The Infosys global supply chain management blog enables leaner supply chains through process and IT related interventions. Discuss the latest trends and solutions across the supply chain management landscape.

« ...SAP SRM 7.0 is here to stay, available to Leverage your SRM footprint: Is your On-boarding business case ready for the Roadmap exercise? – Part 1 | Main | Keep it simple and Stick to basics »

Tapping Collective Maintenance Wisdom - An EAM Route?

I recently got a chance to go through an interestingly titled research report from Bill Polk of AMR going by the headline "Asset Management Algebra: EAM = ROI". In these times of increasingly deficient attention-spans, reading a 2-pager is always better than reading a 20-pager with authors belaboring the same point in multiple ways.

Apart from the usual benefits of EAM (ROCE, efficiency improvements, structured information etc) and its new found importance (movement from tactical to strategic), an interesting point which I haven't come across in many other places was about "Capturing and preserving data from an aging workforce". While implementing EAM systems, we typically think of labor management (thru the EAM app or via a little help from more high brow "Workforce Management Systems or WFMs") as a way to capture skills of the maintenance personnel thus making sure the right party is assigned to the right work order.

Here Bill comes up with a different perspective on the importance of EAM & the skills registry part of the implementation. In his words "Many manufacturing assets - heavy assets, in particular - are designed to run for decades. In fact many companies' assets are designed and built by one generation and then operated and maintained by the next. With key information oftentimes found in the heads or desks of an aging workforce, in addition to the increase in maintenance outsourcing, it’s essential to have easily-accessible central repositories of data"

This is clearly an angle we haven’t heard from our clients yet. You may wonder whether we are getting into the realm of Knowledge Management here as against plain old EAM. That may be the case and the challenge would be in capturing unstructured information gleaned over years of maintenance experience which is asset-specific, location-specific and most importantly person-specific into a structured database as part of an EAM implementation, rather than consign it in some kind of free-form text. For industries with longer asset life cycles (say an electric utility or a nuclear power plant), it may be worth the additional effort.

Elsewhere, Bill also outlines very crisply the key benefit metrics (like reduction of emergency repairs or no. of maintenance incidents prevented). Like any other cost-reduction application (say, indirect spend or HR), its actually possible to create a case for self-funding EAM, which just requires a 6-month wait period before the first set of improvements begin to save money for maintenance managers to start planning for more high-end usages like instrumentation integration real-time or asset health dashboards, which work on a near real-time basis.

Where I am a little skeptical is the fast-case implementation scenario Bill refers to, of 30-60 days lead times for vanilla implementations. Perhaps I’m tempered by the fact that Infosys plays almost exclusively in the Fortune-1000 or Global2K client base where all fast cases could end up being lost cases more often than otherwise. My feeling is that such implementations would be largely in the realm of a capability demonstration PoCs. As Bill himself says elsewhere, the real power of the software can only be harnessed when the EAM workflows go beyond the boundaries of one package and start looking at all the cost elements that impacts asset optimization. That covers ERP, plant systems and any niche applications (MRO procurement, inventory management, workforce management) which can contribute to overall goals of an EAM implementation. We can then say that EAM Algebra is truly the sum of parts being larger than the whole.



I agree with Bill on the skills registry part, which is the most precious one in any maintenance organization. Sources for this could be from the Maintenance manuals provided by the asset manufacturer, asset maintenance history that is recorded and available in the system (if there was one already existing), plant engineers' diary, technicians' individual hand books (which is least shared by them - personally I had got more details in this which never gets captured in maintenance history of the asset) and other generic opinion or known fact that is well known among maintenance and / or operations team but never documented properly (Say known fact, gut feeling!!!, seasonal behaviours etc....). If we are able to collect and tag all these during our implementation we should be fairly decent in capturing and preserving the data from the available sources.



I haven’t seen any EAM (or any such system) implementation being successful, if done in 30-60 days. The usage of EAM systems by Blue collar workers makes it more time consuming to realize the benefits. I have seen, at my 1st maximo client in India, way back in 1999, one of the users trying to operate the mouse of the computer as if it was a television remote, which works when you direct the remote in air towards the TV.

My experience says, it takes minimum 3 months of time, after implementation, to start realizing the benefits of any EAM system.

Coming to skills registry, the kind of data EAM systems “eat” is enormous. Feeding Manufacturer’s recommendations, process manuals, maintenance manuals, best practices into EAM system in “right way” is highly complex and could turn into disaster if not done properly. All EAM systems have intelligent & structured way of managing this data and more intelligent & structured way of making that information available & usable whenever required. I have also seen in one of my earlier projects how the older EAM system of client has data in range of millions of records while the same data could have been managed by having just one third of those records.

In a very crude way, if you compare the EAM & ERP systems on volume of master data initially fed into the system, you would find EAM systems need much, too much of master to be fed into system as compared to ERPs and finally it the master data which makes EAM system intelligent and that intelligent, self speaking data is the crux of IBM’s smarter planet theory.

Thanks Praveen for your views (and Senthil's too). Comments like these enrich the blogpost multiple-fold.

The other day I had a chance to talk to someone from a major oil company (nope, its not BP!) and as we were discussing, he was highlighting the fact that the entire RCM philosophy in one of their South-East Asian sites was based on info around 100 tags which had clean data, while upon assessment they found over 100,000 valid tags overall. Such is the state of master data, forget looking at a catch-all execution system for work management. Sometimes, I feel the vision of an instrumented world is still a few light years away when I see instances like the one above. Thanks again.

Post a comment

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)

Please key in the two words you see in the box to validate your identity as an authentic user and reduce spam.

Subscribe to this blog's feed

Follow us on

Blogger Profiles

Infosys on Twitter