescriptive monitoring system.

Read more news

Modbus TCP update

Today, EKE-Electronics are announcing the rollout of Modbus TCP (Transmission Control Protocol) client capabilities to our CODESYS® implementation, to further boost the interoperability of our TCMS systems.

Read More
Search