I am glad to bring this special issue on Reliability Engineering to the Quality Engineering audience. This is the second time that the ASQ Reliability Division arranged a reliability-focused issue. The first time was the July issue of 2015, and since then, several articles in that special issue had appeared in the top 20 downloaded articles from the publisher's website in 2016, which reflects the need of a reliability-focused issue.
No one would doubt that reliability and quality are two tightly intertwined subjects in industry. In fact, we often call reliability as the product/process quality with a time dimension. Therefore, many quality control and data analysis techniques can be directly used in reliability improvement program. For example, process variation reduction is important to both quality and reliability control. The case study by Voelkel included in this special issue indeed provides a variance component analysis for controlling a process quality characteristic that will eventually affect the reliability of the product. The article by Lv et al. presents a response surface methodology for optimizing dual objectives that involve a quality characteristic and a reliability characteristic and shows the tradeoff between these two objectives.
A broad range of reliability research topics are explored in this special issue. I try to summarize these articles by the models and data analysis methods they used and the application areas they targeted on. First, designing statistically efficient and economically affordable reliability tests has always been the goal of reliability test planning. Optimal testing plans for accelerated life tests and step-stress accelerated life tests and optimal inspection plans for interval censored reliability tests are discussed in the articles by Han and by Lu and Anderson-Cook, respectively. Both articles emphasize on the utility of cost function as either the objective function or the constraint for test plan optimization. To plan reliability demonstration tests, Chen et al. control the consumer's risk, another facet of cost. Second, reliability data analysis and reliability prediction are still the main task for reliability engineers. We have seen more and more research interests in degradation modeling and data analysis in recent years, probably due to the availability of online monitoring data generated by sensors. The articles by Yu et al., Liu et al., and Fugate et al. are all about degradation models, while Yu et al. focus on an evolutional optimization method for model parameter estimation and Liu et al. and Fugate et al. develop novel degradation models for certain datasets. Furthermore, in this special issue we have seen many applications of Bayesian inference methodology on reliability test planning and data analysis. For example, Quinlan et al. propose a Bayesian experimental design method with weighted multiple priors; Banghart et al. apply a Bayesian network model on assessing potential malfunction risks on military aircrafts. Finally, other topics, including reliability function comparison, warranty data analysis, software reliability, and reliability growth models, have been tackled by Stevens and Anderson-Cook, Dai et al., Wang, and Talafuse and Pohl, respectively.
Even though so many aspects of reliability engineering have been touched upon in this special issue, I have to admit this is only the tip of the iceberg. Including reliability-related publications in Quality Engineering has certainly broadened the scope of the journal and will potentially attract more quality and reliability practitioners to the journal. Lastly, I want to thank all the referees who had contributed their time and efforts on reviewing these articles, and hope you will enjoy them.