Reliability: It Ain't All About Maintenance!

Welcome Reliable Plant readers to my new column, which I’ve entitled “The Exponent.” I’m very excited to embark upon this new venture! I chose to name my column “The Exponent” for a very strategic reason. Let’s look at how the dictionary defines the term “exponent”…

1. a symbol written above and to the right of a mathematical expression to indicate the operation of raising to a power
2. a. one that expounds or interprets; b. one that champions, practices or exemplifies

For a plant reliability management professional, the term possesses an interesting juxtaposition of definitions. It at the same time represents a champion or practitioner and the mathematical function that is the foundation of the exponential distribution on which the majority of basic reliability calculations are based. The exponential distribution is used to calculate system or component reliability by raising the natural logarithm – mathematically represented by the term e = 2.718 – to the negative power of the failure rate multiplied by time.

Most basic reliability calculations are based upon the natural logarithm, or some derivative, including Weibull analysis, arguably the most powerful quantitative tool in the reliability engineer’s arsenal. One could say that e is to reliability what pi is to geometry. Both are versatile yet irrational numbers.

My mission for this column is to discuss the foundational elements of plant reliability management and to champion the practice in the field to the benefit of our readers, their employers and other stakeholders. I’ll address both quantitative and qualitative concepts. Some columns will improve your toolbox, others will make you think, some will celebrate success, and still others will serve to help us all learn from our failures in hopes of avoiding recurrence.

Reliability is a functional enabler that transforms root causes into desired outcomes.

The first issue I’d like to address is the industry culture we’ve collectively managed to create whereby our organizations have come to define reliability simply as a de facto synonym for maintenance. It’s far too common for management to respond to questions about reliability by citing who’s in charge of it. “Yeah, we’ve got reliability – Joe takes care of that.” This is hauntingly similar to what occurred during the quality revolution. Industry recognized that quality is an objective that serves the mission of the organization, not a functional department. Only when organizations began to create quality cultures, assigning responsibility to everyone, did organizations break out of the inspection-based quality control mind-set into the mission-focused quality management mind-set. Reliability is no different. If you think your organization has reliability because a couple of folks are working on it, you’re severely mistaken. 

Let’s break this down into a flow model and run a sanity check (see figure below). Reliability is a functional enabler influenced by several root causes which, in turn, affect several symptoms that are mission-serving desired outcomes. Reliability, along with several other functional enablers, permits the plant or factory to achieve its uptime, yield, quality, safety and environmental impact objectives, and ultimately the firm’s mission objectives. However, this nebulous term “reliability” is a byproduct of its root causes, or affecting variables, which include design, procurement, operations and maintenance. Assuming that any one of the influencing variables alone can produce the desired reliability outcome is nonsensical – what I often call pretzel logic because it’s all twisted up.

Take design and procurement, for example. Just like genetic predisposition is the largest factor influencing the likelihood of heart attack, stroke and other failure-inducing ailments in a human being, mechanical and electrical design, which can be extended to include procurement, determines the baseline reliability of a machine. Bad design is like bad DNA in a human being. Mike Johnson, one of Noria’s senior technical consultants, refers to bad design as bad DNA – Design Not Adequate! No matter how you maintain and operate the machine, if it has bad DNA, you’ll never achieve your operational reliability goals and, hence, your operational business objectives of uptime, yield, quality, safety and environmental impact. If these operational goals are met, there is little chance of achieving your mission goals of increased return on net assets (RONA), increased earnings per share, increased share price, etc.

Ray Oliverson of the Hartford Steam Boiler insurance company, attributes about 40 percent of plant problems to design, about 30 percent to operations and about 30 percent to maintenance. I think his numbers make sense. And, there are many interactions within those functional groups. Design influences maintainability and operability, for instance. Maintenance and operations, if they communicated with one another, affect each other. If they communicate effectively with design and procurement, they can help avoid recurrence of design or purchasing mistakes.

The point is, reliability is not a functional task that’s assignable to a person or group. We’re all in this together. We can make the most of our opportunity to contribute to the organization’s mission through teamwork.

It’s time to start changing our approach to plant reliability, which begins with a change in attitude. I look forward to future installments of “The Exponent.” I’m always interested in your feedback and ideas!