International Journal of Reliability, Quality and Safety Engineering

The limitation of the trend test in SPC is reviewed. The current Nelson's run test for the six or seven consecutive increasing or decreasing points cannot detect non-monotonic increasing or decreasing trend that is frequently encountered in industries. This paper proposes the reverse arrangement test (RAT) to replace the current SPC trend test rule in order to detect both non-monotonic and monotonic increasing or decreasing trends. The theory of RAT is reviewed and we point out the errors in the tables from a well-known and most frequently referenced paper on RAT. The corrected tables of accumulated probabilities for each total reverse arrangement for N = 3 to 12 are presented. The false alarm rate from RAT is verified with excellent accuracy by 10⁸ data points simulating stable normal distributed process. Real examples from the semiconductor manufacturing illustrate that 7 non-monotonic increasing points can be detected by RAT with the same false alarm rate as for the 6 monotonic increasing trend, while sometimes none of the current WECO and Nelson's rules can detect such abnormal pattern. Applications of RAT in IC SPC and WLRC (Wafer Level Reliability Control) show RAT is much more powerful than conventional monotonic trend tests in detecting nonconformance.

This paper is focused on the multiobjective design of equivalent accelerated life test (ALT) plans. Equivalent ALT plans are expected to achieve the same statistical performance as a baseline ALT plan yet lead to other desired performance measures such as reduced test time and total cost. Before determining the desired multiobjective equivalent ALT plans, an efficient fast non-dominated sorting genetic algorithm (NSGA-II) is utilized to identify a set of Pareto optimal solutions. To handle a large number of Pareto optimal solutions, a self-organizing map (SOM) and data envelopment analysis (DEA) are sequentially applied to classify the Pareto solutions and reduce the size of the suggested solution set. This integrated approach allows for the tradeoff of information among the Pareto solutions and the reduction in the size of the solution set. It provides a useful tool for practitioners to make meaningful decisions in planning ALT experiments.

Corrosion initiation time of steel reinforcement for partially saturated concrete members subjected to chloride ingress is investigated at five places along Indian coasts using simplified probabilistic analysis and sensitivity of the parameters to reinforcement corrosion is studied. Previously proposed diffusion based chloride ingress model for constant surface chloride concentration and variable diffusion coefficient is used. Corrosion is initiated when the chloride concentration on steel reinforcement exceeds a threshold value. The various input parameters considered in the model are surface chloride concentration (C_s), chloride threshold value (C_th), Concrete cover (X), reference diffusion coefficient (D_ref), age (t), reference age (t_ref), diffusion decay index (m) and monthly temperature and relative humidity correction factors f(T) and f(h) respectively; to reference diffusion coefficient. For prediction of corrosion initiation time temperature and relative humidity data collected by India metrological Department is used and all other parameters are assumed same. A large variation in corrosion initiation time is found and therefore it becomes necessary to consider the temperature and relative humidity data in a region. Life-365 considers fully saturated condition of concrete and thus underestimates the corrosion initiation time of a reinforced concrete member. The deterministic approach uses mean values of the parameters and provides a 50% probability of corrosion initiation but the probabilistic approach provides expected time of first corrosion for chosen risk of corrosion and also gives sensitivity of parameters to probability of corrosion. Corrosion initiation time in ascending order is found at Colaba, Kanyakumari, Santacruz, Chennai and Vishakhapatnam. Corrosion initiation time is useful for owner, designer, or to an organization to take decision in time of priority of repairs, repair strategy, corrosion protection in order to optimize maintenance planning and budgeting. Planned maintenance at the optimum time is the safest and most cost effective approach.

The problem of optimizing accelerated production testing is a pressing one in most electronic manufacturing facilities. Yet, practical models are scarce in the literature, especially for testing high volumes of electronic circuit packs in failure-accelerating environments. In this paper, we develop both a log-linear and linear model, based initially on the Weibull distribution. The models developed are suitable for modeling accelerated production testing data from a temperature-cycled environment. The model is "piecewise" in that the failures in each discrete "piece" of the temperature cycle are modeled as if the testing was in parallel rather than sequential mode. An extra covariate is introduced to indicate age at the start of each piece. The failures in a piece then depend on the stress in the piece itself and the time elapsed to the start of the piece. This last dependence captures the influence of reliability growth and has the result of providing an alternative linear model to the log-linear one. The paper demonstrates a simpler use of Poisson regression. An application, using actual production data, is described. Uses of the Loglogistic, Logistic, Lognormal and Normal distributions are also illustrated.

Software operational profile (SOP) has found its extensive use in applications such as estimation of software reliability, allocation of testing resources, evaluation of software quality attributes, performance analysis and testing of software. However, benefits of the SOP can only be reaped if it is developed during early stages of software life cycle and updated for changes in operational structure with the progress in development of software process. In the early stages of software life cycle usage data is available only in linguistic form. This linguistic data may be obtained from previous release of software usage data, marketing surveys, experts or user's opinion or from combinations of these. Moreover, due to limited data resources and large efforts required to convert the gathered data into point estimates, reluctance is observed towards the development of SOP in spite of its numerous benefits. This paper proposes a fuzzy software operational profile (FSOP) approach, which takes input data in the form of linguistic variables from experts and assigns occurrence possibilities to the input parameters. Linguistic data collected from the experts are processed to obtain the resultant fuzzy operational profile. Further, the resultant profile is defuzzified and normalized to estimate occurrence probability of operations. The proposed method has been applied to various software systems and results obtained are compared with existing practice. Besides simplicity, FSOP can easily be altered for changes in operational architecture and frequencies simply by changing the location and value of the affected linguistic variable.