Could electronics reliability be predicted, quantified and assured?

Abstract

Application of the consistent, comprehensive and physically meaningful probabilistic design for reliability (PDfR) concept can not only help to understand the physics-of-failure of an electronic product, but, most importantly, can enable one to predict, quantify and assure its failure-free performance in the field. The use of the PDfR concept can be helpful also in the development and implementation of the new generation of the most feasible and effective qualification test (QT) methodologies, practices and specifications. The major ten PDfR requirements (“commandments”) for the predicted, quantified and assured reliability of an electronic or a photonic product could be formulated as follows:(1)PDfR approach is an effective means for improving the state-of-the-art in the field, having in mind that nothing is perfect, and that the difference between an unreliable product and a robust one is “merely” in the level of the never-zero probability of failure (PoF).(2)The best electronic product is, in effect, the best compromise between the needs for its reliability, cost effectiveness and time-to-market (completion) for a particular product and application.(3)Reliability cannot be low, need not be higher than necessary, but, for a cost-effective and a timely product, has to be adequate for a particular product and application.(4)When reliability is imperative, ability to quantify it is a must, especially if optimization is considered: no optimization is possible, of course, if the product’s reliability characteristics of interest are not quantified.(5)One cannot design a product with predicted, quantified, optimized and assured reliability by limiting the effort to the highly accelerated life testing (HALT): HALT can test the reliability limits and perhaps to ruggedize the product, but does not quantify reliability.(6)Reliability is conceived at the design stage and should be taken care of, first of all, at this stage, when a “genetically healthy” product is supposed to be created; if the reliability of the product is taken care at this stage, then the subsequent fabrication, qualification and prognostics-and-health-monitoring (PHM) stages will have much better chances to succeed.(7)Reliability evaluations and assurances cannot be delayed until the product is fabricated and shipped to the customer, i.e., cannot be left to the PHM stage: it is too late at this stage to change the design or the materials for improved reliability; that is why, when high reliability is critical (e.g., in the aerospace and military electronics), users have to re-qualify devices to assess their (remaining) useful lifetime (RUL) and to use redundancy in an attempt to build a reliable system out of insufficiently reliable components.(8)Design, fabrication, testing, qualification and PHM efforts should consider, and be specific for, particular products and their most likely actual or at least anticipated applications.(9)Highly cost-effective and highly focused failure oriented accelerated testing (FOAT) geared to a particular pre-determined relevant reliability model and aimed at understanding the physics of failure anticipated by this model is an important constituent part of the PDfR concept and effort.(10)Effective, easy-to-use and physically meaningful predictive modeling (PM) is another important constituent of the PDfR approach; in combination with FOAT, it is a powerful means to carry out meaningful sensitivity analyses (SA), so that the operational reliability of the product is effectively predicted, quantified and assured (“principle of practical confidence”). Analytical (“mathematical”) modeling occupies a special place in the modeling effort, because of its compactness and explicit indication on “what affects what” and what could possibly be done to improve the product’s performance.In the write-up that follows the above requirements (“commandments”) are addressed and discussed in detail.