Thinking of implementing predictive asset maintenance? Expect bumps on the reliability journey

Back of jeep facing mountain-scape

Increasingly, organisations are looking at methods such as Predictive Maintenance (PdM) and Reliability-Centred Maintenance (RCM) to increase cost effectiveness, asset reliability and uptime. This is because traditional maintenance methods such as break-fix, and other reactive approaches are not sustainable for an asset-intensive organisation.

Although maintenance, repair and overhaul (MRO) activities such as PdM and RCM are viewed by most organisations as a cost, MRO spending should be considered an investment for asset reliability. For example, a properly implemented MRO system could save an airline company $4 million to $18 million by minimising delays to flights due to unscheduled maintenance.

So, when should PdM be applied to critical assets in order to allow organisations to achieve optimum MRO?  Let’s review the considerations.

Reliability-Centre Maintenance (RCM)

The RCM process, when applied to systems, ensures that assets continue to operate per user requirements while receiving a safe minimum level of maintenance. This allows organisations to develop more effective maintenance strategies by identifying the assets’ failure modes and the process which the assets degrade before failing.

RCM also applies risk management principles. By assessing the likelihood of failure, both the consequences and impact (i.e. safety, environmental, operational and economic) of when the asset fails can be predicted.

An effective RCM process will address a number of questions, including asset functionality, the number of ways an asset can fail, and the causes for those failure.

Failure Mode and Effects Analysis (FMEA) and Failure Mode, Effects and Criticality Analysis (FMECA) processes will address most of these questions. These models examine how the asset functions, possible failure points, the degradation process that an asset undergoes before it fails, and how failure affects the operation of the system.

However, I will specifically address, “What should be done to predict or prevent the failure from occurring?” How can predictive maintenance assist organisations to make better decisions to balance risk, cost and performance?

First, we need to understand the difference between preventive and predictive maintenance.

Preventative vs. Predictive Maintenance

Both techniques share a common goal; they are designed to minimise the likelihood for the asset to unexpectedly fail. However, there are differences in each approach.

Preventive maintenance refers to repairs or asset replacement activities that are performed based on a scheduled interval, or periodic cycle typically based on the manufacturer’s recommendations.

Predictive maintenance does not apply a cycle or schedule for repair works. Instead, it captures evidence and analyses the condition of the asset. The information is conveyed to an operator who will make an informed decision as to when to perform maintenance.

To illustrate the difference – think about your car service schedule. Preventive maintenance dictates that, according to the manufacturer’s advice, a car is serviced at set intervals, usually time elapsed or kilometres driven. On the other hand, predictive maintenance occurs when there is evidence of an impending failure, like a diagnostic routine that sets a light flashing on the dashboard that the engine is failing.

When compared to preventative maintenance, PdM can deliver significant cost savings. The asset is only repaired when there is fault and is only taken out of production when the need for maintenance is identified.

PdM in Operation

PdM, as demonstrated by the car service example, provides information to justify cause and assists owners to decide if MRO efforts are required.

Now, let’s examine PdM in operation for an asset-intensive industry such as a natural gas plant:

PdM flow chart

Assuming the RCM process has been applied effectively to the plant operating system and assets, this is how the planning and execution of MRO activities should occur:

  1. The asset layer, continually reports on its condition to the process control system (for this example a pressure valve is reporting differential in flow).
  2. The control system reports the asset’s condition to the predictive analytic mechanism which examines the data.
  3. Reports are fed into the organisation’s enterprise asset management (EAM) system, which is used by the organisation to plan, optimise, execute, and track maintenance activities.
  4. A planner then reviews the reported information from the PdM and uses the decision support tools within EAM to decide if the recommended MRO is required and executes the work order.
  5. Once the EAM dispatches the work order, it identifies and allocates the most qualified person in the maintenance workforce and prepares the necessary permits and schedules outages. A dispatch order is also issued to the inventory warehouse for the required tools and parts for the task.
  6. Once the work crew completes the task they close out the work order in EAM and the asset’s condition is reset to the original working status.

The example demonstrates additional benefits gained by incorporating PdM into EAM, including:

  • Better storage management for replacement assets and spare inventory
  • Better coordination for maintenance planning and workflow process management in terms of workforce mobilisation, permit to work, dispatch of tools and plants for repairs
  • Decision makers are better informed
  • Repairs and maintenance activities are justified and evidence based

The journey for an organisation to implement RCM and PdM can be challenging. Consider:

RCM is process-oriented: an organisation’s production system and assets need to be identified and understood, including both tangible and non-tangible components that contribute to a working system.

RCM should not be a static implementation: A continual review of the RCM process to understand its effectiveness is crucial. It’s also important to review RCM and PdM when there is a change to the asset’s function, changes to the environment, regulation and/or legislative changes.

Return of Investment (ROI) can be hard to measure:  In practice, ROI can be notoriously complex, but the non-financial benefits, such as improved safety and reputation, may be enough for an organisation to implement RCM and PdM.

Excitement of technology enablers overshadow the original problem: Predictive analytics, machine learning and Internet of Things (IoT) are all technologies that enable the PdM process. However, they should not become the primary drivers.  Technology enablers are only valid if they help achieve business outcomes; in this case, provide organisations with information and tools for better decision making.

Lack of endorsement to implement RCM and PdM: Multiple divisions are generally responsible for the maintenance and the operations of asset-driven organisations. The RCM process requires inclusive buy-in from the departmental heads to avoid a poorly implemented system and minimise user rejection.

Quality data hasn’t been captured:  The right data is needed to better support the decision making process. Organisations need to assess if the data collected is contributing to the prevention efforts, and if the data is improving the overall decision making process.

PdM should not be considered for all assets: Not all assets will realise a benefit from PdM. It is only a good investment if the prevention costs is less than the repair costs.  A break-fix reactive approach may be more suitable for small-cost components were failure has little/no impact on the primary function of the asset.

Final Thought

MRO efforts, when poorly managed, increase costs and erode profit margins. Organisations cannot simply ignore all maintenance efforts. When assets fail, they contribute to lost production time and create safety hazards that could result in loss of life, injuries and illness and/or breaches against regulatory and legislative requirements.

The implementation of RCM and PdM can help organisations obtain a balance between maintenance and asset reliability. For asset-driven organisations, the benefits can be immeasurable.  Besides an increase in cost effectiveness, asset reliability and uptime, organisations can develop a better understanding of the risks of asset failure and the implications.

 


David LimDavid Lim is a Senior ICT Consultant with more than 11 years’ experience across multiple sectors including transport, resources and telecommunications. David has successfully designed and delivered a wide range of critical communication infrastructures that are capable of supporting mission-critical operation technology and enterprise-level user applications. David has developed a keen interest and expertise in strategic planning and digital transformation of the networks to meet emerging business needs. He has provided advisory guidance and consultancy on asset investment planning to balance between risk, cost and performance.

Comments

  1. Brilliant article. Great to see. Thanks David.

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: