Fact or fiction? How do you take decisions on maintenance?

Technological advances relating to how asset managers capture, organise, and use asset related data is, without a doubt, the most important advance that we have made within the twenty years that I have been consulting in this area. The implications of this are immense and it has the ability to permanently change the way that your company manages its physical asset base.

On the other hand new technologies can also bring with them new dangers. Not the least of which is the potential for allocating a lot of time and money on tools, services and activities that do not support the central goals of asset management. In recent time I have seen many corporations spending literally millions of dollars on areas of asset data management that are dubious at best, counterproductive at worst.

The humble CMMS is the centre of most asset maintenance efforts to capture, store and analyse asset data, in larger operations this has been superseded by Enterprise Asset Management, EAM, and Enterprise Resource Planning systems, ERP, but the goal remains the same. These are supplemented by mobile working and bar coding solutions, GIS integration, RFID tagging systems, online condition monitoring, plant management systems, and a range of detailed analytical software tools.

These technologies are now abundantly available; their costs are becoming affordable in terms of generating a good return on investment from them, and they are increasingly easy to use. Yet there are still asset maintenance departments that operate without even a CMMS, or worse, use only a fraction of their existing system.

A common issue is where companies implement a system, starting with asset register information, and then never progress from here to truly effective and dynamic work flow management. They remain stagnant, going in ever decreasing analytical circles focussing on collecting reams of static data, often without a real cost-benefit analysis of how they are going to use it.

So why is this so important? If you ask any of the hundreds of companies specialising in data management and cleansing they will tell you all about how it can highlight areas of inefficiency, aid effective reliability management when applied through a framework of RCM logic, reduce inventory, provide defensible asset replacement plans, and increase profitability. But all of these are actually side issues.

Even today, many maintenance companies take decisions, often very large decisions, based on expert judgement, opinion, and anecdotal information. If we look at this critically we can see that often these decisions are taken based on strength of character, political manoeuvrings, and coalition building. In summary we revert to the situation where “the squeaky wheel gets all the grease”.

When a company begins to take decisions based on data rather than on opinion, the entire dynamic of the company changes. Instead of influence and story-telling, decisions become based on fact; anecdotal benefits are replaced by provable and supportable business plans.

One of the key benefits of data based decisions is that projects stop being initiated based on spurious claims and start being judged based on their ability to achieve identifiable targets. This alone is a strong reason for any company to seriously get into the data-capture and analysis business!

So how does your company get to the point where they are able to take asset maintenance and asset management decisions based principally on data? The following are some tips that may be of use to you, I have used them over the years to help numerous companies to advance their decision support frameworks.

• Don’t focus on the quality, volume and integrity of the data until you know what data should be collected!

It has been my experience that a great deal of money is spent, often unwisely, on programs aimed at perfecting data quality, integrity, and on monitoring volumes of data. If this does need to be done, and often that is a point for debate, then it should only be done after you are sure what data you should be collecting in the first place.

Many companies spend several man years in collecting static, nameplate, data that will be of little or no use. And while doing so they often ignore other areas where data can give an immediate effect and impact.

If you want to define your data sets, then I strongly recommend that you look at what you need it for first. This is a top-down approach that has helped me immensely in the past. By defining the goals and requirements, you are easily able to define the supporting information. (See The Maintenance Scorecard for more information on how to apply this practically)

As a side note, ensuring you have a good understanding of the baseline maintenance regimes through something like RCM will help significantly in this exercise.

• Be careful of trying to lock users into specific way of working. Flexible ways of working, combined with process controls and exception reporting will always provide better results over the long term than locking an interface.

I mention this because I have seen many cases where a lot of time and effort has been dedicated to locking an interface. The goal is to ensure that users only fill out certain fields, to make sure they fill all of the fields out, and to make sure that they don’t enter things that don’t comply with the rules you have imposed.

The intention is great; the reality is less than great. When faced with a screen that requires certain information before allowing them to proceed, users often get frustrated and annoyed and instead of entering what they are required to; they enter garbage or nothing at all!

So instead of improving the asset information this approach often reduces it altogether. A better approach is to train people on what to do and why, ensure that exception reporting is part of somebody’s role so that errors in data entry can be picked up in the short term, and make sure there are feedback loops in place to tell people where mistakes where made.

• Integrate data quality activities into the day-to-day work of existing roles, don’t put it off to one side and assign it to an “asset data” department, or worse, to the IT department.

Asset managers and maintainers are always the best placed to take decisions and exercise judgement over asset information. It can be done through training clerical people, but this will always be limited in application.

As such asset maintainers and managers need to be the ones reviewing asset data exception reports to ensure they meet minimum criteria. So tying this activity into their day-to-day roles is often an easy thing to do. (As long as it doesn’t become yet another data review and survey exercise) Areas where exception reporting is useful include:

• On work order creation, ensuring that all the relevant minimum information has been listed, that the coding represents how the processes were intended to be used, and that any free text meets “good information” standards.
• During planning to ensure that planning codes and information is correct
• During work order completion to ensure that all relevant information is in the correct place
• Review static information on change to make sure that new information follows existing guidelines

None of these are particularly difficult if they are done regularly, otherwise there may well be a need for a very expensive consultancy to come in a “scrub the data”.

Some tools that may be of use:

24 hour reports – A report of all of the corrective and reactive works that have arisen within the last 24 hours to management for their discussion at morning meetings. The outcome will be agreement over some of the work orders, redefining the priorities, and requests for further information. All aids in making sure the data is right from the get-go.

Backlog reports – Regular reviews of the work order backlog to ensure that garbage work orders are identified and either corrected or eliminated, useful for the weekly or bi-weekly planning meetings.

Work order closure reports – Run daily and checked to make sure that the information is correct and that they are going to be of use for future use. A separate but similar report is that of end-of-month cost reports. In similar fashion these allow planners and managers to run through all of the items to ensure they have been attributed to the correct cost centres. When doing any activity based reviews always bear in mind how the data could be used in future.

• Check out mobile and other automatic data capture technologies.

Today there are a lot of technologies available that do not cots an arm and a leg to implement. This is particularly true in the field of mobile technology, barcode scanning, and smart tags. Depending on how they are used all of these items can assist you to capture reams of important data without one key being pressed.

• Do not focus on volume!

One of the great misconceptions about asset data is that rising volumes of data is a good thing. In some areas it is okay while in other areas it indicates second rate performance.
Part of the role of any maintenance department is to effectively manage critical failures. This means that they put in place strategies, plans, and initiatives that are aimed at reducing the amount of critical failures an asset has. So, instead of increasing asset failure data, over time it should be decreasing.

Other areas are different but similar principles apply. For example, work order volumes have nothing at all to do with any measure of maintenance performance. Every company uses work order differently, some at the system level, some at the plant level, some at the asset level and some at the component level. Furthermore, some companies have different work orders for different disciplines even though they are on the same task. Volume is irrelevant, hours worked is vital!

There are many ways to get better data without spending a fortune doing so. Like everything else in asset maintenance the core of it focuses on process development, being smart about the way we use our resources, and having a good understanding of why we are doing certain activities in the first place.

The costs of doing this should be small, and the gains include taking a fact based approach to asset maintenance in your company. Definitely a good return on investment!

I hope this information is of some use to you and good luck with your data improvement initiatives!