Mean time before failure

Reliability data, e.g. mean time before failure. 

The lying down of levels of quality and reliability necessary to ensure product success and acceptability in a particular market is a cause for increasing concern. They are the most difficult aspects to quantify in absolute terms, although statistical data fi-om company product precedents are helpful here. There are expressions used such as mean time before failure (MTBF) and mean time to repair (MTTR) that are used with mechanical, hydraulic, pneumatic, electrical, and other products. Nonetheless, some quantitative expression must be made in respect of quality and reliability at the initial design specification stage. A Company must ensure adequate feedback of any failure analysis to the design team. 

In order to determine the value of 4> it is necessary to subject at least two groups of specimens to lifetests at different temperatures and to measure a parameter, such as the mean time before failure (MTBF), related to the rate of degradation. For most thermally induced failure mechanisms, the value of will lie in the range 0 45-l 2eV. 

Second, guaranteeing availability as high as 99-9999 percent requires a tremendous amount of performance data on every single piece of equipment. Accurate risk-assessment analysis requires reliable data such as mean time between failure (how long a component is likely to run before breaking down) and mean time to repair (how long it will take to fix a component that has broken down). Analysts would prefer to have as much as i million hours of data on each and every system component. That takes years to... 

A representative stress-strain creep curve for a 5X CMC specimen is shown in Figure 16. In analyzing these hmited data, it was concluded that extrapolating the estimated creep rates to longer times and lower and high stress levels could not be done with accuracy. However, as a means of comparing this composite system with others, the data can be used to estimate the stress level to exceed a time-before-failure of 100 hours. These data indicate that a stress level of 11 ksi will just exceed the 100-hour target. 

For the purpose of this case example let s assume that the mean time to failure (MTTF) must be at least two years. The reliability target may also be defined as a minimum fail free operational time, where the probability of a failure before say 6 months is required to be less than 10 %. 

Consider a transfer line with m stages and m - 1 banks of capacities Zi, Z2, , Zm-i- The number of cycles of operation before failure and the repair times aU have geometric distributions with mean 1/flj and 1/bi respectively for stage i, I = 1,. . . , m. The parameters of stage i are then (a, b ). 

Formulas for MTTF are derived and often used for products during the useful life period. This method excludes wearout failures. This often results in a situation in which the MTTF is 300 years and useful life is only 40 years. Note that instruments should be replaced before the end of their useful life. When this is done, the mean time to random failures will be similar to the number predicted. 

Log-normally distributed with mean equal to thee MTTR and std. dev. equal MTTR/2. Components w/several failure modes as bad as old i.e. set in the condition they were right before failure. Components with one failure mode as good as new . Duration of replacement and regular repair is the same when spare parts are available. Spare part order time is added when not available. 

But a hazard will only result if there is a failure of the safety-related system that is followed by a demand before the next proof test. The mean time to a hazard (MTTH) is therefore ... 

As said many times before in this book, failures do not necessarily mean hazards, and we can have hazards without failures. When looking for safety data, specifically hazard information, it is important to treat it with the same trepidation that you treat taking failure data from one context to another. A system operating in one condition will not necessarily have the same hazards operating in a different one. 

Handling these three types of failures under the specified operating conditions of the components is what we call maintenance. Preventive maintenance aims to eliminate potential failures before they occur, e.g. by measuring wear, or by replacement or retraining prior to rapid increases in the failure rate corrective maintenance rectifies failures on occurrence. As components fail and are repaired or replaced, the performance of the system will fluctuate, and at some point in time it is possible that the performance becomes so poor that we say the system has failed. This leads to the definition of system failure rate, X, as the inverse of the mean time between such failures and, given the specified operating conditions. 

Residues at the MRL in a food commodity usually result in intakes well below the ADI. As a result, residues above the MRL do not represent any immediate risk to a consumer. A residue will need to be many times greater than the MRL in a food before the ADI is approached by the consumer. However, in the face of strict trade requirements, a residue above the MRL could mean failure to meet specifications and loss of access to export markets. 

It was pointed out in Chap. 8, Sect. 2.1 that there are primarily two reasons for the failure of the diffusion equation to describe molecular motion on short times. They are connected with each other. A molecule moving in a solvent does not forget entirely the direction it was travelling prior to a collision [271, 502]. The velocity after the collision is, to some degree, correlated with its velocity before the collision. In essence, the Boltzmann assumption of molecular chaos is unsatisfactory in liquids [453, 490, 511—513]. The second consideration relates to the structure of the solvent (discussed in Chap. 8, Sects. 2.5 and 2.6). Because the solvent molecules interact with each other, despite the motion of solvent molecules, some structure develops and persists over several molecular diameters [451,452a]. Furthermore, as two reactants approach each other, the solvent molecules between them have to be squeezed-out of the way before the reactants can collide [70, 456]. These effects have been considered in a rather heuristic fashion earlier. While the potential of mean force has little overall effect on the rate of reaction, its effect on the probability of recombination or escape is rather more significant (Chap. 8, Sect. 2.6). Hydrodynamic repulsion can lead to a reduction in the rate of reaction by as much as 30-40% under the most favourable circumstances (see Chap. 8, Sect. 2.5 and Chap. 9, Sect. 3) [70, 71]. 

Redundancy. The existence of more than one means for accomplishing a given task, where all means must fail before there is an overall failure to the system. Parallel redundancy applies to systems where both means are working at the same time to accomplish the task, and either of the systems is capable of handling the job itself in case of failure of the other system. Standby redundancy applies to a system where there is an alternate means of accomplishing the task that is switched in by a malfunction-sensing device when the primary system fails Ref Anon, OrdTechTerm (1962), 251—52... 

Droplet growth rates and viscosity decline rates both are exponential processes, following a straight line on a semi-log plot (log x or log vs. time), where is the mean droplet diameter. Emulsion failure is also associated with a certain minimum viscosity, depending on water content, crude-oil content, temperature, etc. Viscosity and mean droplet size may be projected to estimate the time remaining before emulsion failure. The ultimate droplet size and viscosity should be determined experimentally for the same formulation in a pilot-plant pipe loop. 

---