Product Specific Failure Data

It is clear that some are uncomfortable with the level of accuracy in the failure data estimated from industry databases and experience. Questions about failure rate versus stress conditions in particular applications come up. Questions about specific products are constantly being asked especially when one must attempt to pick a better product to achieve higher safety. 

Fortunately, several instrumentation manufacturers are providing detailed analysis of their products to determine a more accurate set of numbers useful for safety verification purposes. A Failure Modes Effects and Diagnostic Analysis (FMEDA) will provide specific failure rates for each failure mode of an instrumentation product. The percentage of failures that are safe versus dangerous is clear and relatively precise for each specific product. The diagnostic ability of the instrument is precisely measured. Overall, the numbers from such an analysis are indeed product specific and provide a much higher level of accuracy when compared to industry database numbers and experience based estimates. 

A FMEDA is sometimes done by the instrument manufacturer but typically done by third party experts. Often a product manufacturer does the work as part of an IEC61508 functional safety certification effort. Many different types of instruments have had this analysis done. A listing of 

It should be emphasized that a FMEDA provides failure rates, failure modes and diagnostic coverage effectiveness for random hardware failures. If done properly, it does not include failure rates due to systematic causes including incorrect installation, inadvertent damage, incorrect calibration or any other human error. 

---

For routine production, it is important to adequately record process details (e.g., time, temperature, equipment used) and to record all changes that have occurred. A maintenance log can be useful in performing failure investigations concerning a specific manufacturing lot. Validation data (along with specific test data) may also determine expected variance in product or equipment characteristics. 

Unsatisfactory Data If the stability data under the ICH conditions fall outside the acceptance criteria while data from the parallel study under the previously approved conditions or 25°C/ambient humidity, whichever applies, are satisfactory during the previously approved expiration dating period, and the added humidity is determined to be the cause for the stability failure, the product will still be considered to be in compliance with the regulatory specifications approved in the application. If the applicant decides to adopt the ICH conditions, a Changes Being Effected Supplement with a shortened expiration dating period or a Prior Approval Supplement with revised product specifications may be submitted where justified. Other measures (e.g., more protective container and closure or product reformulation) may be considered through a Prior Approval Supplement. 

Many companies have an internal expert who has studied these sources, as well as their own internal failure records, and maintains the company failure rate database. Some use failure data compilations found on the Internet. While the data in industry databases is not product specific or application specific, it does provide useful failure rate information for specific industries (nuclear, offshore, etc.) and a comparison of the data provides information about failure rates versus stress factors. 

Although product specific FMEDA reports offer superior data sources when compared to industry databases, they still do not account for application specific stress conditions that may affect actual failure rates. Ideally in the future manufacturers will be able to provide not only point estimates of failure rates but perhaps even equations with application specific variables to more precisely calculate the needed numbers. That wiU happen if there is demand and the needed data is collected. 

S Predicted cumulative failure probability (CFP) showing initial production period results based on standard materials data, and refined results from component specific materials data obtained by post exposure testing (PET) of samples taken at shutdown. 

Example. In the case of the BVA project, the taxonomy of (Beizer 1990) and historical failure data from similar projects were used as the basis for generating a product-specific taxonomy. An example of die BVA defect taxonomy is given in Table 3. 

The safety analysis techniques described in this book will facilitate ship safety assessment in various situations. They ean be tailored for safety analysis of any maritime and offshore engineering product with domain-specific knowledge. As some of these approaches described are subjective in nature, they may be more applicable for many engineering applications that lack reliable failure data. 

In Weibull analysis, the practitioner attempts to make predictions about the life of all products in the population by fitting a statistical distribution to life data from a representative sample of units. The parameterised distribution for the data set can then be used to estimate important life characteristics of the product such as reliability or probability of failure at a specific time, the mean life for the product and failure rate. Life data analysis requires the practitioner to ... 

The application of appropriate data to product design can mean the difference between the success and failure of manufactured products made from any material (plastic, steel, etc.). There are different sources of information on plastics. There is the data sheet compiled by a manufacturer of the material and derived from tests conducted in accordance with standardized specifications. Another source is the description of outstanding characteristics of each plastic, along with the listing of typical applications. 

If one is less restrained in setting specification limits, a balance can be struck between customer expectations and the risk and cost of failure a review of available data from production and validation runs will allow confidence limits to be calculated for a variety of scenarios (limits, analytical procedures, associated costs see Fig. 2.15 for an example). 

Deviations and failure investigation data—The failure of any batch to meet any specification, including batches failing in-process, release, or finished-product (shelf life) specifications is a crucial event. Such events must be reported and captured in the APR system. The identification of the root cause and the determination of corrective... 

The PAI inspection team must determine if there is valid and scientific justification for the failure to report data that demonstrate that the product failed to meet its predetermined specifications. Inspections should compare the results of analyses submitted with results of analyses of other batches that may have been produced. The PAI will determine if data submitted in the application are authentic and accurate and if procedures listed in the application were actually used to produce the data contained in the application. These procedures must be specific and well documented. The PAI, in addition, focuses on the following ... 

The investigation of failures of manufactured components and systems, especially in the electronics and aerospace industries, has generated a variety of statistical models on which data analysis may be based. Each model uses a specific distibution of failure probabilities, and it is important to select a model that matches the actual distribution inherent in the product concerned. In the case of dielectric breakdown, where a large number of quite different modes of failure are known to occur, sometimes even together, the application of a particular statistical failure model must be approached with great caution. Nevertheless, one treatment, based on a Weibull distribution of failure probability, has taken root, and is most generally used in practice. For a dielectric, the Weibull failure probability function has the form... 

Occasionally anecdotal data come to light on the cash outlays required for the development of specific NCEs. For example, in depositions filed for a patent infringement lawsuit, Genentech claimed it had spent 45 million to develop Protropin , its human growth hormone product, (494) and Eli Lilly certified that it had spent 16 million between 1980 and 1987 on its effort to develop its version of the drug (495). In another example, a 1980 report of the development cost of an oral systemic drug for chronic use estimated 21 million in outlays in the clinical period (226). Unfortunately, anecdotal estimates of this kind do not help verify industrywide costs, because they are self-selected and do not reflect the cost of failures or basic research. 

---