Make the business case for predictive maintenance and condition monitoring

JN: Reliability practices affect not just the immediate organization or a local plant, but many other constituents. Examples include investors, customers, suppliers, neighboring plants, communities, and retirees. So the measure of financial impact may be measured better by the cost of not being reliable. For example, if a customer can’t count on you for a quality product or an on-time delivery, the customer is going to look for an alternative supplier, and it’s going to cost you that revenue and maybe your whole livelihood.

If a plant has an environmental accident that affects nearby residents and businesses, the penalties may be quite severe, reducing the bottom-line profits that benefit stakeholders, to say nothing of the ill feeling of those affected. This can be quite widespread. Consider for example the British Petroleum Deepwater Horizon disaster just about four years ago with financial ramifications that still linger today.

If an investor senses an unreliable operation, that investor is going to sell the applicable stock, even at a loss, which may in turn lower value of the overall enterprise as viewed by the market and all who are counting on that value to remain constant and provide income in the form of dividends or growth in capital worth.

Every one of those mentioned and some I’ve not mentioned are affected by a lack of reliability of processes, practices, and the consequences thereof.

PS: Which people within an organization need to be aware of the effects that reliability-centered maintenance practices are having on profitability?

JN: First, management at the highest level should be aware that RCM is important in the 20% of those systems that have the most impact on profitability and that resources to ensure the most effective outputs of an RCM study are provided. By most effective outputs, I mean non-intrusive, condition-directed tasks to mitigate or eliminate failures before they cause unplanned downtime, poor quality, unnecessary waste of energy, or excessive scrap.

Second, the people closest to the systems should have what I call an RCM mentality. This includes the maintenance personnel but equally the operators, who in many cases are starting to apply total-productive-maintenance-inspired practices by undertaking minor maintenance and lubrication tasks and acquiring a sense of ownership and partnership with maintenance in sustaining the highest level of reliability built into the system by its designers or later added by owners.

Operators and predictive condition monitoring team personnel should have a close working relationship that ensures that machinery is made available for monitoring under conditions of loading that provide the most meaningful information and the earliest indication of the onset of problems. That ensures the problems can be handled when the cost to mitigate is low and the risk of collateral damage can be totally avoided. This also means, when an operator senses something trending in wrong direction, a call is made to the PdM team to see if adverse conditions can be further defined and effort begun to correct it. The PdM team must educate the operators on the capabilities they possess and the operator must be imbued with the concept of early detection as the best way to decide upon, start planning for, and executing corrective action, rather than continuing production to the point of complete failure. This is the best way to keep control in the hands of people around it and not surrender to the whims of the machines.

Those responsible for lubrication must take a comprehensive view of all the best practices that ensure highest reliability such as testing new supplies of lubricants to ensure before use that particle count and moisture content meet or are better than specifications for in-service lubricants. RCM studies that are done well will call for tasks that are lubrication-related, especially when failure-modes-and-effects-analysis steps indicate lubrication cause relationships. Even without an RCM study, however, if personnel responsible for lubrication have the RCM mentality I mentioned, they will take steps like testing lubricants on receipt and invoke other measures to avoid creating problems with best practices that are designed to do so.

PS: Because of their different functions, should a maintenance department and a reliability department be separate groups of people in a plant?

JN: I have observed that the most cooperative arrangement is to have the reliability group reporting to the senior-most maintenance manager, especially in organizations where maintenance and operations, and engineering and stores departments, are co-equal with all reporting to the same senior manager. The maintenance department makes the greatest contribution of time and labor to reliability issues and their resolution — RCM analysis, root cause analysis, reliability reviews of new designs for application of PdM, lubrication best practices, or oversight or actual installations of modifications that improve designs for increased reliability. The senior manager in maintenance can best decide where those personnel in the department should spend their time. Obviously, if they’re constantly in crisis mode, finding time for reliability-improvement initiatives is hard, but the sooner the organization can get ahead of the failure curves and allocate personnel to concentrate on improving reliability, the better life in that organization is going to be. Once on that path, most organizations never want to go back to the old fix-when-fail approach, especially if the organization replaces detrimental incentives, such as rewards for fixing things fast, with those that enhance productivity and reliability.

That said, no RCM study, root cause analysis, or other important process affecting reliability should proceed without an operator who understands the system being analyzed. If needed the maintenance manager can intercede with the operations manager to make operator personnel available for participation in reliability-related activities. The operations manager is more likely to cooperate if it’s clear that maintenance is committing personnel to the effort. Should there be a reluctance to cooperate, their superior can arbitrate the matter.

PS: How does a plant manager get operations personnel to work with maintenance and reliability workers toward increased productivity?

JN: First, the plant manager must keep everyone’s eyes on the prize — maximizing designed-in reliability. That in turn means that all work processes must be reliable. Work processes are the high to medium level step-by-step, input-output diagrams that outline how personnel work. Procedures are derived from processes to provide the detail and capture the knowledge of those most closely involved with performing the work.

Plant managers must ensure that everyone understands the processes and provides resources to ensure that procedures are provided that are up-to-date and readily available in forms needed by users — in printed or digital form by Ethernet or wireless communication links. A process for responding promptly to recommendations for improving procedures and rewarding those who make good-faith inputs for doing so, whether adopted or not, must be resourced and continuously supported.

PS: What influence does unplanned downtime have on a plant that is using its excess capacity to produce goods for other companies — sometimes private labels, sometimes competitors, or sometimes related non-competitors?

JN: Unplanned downtime very often translates into missed delivery dates, rushed or overtime-staffed makeup runs that may affect product quality, lower profits for both customer and producer, and lack of strong management in the eyes of investors in public firms or owners of privately held businesses. This in turn drives customers, such as private-label vendors, to look for more reliable suppliers. Competitors may decide their future lies in expanding their own internal production capacity in the belief that they can do a better job of producing quality products on time and in the quantities the market demands. When this happens, everyone associated with the poorly performing plant suffers due to lost profits, jobs, wages, and other benefits, including some or all of their pensions when a plant closes.

PS: With the new capabilities offered by remote monitoring of equipment, how does a plant turn overwhelming amounts of data into useful information that can increase production?

JN: Cloud computing and big data analysis and management are starting to respond to the need for making something useful out of large amounts of data. Companies producing supervisory control and data acquisition (SCADA) software are striving to provide more sophisticated, secure, and easily understood presentations of information from an increasing list of sophisticated sensors. Platform as a service (PaaS) and/or software as a service (SaaS) and enterprise resource planning (ERP) software located in the cloud provider’s facility has many advantages over in-house capabilities including but not limited to the ability to scale up to handle large amounts of new data quickly or add a new analysis capability without any development or capital investment on the part of users. I’m a little leery of having control functions in the cloud for reasons of cybersecurity, but I believe that analysis and tracking functions not linked to control can be tolerated, given that cloud facility operators have protection against hackers and methods of validating data sources as legitimate.

Wireless communications and multi-path networks such those using the Zigbee protocol now have international standards that allow multiple suppliers to provide compliant products that are far less expensive to install, maintain, and use in a plant than those that are wired. Even the need for energy sources is being supplanted by ambient energy-harvesting devices that use things such as normal machinery vibration, temperature differentials, and other methods to generate electrical current for powering sensors and communications devices and obviate the need for batteries that must be changed periodically.

PS: What are the pluses or minuses of leasing a service, such as compressed air or machine uptime, from an OEM, rather than buying equipment and maintaining it yourself?

JN: The latest business models that provide full lifecycle utilities such as compressed air or machinery maintenance seem quite viable to me especially in the developed world where underlying electrical, clean water, sewer, Internet communications, and transportation links are secure, reliable, and rarely interrupted, except in disaster situations such as hurricanes, tornados, 100-year floods, and major snow events where everything else is affected, also. Such arrangements allow both utility supplier and customer companies to concentrate on their core competencies, avoiding diversions from matters not in their regular areas of expertise.

On the negative side, such arrangements must be carefully negotiated with built-in assurances that, when the primary source of supply of a utility or machine availability is interrupted, the suppliers have backup plans for restoration within prescribed timeframes with penalties for missing deadlines. Those involved with putting such contracts in place must understand the needs of their own companies and carefully assess the claims of prospective services suppliers to determine if what’s being promised can be delivered. Financial analysis is important here to determine if outsourcing makes sense.

At the same time, operators of production equipment must be cognizant of their responsibilities for following best practices, not be incentivized improperly — for example, to exceed throughput or agreed-to operating cycle limits in order to enhance their production-related bonuses — and be actively engaged in controlling and constantly looking for ways to involve the processes they are controlling.

PS: How does a plant decide whether to contract services for equipment maintenance and reliability or to hire and train internal staff as a core competency?

JN: This all depends upon the culture of the organization. If it’s a learning organization that believes in and funds meaningful training on a continuing basis at all levels from mill deck to the C-level, then a rational decision can be made concerning whether to outsource to a contractor or train and support internal staff to perform a particular function. The decision often comes down to whether it’s more trouble to accommodate outside contractors or train and support internal staff. Workarounds to accommodate the contractor can only be tolerated so far before frustration sets in. Both financial and human-factor considerations must be taken into account. Questions arise. It’s best to know the answers in advance. For example, is a contractor required to follow internal procedures that have been developed to execute the function being outsourced, or are the procedures to be provided by the contractor? As the contractor continues on the job, is there a requirement for increased productivity and concomitant reduction in cost of the services provided over time or not? Does the contracting agreement require periodic assessment of performance, contain a short time kick-out clause if not satisfactory, and have a long enough term to be effective in achieving improvement and increase in value to the customer?

Read more Big Picture Interviews with industry leaders