Exploring Reliability Culture
Years ago I had the opportunity to assess the reliability programs of two organizations. They made similar products for different segments of the market, and they both had about the same size an organization. Two years previously, both organizations lost the reliability professional from their staffs. Furthermore, both teams were located in one building, one upstairs and the other downstairs, which made scheduling the assessment interviews convenient.
Upstairs / Downstairs
Though the course of the interviews I enjoyed the conversations more with the organization upstairs. They started on time and were not interrupted. One of the first things I noticed was that the office plants were common, green and healthy. The engineers and managers knew how to use a wide range of reliability tools to accomplish their tasks. For example, the electrical design engineer knew about derating and accelerated life testing, and she also knew about the goal and how it was apportioned to her elements of the product. Each person I talked to upstairs knew the overall objective and how they provided and received information using a range of reliability tools to make decisions. They enjoyed a very low field failure rate and simply went about the business of creating products.
Downstairs was different. The interviews rarely started on time, and most were interrupted by an urgent request usually involving an emerging major field issue or customer complaint. I didn’t see any office plants, just plenty of coffee pots. The engineers and managers knew that ‘Phil,’ the former reliability engineer with the team, did most of the reliability tasks. “That was Phil’s job” or “Phil used to do something like that.” when I asked about stress testing or risk assessment. Most did not know what HALT or ALT was and didn’t have time to find out. There was a vague goal, and all agreed that it wasn’t measured during product development, and so was meaningless. The downstairs team had a very high field failure rate and the design team often spent 50% or more of their time addressing customer complaints.
The only salient difference between the teams and their history was the behavior of the former reliability professionals with each team. Upstairs, Mabel was a reliability professional well versed with a wide range of reliability tools and processes. She provided direct support along with coaching and mentoring across the organization. She encouraged every member of the team to learn and use the appropriate tools to make decisions. The team became empowered to make decisions that led to products meeting their reliability goals.
Downstairs, Phil was another reliability professional well versed with a wide range of reliability tools and processes. He directly supported the team by doing the derating calculations, asking vendors for reliability estimates, designing and conducting HALT or ALT as needed, and the myriad of other tasks related to creating a reliable product. He provided input and recommendations for design changes that would improve reliability, and he was a key member of the team. Phil was not a coach or mentor, however, and as he moved to a new role, his knowledge and skills went with him. He preferred just to do it himself and often found he had little time to teach others about reliability engineering tasks.
The difference between the organizations was in the culture. The difference showed in who had and who used reliability engineering knowledge. When the entire team has knowledge appropriate for their role on the team, they can apply those tools to assist making design decisions. Without that knowledge, design teams will use the tools and knowledge they have to make design decisions. Without the consideration of reliability-related information, the design decisions are made blind to the impact.
Reliability occurs at decision points during the design process. When components are selected, when structures are finalized, or when all risks have been addressed. Near the end of any product development process, the team asks if the product is ‘good enough’ to start production and introduce the product to the market. Having a clear goal with an appropriate measure of the current design’s ability to meet that goal provides the reliability aspect of ‘good enough.’
Every organization and product are different. The markets, expectations, and environments are all different. Every product achieves some level of product reliability. The culture is only one factor, yet I suspect you would agree that working upstairs would be preferable.
Do you work upstairs? How is the culture concerning reliability?
Transition of Reliability Culture (podcast)
To change the Reliability Culture (article)
Reliability Maturity (ebook)
Also published on Medium.