Fred Schenkelberg is a reliability engineer who has worked in the field for over twenty years. He is a West Point graduate. He majored in Physics. He holds a Masters in Statistics from Stanford University. He began his career in reliability engineering at Raychem Corporation. He moved to Hewlett Packard (HP). At HP he helped create the Corporate Reliability Program. In 2004 he founded FMS Reliability. FMS Reliability is a management and reliability engineering consulting firm. He is the founder of the reliability engineering professional development site Accendo Reliability.
- Can you explain what a reliability Engineer is and does?
Not really. The field is broad and the specific tasks a reliability engineer may take on include doing test data analysis, conducting a root causes analysis, recommending enterprise-wide policies concerning warranty reporting or vendor contracting. Of course, there are variations depending on the organization and industry, yet in general reliability engineers work to influence decisions across an organization in order to realize the desired reliability performance of products or systems.
One of the best aspects of reliability engineering is the potential breadth of influence. One may work to improve preventative maintenance practices for a specific asset in the morning, then consult on improving the design of a custom piece of equipment in the afternoon. Or one may work to estimate the sample size for a detail accelerated life test in the morning, then join a meeting of engineering directors to discuss improvements to the internal design for reliability guidelines.
In some ways, reliability engineers enable others to understand and incorporate the appropriate reliability related information into their decision making. The intent is to allow each decision to properly balance all the tradeoffs, along with considering reliability performance ramifications.
- How is reliability engineering related to a company’s Enterprise Risk Management process?
In general, most of the work done by reliability engineers is at a tactical level – specific tasks and activities to move a product or program along. While ERM tends to focus on strategic concerns. Where they overlap, I think, is in the identification and assessment of reliability related risks.
For example, a faulty product that has the unintended potential to harm customers or the environment may cause significant harm to the brand reputation.
Many of the tools used to identify risks to the ability of an organization to create a reliable product, such as FMEA, are also useful when trying to prioritize risks facing an enterprise. While the focus is tactical, understanding when to report into the ERM is essential.
- What are the challenges that a reliability engineer will face in the future and what skills will be needed to meet the challenges?
I see three major challenges facing reliability engineering going forward:
- Deterioration of access to professional development,
- Deterioration of statistical thinking, and
- Acceleration of time to market, cost, and feature set priorities.
Deterioration of access to professional development
Few that work as reliability engineers are actually graduates of a reliability engineering program. Many have been asked to take on the role and found it interesting and rewarding. In the past, professional societies and conferences provided a means to learn the trade, build a supporting network, and learn or improve skills on the job.
Today, even before COVID, conferences have tended to focus on publishing academic focused work and rarely provide practical information to implement with actual problems. Attendance has been on decline for years. Even before COVID live seminars or workshops had a decreasing ability to meet minimum attendance for the event to happen.
Professional society are in retreat and offering less and less by way of useful or educational material. Trade magazines are increasingly catering to their advertisers and reducing the articles and tutorials that have trained generations of prior engineers.
The ability to google most any question or to quickly find an online calculator to solve a sample size problem is great, if one asks the right questions or understands that they are using the appropriate formula.
The gap is the floundering ability of our current system to provide advice, guidance, and support to those just learning which questions to ask or which tools to use.
Going forward, reliability engineers need to take control of their own professional development and be deliberate about it. Rather than a quick google search, an essential skill is to focus on understanding and mastering the necessary knowledge. Dive deep often rather than just doing a cursory overview. In reliability engineering to be good as what you do you can never stop learning.
Deterioration of statistical thinking
A few years ago, when addressing a younger group of engineers, I was shocked to learn that only one undergraduate program of the 30 or so engineers included a probability & statistics course. Given that probability & statistics courses rarely garnered joyful learning, the overall effect of not offering those courses is not large. Yet, not including basic understanding around the language of variation, given that everything varies, is only eroding our engineers’ ability to understand and work in the real world.
Reliability work includes a good amount of statistics. ‘What is the sample size…?” is a the most common question I’ve heard in my career. It’s a statistics question. To design a meaningful engineering experiment, we must understand statistics.
The engineering definition of reliability includes the idea of probability of success. We have to understand probability, too.
The problem is fewer mechanical and electrical engineers have a statistical thinking capability. Thus, when we present a regression analysis of field failure data, one must devote time to teaching what the analysis is and how to interpret the results.
Going forward reliability engineers need to become effective teachers (no, champions) of statistical thinking. We need to enable our peers and managers to use probability and statistical knowledge on a regular basis.
Acceleration of time to market, cost, and feature set priorities
This challenge seems perennial and yet it continues at a quickening pace. Reliability work takes time. The more we rely on simulations and higher acceleration factors, the more risk we take that our work will not be meaningful (not reflective of what will really occur), or that we miss something essential.
I do not see an easy way to address this challenge, other than we need to focus on applying the right set of tools at the right time such that reliability engineering work actually adds value within the given set of constraints. We need an expansive toolbox of skills so we can recognize the need for just the right approach quickly.
We also need to realize the reliability performance is not the only objective. Reliability fits in with time to market, cost, and feature objectives. Being able to work with others, to influence, and to balance priorities to the benefit of customers and the organization is essential.
This goes back to professional development.