New control charts for monitoring the Weibull percentiles under complete data and Type‐II censoring

In this paper, we propose 3 new control charts for monitoring the lower Weibull percentiles under complete data and Type‐II censoring. In transforming the Weibull distribution to the smallest extreme value distribution, Pascaul et al (2017) presented an exponentially weighted moving average (EWMA) control chart, hereafter referred to as EWMA‐SEV‐Q, based on a pivotal quantity conditioned on ancillary statistics. We extended their concept to construct a cumulative sum (CUSUM) control chart denoted by CUSUM‐SEV‐Q. We provide more insights of the statistical properties of the monitoring statistic. Additionally, in transforming a Weibull distribution to a standard normal distribution, we propose EWMA and CUSUM control charts, denoted as EWMA‐YP and CUSUM‐YP, respectively, based on a pivotal quantity for monitoring the Weibull percentiles with complete data. With complete data, the EWMA‐YP and CUSUM‐YP control charts perform better than the EWMA‐SEV‐Q and CUSUM‐SEV‐Q control charts in terms of average run length. In Type‐II censoring, the EWMA‐SEV‐Q chart is slightly better than the CUSUM‐SEV‐Q chart in terms of average run length. Two numerical examples are used to illustrate the applications of the proposed control charts.     

EWMA chart based on Bayes‐conditional pivotal quantity for Weibull percentiles under complete data and type‐II censoring

Bayes‐conditional control chart has been used for monitoring the Weibull percentiles with complete data and type‐II censoring. Firstly, the Weibull data are transformed to the smallest extreme value (SEV) distribution. Secondly, the posterior median of quantiles is used as a monitoring statistic. Finally, a pivotal quantity based on the monitoring statistic with its conditional distribution function is derived for obtaining the control limits. This control chart is denoted as Shewhart‐SEV‐ . In this study, we extend this work based on an exponential weighted moving average model named exponential weighted moving average‐SEV‐ for monitoring the Weibull percentiles. We provide the statistical properties of the monitoring statistic. The average run length and the standard deviation of run lengths, computed by the integral equation approach, are used as performance measures. The results indicate that the proposed chart performs better than the Shewhart‐SEV‐ . The breaking strength of carbon fibers is used to illustrate the application of the proposed control chart.     

One‐sided control chart based on support vector machines with differential evolution algorithm

The statistical learning classification techniques have been successfully applied to statistical process control problems. In this paper, we proposed a one‐sided control chart based on support vector machines (SVMs) and differential evolution (DE) algorithm to monitor a process with multivariate quality characteristics. The SVM classifier provides a continuous distance from the boundary, and the DE algorithm is used to obtain the optimal parameters of the SVM model by minimizing mean absolute error (MAE). The average run length of the proposed chart is computed using the Monte Carlo simulation approach. Several simulated cases are conducted using a multivariate normal distribution with 10 and 20 dimensions and three different process shift scenarios. In addition, we consider two non‐normal distribution cases. The ARL performance of the proposed chart is better than the distance‐based SVM chart. A real example is used to illustrate the application of the proposed control chart.     

A comparison study of control charts for Weibull distributed time between events

Monitoring decreases in the mean of Weibull time between events data to address process quality deteriorations is an important task in reliability analysis. Two new control charts such as Weibull exponentially weighted moving average and mixed cumulative sum‐exponentially weighted moving average by transforming the Weibull data to the exponential data are proposed and compared with 2 existing control charts such as Weibull cumulative sum and mixed exponentially weighted moving average‐cumulative sum. The performance comparison provides a way to select a specific control chart in a given situation. The average run length and the standard deviation of the run length are used as performance measures. The relative mean index is also utilized to measure the overall performance. The smaller the value of the relative mean index, the better the performance of the control chart and vice versa. Two illustrative examples are provided to show the applications of the proposed control charts.     

Deuterium isotope analysis of methanol oxidation on mixed metal anode catalysts

The widely accepted mechanism for methanol oxidation on Pt based catalyst surfaces has held that the rate determining step is activation of water, and/or oxidation of surface-bound CO to CO₂. In fact on pure Pt, water activation is always rate limiting at potentials negative of 0.6 V. Anode potentials greater than 0.4 V are outside the useful potential window of direct methanol fuel cells when using Nafion 117 at 60 °C. Enhancement of the water activation kinetics on Pt has been effected by the use of oxophilic transition metal promoters including Ru, W and Sn. For decades the search for improved methanol oxidation electrocatalysts has focused on water activation. A systematic deuterium isotope study on Pt black and two active mixed metal catalysts (PtRu and PtRuOsIr) shows that for each catalyst there is a characteristic transition potential above which the primary reaction in the rate-determining step changes from water activation to CH bond activation. On the mixed metal catalysts, this crossover potential is ca. 0.35 V, which is within the direct methanol fuel cell potential window (0-0.400 V). This study confirms that on these active catalysts there is a potential above which further improvements in water activation must be concomitant with acceleration of CH bond activation. Thus the catalyst search strategy involving Pt promoter metals must also consider the kinetic importance of CH bond activation.     

A double exponentially weighted moving average chart based on likelihood ratio for monitoring an inflated Pareto process

In this paper, we present a new chart called a likelihood ratio based double exponentially weighted moving average (LR_DEWMA) chart to monitor the shape parameter of the inflated Pareto process. Three other control charts such as the Shewhart type, the classical cumulative sum (CUSUM), and the likelihood ratio based EWMA (LR_EWMA) charts are also investigated. The performance of the control charts is evaluated by the average run length (ARL) and standard deviation of run lengths (SDRL) computed through the Monte Carlo simulation approach. Moreover, the median run length (MRL) and some other run length (RL) percentiles are also considered in some cases. Different charts have shown the best performance in different cases. In detecting smaller shifts, while the LR_DEWMA chart outperformed the other charts in terms of ARL and MRL, the CUSUM chart has shown the best performance in terms of SDRL and IQR of RLs. The application of the proposed control charts is illustrated using a chromatography analyses data from the food industry.     

Clustering hypervisors to minimize failures in mobile cloud computing

Resource virtualization has become one of the key super‐power mobile computing architecture technologies. As mobile devices and multimedia traffic have increased dramatically, the load on mobile cloud computing systems has become heavier. Under such conditions, mobile cloud system reliability becomes a challenging task. In this paper, we propose a new model using a naive Bayes classifier for hypervisor failure prediction and prevention in mobile cloud computing. We exploit real‐time monitoring data in combination with historical maintenance data, which achieves higher accuracy in failure prediction and early failure‐risk detection. After detecting hypervisors at risk, we perform live migration of virtual servers within a cluster, which decreases the load and prevents failures in the cloud. We performed a simulation for verification. According to the experimental results, our proposed model shows good accuracy in failure prediction and the possibility of decreasing downtime in a hypervisor service. Copyright © 2017 John Wiley & Sons, Ltd.     

Analyzing the influence of changes in land use and management practices on the lag time of peak flows for tropical watersheds of Ethiopia

Changes in peak flow response time governed by rainfall and physical configurations of watersheds have been the topic of many studies, but other factors are also important. We aimed to analyze the dynamics in the lag time (T_L) of peak flow for a tropical watershed (Kecha) and its sub-watersheds (Dokmit, Zenjero Maderia, and Wotit Minch) in Ethiopia, as influenced by land-use changes from 1982 to 2017 and the implementation of soil and water conservation (SWC) practices since 2011. Layered maps of land use, soil type, and SWC practices were used to determine the curve number, and the Natural Resource Conservation Service hydrologic model was used to estimate T_L. We compared the estimated values against measured median values of T_L for 30 rainfall–runoff events. The estimated T_L of 1982, 2005, and 2017 varied from 9 to 19 min, 8 to 18 min, and 10 to 22 min, respectively, in the Dokmit, Zenjero Maderia, and Wotit Minch sub-watersheds. The smallest (8 min) and greatest (22 min) values of T_L were observed during 2005 and 2017, respectively. These results are likely attributable to the increased amount of cultivated area at the expense of bushland and forest in 2005 as compared to 1982, and the implementation of SWC practices in 2011–2017. Dokmit had the lowest T_L values of the three sub-watersheds, most likely because of its greater coverage of grazing and cultivated lands and degraded Nitic Luvisol. The variation in T_L values among the sub-watersheds was related to changes in both land use and SWC practices but also to the presence of degraded areas inherited from past human activities. Overall, such spatiotemporal flow response time analysis can provide useful information for the proper design of sustainable development strategies for particular niches in tropical regions of Ethiopia and elsewhere.