The Challenges of Predictive Maintenance – Unlocking the Potential of PM

Whilst the benefits are clear, the challenges of implementing Predictive Maintenance often go unspoken. This mini blog series will deep dive into these challenges, exploring what Predictive Maintenance is, the hurdles it presents (especially with data), and how to evaluate its true impact.

Traditional Approaches to Asset Maintenance

Asset maintenance typically follows one of two approaches: Reactive maintenance or preventive maintenance.

Reactive Maintenance: This approach is simple in so far as maintenance only takes place once the asset has failed, prompting the maintenance activity to take place. Whilst this is simple, this method is the costliest due to the unplanned downtime, emergency repairs and potential operational disruptions.

Preventive Maintenance: This strategy relies on the regular scheduled maintenance tasks based on the known lifespans of the asset’s components, aiming to keep the assets running smoothly throughout their life. However this can lead to unnecessary repairs and replacements, which again is costly.

So Why Optimise Maintenance?

General asset maintenance costs businesses millions of pounds per year. Add to this that asset downtime can cost businesses significantly through various means such as loss of stock, trading interruptions and reputation leads to a significant avenue to optimise.

With technology at the forefront, the overarching aim of modern maintenance is to ensure reliability to be able to service customers and remove the need for traditional asset maintenance. This requires that the maintenance activities are aligned with the assets through continually monitoring the assets operation via telemetry received through internet of things (IoT) enabled sensors affixed to the assets.  This enables a dynamic link between real time asset performance and potential maintenance activities.

What is Predictive Maintenance?

Predictive Maintenance is the prediction ahead of time for whether an asset or component will fail in the near future, therefore allowing maintenance activities to be scheduled in at the right time to fix issues before they occur and become costly.

Predictive maintenance utilises the continuous monitoring of asset telemetry to allow for early detection of asset failures, reducing costs and risks.

The whole aim is to strike a balance between maintenance costs and failure rates, minimising the overall costs with prevention efforts, as illustrated in the graph.

Figure 1: Visualisation of the overall cost for reactive, preventive, and predictive maintenance regimes.

As can be seen, preventive maintenance aims to minimise failures but is costly due to frequent replacements and prevention costs, whereas reactive maintenance incurs high repair costs due to frequent failures. Predictive Maintenance on the other hand allows the total cost to be minimised by striking a balance between the repair and prevention costs and total number of failures.

What Predictive Maintenance is Not!

Despite the growing appeal of Predictive Maintenance, many of the implementations can actually be categorised as either:

Rule-Based Logic: Where we use predefined logic-based rules to flag anomalies or issues often at the point they are occurring.

Anomaly Detection: Which highlights deviations in the data, that may or may not be indicative of a true issue, without predicting failure timelines.

These approaches, whilst useful, lack the true predictive capabilities, i.e. there is no time element associated with their outputs, and often are picking up issues as they occur or just detect inherent noise within a data set.

Whilst they are not predictive, they do serve a purpose and can be used as a precursor to Predictive Maintenance by allowing for the gathering of failure data to be used with a training dataset for machine learning purposes.

The Transition to Predictive Maintenance

While Predictive Maintenance holds the promise of predicting asset failures before they occur, transitioning from traditional methods demands more than just technology. It requires a shift in mindset.

To optimise maintenance, the predictive solutions must demonstrate cost benefits over reactive and preventive methods. This involves not only adopting AI technologies, but also redefining how maintenance is perceived within businesses.

A significant obstacle lies in changing the mindset around maintenance, often viewed as a necessary disruption to operations. Gaining stakeholder buy in can be challenging when decisions rely on the output from computational insights rather than visible issues, yet embracing this shift is crucial to unlocking the full potential and benefits of Predictive Maintenance.

Where Can Predictive Maintenance Be Applied?

Predictive Maintenance can be used anywhere where telemetry data is being recorded for assets, processes, components and where assets undergo maintenance activities.

To be able to use Predictive Maintenance, the following components are needed as a minimum:

• Time series telemetry data – Data from sensors collected at regular intervals.
• Failure data – Historical failures data records including timestamps for model training.

Looking Ahead

Predictive Maintenance shows a promising advancement to optimising asset maintenance and reducing the overall costs associated with traditional maintenance methods. By Predicting failures ahead of time, businesses can intervene proactively, reducing downtime and maximising asset performance.

That said, the journey is not without further challenges, especially with the handling of and interpreting of data. In the next instalment, we’ll look at the critical data challenges faced when implementing Predictive Maintenance and how these can be overcome.

Andrew is Innovation & Data Director for LoweConex. LoweConex is your Centralised Data Platform, used to help organisations ingest, standardise, unify and enhance unstructured and structured data at scale. Using their technology solutions, customers can achieve energy reductions, drive maintenance cost efficiencies, ensure compliance with regulation and maximise business profitability. Learn more about Switch, our IWMS for Critical Assets, and how it is helping leading brands centralise data essential to enabling predictive maintenance.