Reliability bathtub curve

The lifetime of a population of units at the component, board, box, or system level can be divided into three distinct periods. This is most often defined by the so-called reliability bathtub curve (Fig. 6.16). The bathtub curve describes the cradle-to-grave failure rates or frequency of failures as a function of time. The curve is divided into three distinct areas early failure rate (also known as infant mortality), the useful life period, and the wearout failure period. The infant-mortality portion of the curve, also known as the early life period, is the initial steep slope from the start to... 

The reliability bathtub curve, shown in Rgure 2.5, represents the change in probability of failure over time of a component. The bathtub can be divided into three regions bum-in period, useful life period, and wearout period. 

The LINUX cluster had been managed under continually increasing load for a period of months. Systems administrators had responded by incremental addition of PCs, but the performance had shown continued decline. The system was loaded near capacity, which is often far below theoretical, 70% of maximum being a common benchmark. Another feature of the system, which is especially relevant to clusters, was the impact of component failure. For computers, this is likely the disk drive. Drives have a mean time between failure of one to two years and follow a bathtub curve of burn-in failure, followed by reliable performance and then high failure near the end of the lifecycle. Cluster design must take into account the need for graceful degradation. 

Fig. 5.9.6 Schematic bathtub curve of failure rates over time with increasing reliability, the curves are lower and less steep at the end of a product s life...

The shape of the plot in Figure 3-2 is characteristic of many components and well known to reliability engineers. The shape is called the "bathtub curve." Three regions are distinct. In the early portion of the plot, failure rates are higher. This area is called "infant mortality." The middle portion of the curve is known as "useful Ufe." The final portion of the curve is called "end of tife" or "wearout region."... 

Statistical values can also be used to determine expected periods of optimum performance in the life cycle of products, systems, hardware, or equipment. For example, if the life cycle of humans were plotted on a curve, the period of their lives that may be considered most useful, in terms of productivity and success, could be represented as shown in Figure 5.4. This plotted curve is often referred to as the bathtub curve because of its obvious shape. A similar curve can be used to determine the most productive period of a product s life cycle according to the five known phases of that life cycle, as discussed in Chapter 3. The resultant curve, known as a product s reliability curvey would resemble the curve that appears in Figure 5.5. During the breakin period, failures in the system may occur more frequently, but decreasingly less frequently as the curve begins to level toward the useful life period. Then, as the system reaches the end of its useful life and approaches wearout, more frequent failure experience is likely until disposal. 

Hence, statistical evaluation of failures that occur during a product s life cycle help develop a failure curve that, because of its shape, is referred to as a bathtub curve. When considering the usefulness of a product, the curve becomes a reliability curve for that product. 

D.J. Wilkins, The Bathtub Curve and Product Failure Behavior Part One - the Bathtub Curve, Infant Mortality and Burn-in, ReliaSoff Reliability Field Consultant, November 2002. Reliability hotwire e magazine issue 21. 

Bathtub Curve A graphical representation of the life cycle of products, systems, or individual components in terms of frequency of failures relative to periods of usefulness. In system safety, it is also known as a reliability curve. 

A plot of the failure rate of a product as a function of time typically takes the shape of a bathtub curve (see Fig. 57.2). This curve illustrates the three phases that occur during the lifespan of a product from a reliability perspective. In the first, infant mortahty phase, there is an initially high but rapidly declining failure rate caused by infant mortahty. Infant mortality is typically caused by manufacturing defects that went undetected during inspection and testing and lead to rapid failure in service. Burn-in can be used to remove these units before shipment. The second phase, the normal operating life of the product, is characterized by a period of stable, relatively low failure rates. 

In the automobile industry, AEC (Q) 100 is used for complex components. It is a standard for the qualification of electric components. Simple components as resistors or capacitors are not covered in this standard. Since these simple components would often push aU statistic boimdaries through their variety of elements, such statistic observations are often insufficient for safety engineering. The risk for such simple components is that harmful components can be delivered to the production undetected. This is why the eligibility and whether the components are actually sufficiently dimensioned for their case of application are tested in the context of the qualification of the entire electric assembly group. The value for failiue rates is taken from the reliability handbooks. However, for the correct qualification including the proof of lifetime efficiency of the entire electronic assembly group it is assumed that the simple components is within the constant phase of failure rates of the bathtub curve. 

No definition of reliability is complete without an understanding of the bathtub curve. It is generally accepted that a component s (or part or sub-system) failure rate... 

FIGURE 512 Classic bathtub reliability curve showing the three stages during the life of a product from a reliability perspective infant mortality, steady-state, and wear-out. 

Hardware usually follows a bathtub shaped curve with a constant reliability for most of its operational life. With software the situation is different, as the reliability does not alter with time. Assuming the stated conditions remain unaltered, then software is not subject to rusting, corrosion, or whatever and that usually simplifies calculations considerably. Software reliability may fluctuate as it is modified and during testing and debugging. 

---