Quality professionals can leverage the same tools and methods used by reliability engineers
by Matthew Barsalou
There’s a light bulb that’s been in use at a fire station in Livermore, CA, since 1901. The light bulb has never been turned off, although it has gone out due to power outages and a week of renovation, as well as a brief trip to a new fire station in 1976.1
That means the light bulb has been lit for more than 1 million hours. For comparison, new energy-saving compact florescent lamps can last more than 15,000 hours, and a normal light bulb typically lasts only 3,200 hours. Modern LED light bulbs are touted to have a lifetime of up to 46 years,2 but the California fire station’s “Centennial Light Bulb” from more than a century ago has a proven lifetime of more than twice that. It would be more than fair to describe the Centennial Light Bulb as “reliable.”
To ensure the reliability of products, there is an entire subfield of quality called reliability engineering. Reliability engineering originated in the United States after World War II. There was a shortage of consumer goods after the war, and U.S. organizations could seemingly ship any product, regardless of the level of quality. Eventually, the consumer goods shortage ended, and customers began complaining about the poor quality of products.
One of the responses to this situation was the creation of reliability engineering as a field, which concentrated on how well products performed after they had been shipped to the customer. This contrasts with the quality assurance of that era, which concentrated on the quality of products prior to shipping to the customer.3 The idea of quality assurance ignoring product performance after it is in the customer’s hands may be shocking in today’s world in which quality management is concerned with the performance of the product in the customer’s hands and the customer’s satisfaction.
According to authors Patrick D.T. O’Connor and Andre Kleyner, the objectives of reliability engineering are using engineering concepts to reduce field failures, identifying and correcting failures that do occur, finding ways to deal with failures when the root cause has not been found, and applying methods for the estimation of potential reliability failures of new designs and analyzing reliability data.4 The concept of reliability engineering is especially relevant today when we are guided by the principles of quality management and customer satisfaction.
Hazard rate and bathtub curve
One useful concept of reliability engineering is the bathtub curve. The bathtub curve is not intended to determine the operating life of an individual product or system. It can, however, be used to understand the life of many products or systems.
The y axis depicts the failure rate and the x axis depicts time. The “instantaneous failure rate of a device as a function of time” is the hazard rate.5 There are three general hazard rates, and they are depicted in the bathtub curve in Figure 1:
- Early failures, which are the result of quality problems such as assembly mistakes or defective components. These failures occur either before the product is shipped to the customer or not long after the customer starts using the product.
- Random failures, which are the rarest type and include root causes (such as electrical overload in light bulbs).
- Wear-out failures, which happen when the product or system begins to deteriorate because of age (for example, a light bulb’s filament generally wears out after thousands of hours of use) and the end of its useful life.
The Weibull distribution
The bathtub curve is a useful model for understanding the failure phases of a product. But a Weibull analysis also can be used to make predictions regarding batches of parts. The Weibull distribution is a frequently used probability distribution in reliability engineering, and the shape of the distribution is determined by the hazard function. A shape of less than one indicates early failures, a shape of one indicates a constant failure rate, and a shape greater than one is a sign of increasing failures over time.
Figure 2 shows Weibull distributions with shapes equal to 0.5, one, two and three. Notice the lower values have more failures earlier than the higher values.
MTTF, MTBF, MTTR and availability
Another concept is mean time to failure (MTTF), which is used to find the average time it would take for a nonrepairable product or process to experience a failure. Mean time between failures (MTBF) is used when the product or system could be repaired, and is used to determine the average time between repairs when the failure rate is consistent.6 Light bulbs are an example of an item that would use MTTF because they are generally not repairable.
On the other hand, a manufacturing system used for producing light bulbs occasionally may break down and be repaired to put it back into service. Here, MTBF would be used.
The time it takes to get a repairable system back into operation is the mean time to repair (MTTR). The availability of a system can be calculated when MTBF and MTTR are known. Availability is calculated as:
Quality and reliability
The tools and methods of a reliability engineer are not necessarily limited to reliability engineers. One typical reliability engineering method is now ubiquitous in the field of quality: failure mode and effects analysis. Additional useful reliability engineering tools include reliability block diagrams, fault tree analysis, and tolerance and worst-case analysis.
Quality professionals can benefit from applying the same methods to ensure that products function as intended for the product’s expected lifetime. In other words, the combination of quality and reliability methods can help ensure we produce reliable products.
- Centennialbulb.org, http://centennialbulb.org.
- Consumer Reports editors, “Choosing the Best Energy-Saving Light Bulbs,” Consumer Reports, August 2013, http://tinyurl.com/consumer-reports-light-bulbs.
- Joseph M. Juran, “A History of Managing for Quality in the United States of America,” which appeared in A History of Managing for Quality: The Evolution, Trends and Future of Managing for Quality, Joseph M. Juran, ed., ASQC Quality Press, 1995.
- Patrick D.T. O’Connor and Andre Kleyner, Practical Reliability Engineering, fifth edition, John Wiley & Sons Ltd, 2012.
- Forrest W. Breyfogle III, Implementing Six Sigma: Smarter Solutions Using Statistical Methods, second edition, John Wiley and Sons, 2003.
- O’Connor and Kleyner, Practical Reliability Engineering, see reference 4.
Matthew Barsalou is a certified Six Sigma Master Black Belt located in Germany. He has a master’s degree in business administration and engineering from Wilhelm Büchner Hochschule in Darmstadt, Germany, and a master of liberal studies from Fort Hays State University in Hays, KS. Barsalou is an ASQ senior member and holds several certifications.