Using big data to schedule unplanned maintenance

Using Big Data To Schedule Unplanned Maintenance

It is estimated that from each and every 787 flight, 500GB of data is collected, and everything from the cabin pressure to the pressure of the tyres is recorded. What happens to that massively growing amount of 'big data'? The tangible benefits of processing and applying big data in the Aerospace and Defence (A&D) industry, in particular for the military, is only just being seen. Brendan Viggers, Product & Sales Support for the IFS A&D Centre of Excellence, looks at how the A&D industry is using big data to revolutionise unscheduled maintenance.

Big data, as a concept, has been around for a long time but is now coming of age due to our ability to know what to do with it!

Big data is the term used to describe a massive volume of both structured and unstructured data from an equally varied set of sources so large that it's difficult to process using traditional database and software techniques. Big data is usually described in a five step model, data – filter – context – process – analysis, but more importantly has the potential to help companies improve operations and make faster, more intelligent decisions if processed and applied correctly – allowing A&D organisations to realise a number of genuine benefits.

We have big data but no big picture

With a wider view, analysis of these data sets can find new correlations to spot business trends. These data sets

 

 

continually grow in size in part because they are increasingly being gathered by numerous information-sensing mobile devices.

The problem to date has not come from big data itself, it has come through the interpretation of that data – we simply do not know which patterns are relevant and which aren't, and how to turn them into a realisable benefit. Perhaps the most well-known example of the use of big data is Google! Through reading and indexing huge amounts of information across the Internet, Google is able to compare against your search terms and find what you are looking for. But look past the first page of results and you start to see the real problem because what we see is mostly irrelevant.

This is precisely what we have seen so far in the A&D industry. There is a massive wave of valuable and untapped data collected from land, sea and air assets, but the industry has to decide what best to do with it. So far we have the big data, but where is the big picture?

How this effects A&D – Removing the noise and informing actions

With all this data you have to be able to filter out which events are causing ‘noise’ and therefore makes it difficult to make decisions – to see the wood from the trees.

If we take the 787 for example, what do you do with 500Gb of data which is accumulated from every flight? Your systems and applications have to be smart enough to filter out the information for you and present you with a set of data which you can act upon. It could well be that 499.9Gb of that data is just standard flight information from multiple devices and the 0.1 remaining is actually data that you should apply.  IFS Labs, the research unit within IFS R&D which explores new areas of application functionality and provides a testing ground for new initiatives and ‘proof of concept’ projects not yet proven for large scale deployment, has already made great strides in helping to define which information is relevant and how it can be applied. This goes further than simply indicating trends and adds a sixth and seventh step to the big data approach, data – filter – context – process – analysis – action - benefit.

Monitoring the health of assets

There have already been some limited applications of the value of big data through the use of predictive analytics which focus on better monitoring of usage patterns and the essential tracking and analysing of the health of equipment and troops on the ground in real-time. This is where the 'action' and 'benefit' step of the big data model can provide operational and budgetary improvements. For example Health Usage Monitoring (HUMS) data from a fleet of scout tanks can be applied to form an optimised fleet maintenance schedule – with the benefits of saving time, maximising resources and reducing costs.

But this can go even further. In particular, the greatest potential for big data in A&D is when predictive analytics is used in relation to the ongoing maintenance of equipment. The real improvement that is coming from the sophisticated analysis of big data for A&D is that of predicting unplanned maintenance. Why? Because you can foresee the maintenance requirements of any asset and then at the fleet level reduce the time that assets are out of action, and preparations can be made to ensure that it has little to no impact upon operations.

Predicting Unplanned Maintenance - The future of big data in A&D – Wood from trees

Unscheduled or unplanned maintenance is a real problem in the A&D industry as it means that assets have to be withdrawn from operations often at short notice. If we are to apply this to an extreme, but not unrealistic, scenario in an active combat environment, this could be incredibly serious. If potentially life-protecting equipment is withdrawn at short notice it can halt mission success –  lives are at stake.

This is where the benefits of predicting unplanned maintenance can really make a difference. Picking out the relevant data patterns being fed back out of the myriad of information which is being produced through big data will allow the true benefit of scheduled maintenance to be extended into every aspect of the support chain. More information will be available to understand when parts need servicing or replacing and, along with other technological advancement, potential down-time can be drastically reduced.

Big data already at work – Proven benefits

Previous examples of predictive analytics, using techniques such as Reliability Centred Maintenance, when applied to a real life scenario, have reduced combat systems maintenance in the UK for the MOD typically by half annually with cost savings estimated to be around 20% during the use of the solution.

Extending the current support chain – Joining the dots

Building on this, a completely streamlined support chain which is pre-emptive rather than reactive from the use of big data is the big dream for the A&D industry. Eliminating all unscheduled maintenance may be considered unrealistic - but big data can bring A&D as close as possible to that goal.

We have already seen that planned support is able to significantly reduce asset down-time and costs through careful planning. Big data is a concept and technological possibility that now needs to prove itself. There is an incredible amount of information that has been gathered and is waiting to be put to use. The dots just need joining up in the correct order to understand – but not necessarily to draw – the big picture to turn into real budgetary benefits.

This needs to be done in a measured and timely way as not to disrupt the current support eco-system. Flexible modular software solutions such as IFS Applications make it possible for this to happen. Collecting the data and feeding through an application allows jobs to be correctly prioritised, unlike a lot of inflexible and complex traditional software solutions, the architecture provides the agility for organisations to focus on the applications that are important right now.

The future of big data in A&D

The scope of millions of intelligent devices that can communicate through the Internet, driven by emerging disruptive technologies such as the Internet of Things, opens up new possibilities to move from our traditional reactive type of business model to become much more proactive, where A&D can apply the data they collect to predict and prevent faults before they happen – particularly key for A&D organisations.

The R&D facility at IFS Labs is continually working on new developments to help organisations to manage data in a number of areas. One of its latest research projects is IFS Pulse, a new developmental dashboard for IFS Applications that provides real-time interaction of all the key data streams from social media and RSS feeds to ERP data – providing a granular level of detail on user activity at any given time.

While this is developmental at this stage, IFS Labs is adding the process and action steps which go further than simply indicating trends from existing data. This is giving real insight into where the future lies in terms of real-time applications for this data and unplanned maintenance  - reducing costs and revolutionising the current A&D support chain.

Hide comments

Comments

  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
Publish