Khalifa N. Al-Khalifa

Condition-based maintenance for continuously monitored degrading systems with multiple failure modes

This article develops an optimum Condition-Based Maintenance (CBM) policy for continuously monitored degrading systems with multiple failure modes. The degradation of system state is described by a stochastic process, and a maintenance alarm is used to signal when the degradation reaches a threshold level. Unlike existing CBM models, this article considers multiple sudden failures that can occur during a systems degradation. The failure rate corresponding to each failure mode is influenced by either the age of the system, the state degradation of the system, or both. A joint model is constructed for the statistically dependent time-to-maintenance due to system degradation and time-to-failure of different failure modes. This model is then utilized to obtain the optimum maintenance threshold level that maximizes the systems limiting availability over its life cycle or minimizes the long-run cost per unit time. A numerical example, using real-life data from a reliability test of communication systems, is provided to demonstrate the application of the proposed approach.

Using the competing values framework to investigate the culture of Qatar industries

The last decade has witnessed a dramatic increase in the level of awareness and the incidents of implementation of total quality management (TQM). However, not all the applications have been successful. Amongst the many reasons given for the failures, the literature often points to the constraining effects of an organizations culture as a major barrier. In fact, many published studies that deal with the implementation of TQM almost exclusively acknowledge the importance of cultural factors on the success or failure of the venture. The focus of this study was to investigate the degree of compatibility of the culture that exists in the industries of one of the Arab countries, Qatar, and that required for implementing TQM. This paper, in particular, discusses how the competing values framework (CVF) can be used in this respect. (The authors had already used the framework to help identify the necessary TQM cultural profile.) The medium used to gather the information was a questionnaire, which was completed by 141 organizations. The results indicated that many organizations in the country were not characterized by just one cultural type, but a mix of two, which did not match the cultural profile characteristics that support TQM. Implementing organizational culture change is not an easy task, but this assessment of the current position, using the CVF, has highlighted where changes are needed to support a total quality approach.

Critical success factors of TQM: a UK study

This paper has presented the results of a study on Critical Success Factors for Total Quality Management (TQM) implementation in UK organisations. A survey instrument (10 Success Factors) developed by Yusof and Aspinwall for use in the automotive sector was used after some modification. The findings have supported previous study and validated the instrument as a reliable tool. The results indicated that the Manufacturing sectors in UK is embarked towards quality excellence. Suggestions were made to replicate the study in some developing countries.

First and Last Triggering Event Approaches for Replacement With Minimal Repairs

To make effective use of a running system, this paper proposes that, when replacement times could be scheduled at a planned time T of operation, and at a number N of minimal repairs, the newly proposed approach of whichever occurs last should be employed into optimizations for policy consideration. The replacement policy with the classical approach of whichever occurs first is also used to make comparison with the approach of whichever occurs last. Further, using the overtime technique, a replacement scheduled at time T in this paper is modified to be done at the first failure after T has arrived, to avoid probable interruptions for running jobs. That is, the system is replaced at failure N before time T, or at the first failure after T has arrived, whichever occurs first, and last, respectively. The policies acting on the approaches of first and last with minimal repairs are analyzed in detail, and their expected cost rates are formulated. Optimal solutions for separate and bivariate T and N are obtained and compared analytically.

Criticality measures for components with multi-dimensional degradation

Failures of engineering structures and equipment are often attributed to the failure of a single component. Hence, it is important to identify critical components in a system and understand how a components criticality changes over time under dynamic environments. This article investigates the criticality analysis for components with multiple competing failure modes due to degradation. The component degradation is modeled as a k-dimensional Wiener process. A component fails when any of the k degradation processes associated with that component attains a certain threshold level. Motivated by Nelsons cumulative exposure model, a relationship between both the mean and diffusion of the degradation process and environmental conditions is established. Expressions of a components criticality measures are derived. Numerical examples are presented to illustrate the criticality analysis.

Approximate methods for optimal replacement, maintenance, and inspection policies

It might be difficult sometimes to derive theoretical and numerical solutions for analytical maintenance modelings due to the computational complexity. This paper takes up several approximate models in maintenance theory, by using the cumulative hazard function H(t) and the newly proposed asymptotic MTTF (Mean Time to Failure) skilfully. We firstly denote by tx the time when the expected number of failures is x. Using H(tx)=x, we estimate failure times, model age and periodic replacements, and sequential imperfect maintenance. Motivated by the asymptotic method of computation of MTTF, we secondly model the expected cost rate for a parallel system when replacement is made at system failure, and give approximate computations for the sequential inspection policy. Optimizations of each model are obtained approximately in an easier way. When failure times have a Weibull distribution, it is shown from numerical examples that the obtained approximate optimal solutions have good approximations of the exact ones.

Adaptive cumulative sum charts with the adaptive runs rule

Adaptive cumulative sum (ACUSUM) charts, which adjust the reference value dynamically based on estimated shift size, provide good performance in detecting a range of mean shifts. However, when the range is wide, ACUSUM may not perform well for small shifts over the range. An adaptive runs rule, which is motivated by the concept of supplementary runs rule, is proposed, in order to make control charts more sensitive to small mean shifts. The adaptive runs rule assigns scores to consecutive runs based on the estimated shift size of the mean. The ACUSUM chart is supplemented with the adaptive runs rule to enhance its sensitivity in detecting small mean shifts. The average run length performance of the ACUSUM chart with the adaptive runs rule is compared with those of cumulative sum and variants of adaptive charts including ACUSUM. The experimental results reveal that the ACUSUM chart with the adaptive runs rule achieves superior detection performance over a wide range of mean shifts.

ISO 9000 performance among the Malaysian companies: the effects of motives

This paper presents the results of a study to identify the real motives behind ISO 9000 Certification. Towards this goal, a survey of some 300 ISO 9000 certified companies has been carried out for Malaysian manufacturing companies. Statistical tests on its relationship with performance improvement and the overall benefits and the barriers on gaining ISO 9000 certification are done. The paper also reveals other factors influencing the performance of ISO 9000, for example, the period of time after the company has been certified and size of company. Finally, factor analysis is applied to identify the main factors in certification motives, benefits, barriers, and their in-between relationship.

Multivariate statistical process control charts based on the approximate sequential χ2 test

Similar to the univariate CUSUM chart, a multivariate CUSUM (MCUSUM) chart can be designed to detect a particular size of the mean shift optimally based on the scheme of a sequential likelihood ratio test for the noncentrality parameter. However, in multivariate case, the probability ratio of a sequential test is intractable mathematically and the test statistic based on the ratio does not have a closed form expression which makes it impractical for real application. We drive an approximate log-likelihood ratio and propose a multivariate statistical process control chart based on a sequential χ2 test to detect a change in the noncentrality parameter. The statistical properties of the proposed test statistic are explored. The average runs length (ARL) performance of the proposed charts is compared with other MCUSUM charts for process mean monitoring. The experimental results reveal that the proposed charts achieve superior, both zero-state and steady-state, ARL performance over a wide range of mean shifts, especially when the dimension of measurements is large.

Economic cost models of integrated APC controlled SPC charts

In this paper, an economic cost model is proposed for processes integrating both automatic process control (APC) and statistical process control (SPC) for quality monitoring and control. Both the special cause and common cause variations are reduced by applying integrated APC and SPC. Traditionally, the integrated processes using APC and SPC are evaluated by the average run length (ARL). However, ARL may not be appropriate as a measurement of the economic design since it does not take into consideration the run length variation. Also, there are few studies that compare the cost models of such an integrated control system and the effect of cost parameters using different APC controllers. Therefore, we develop an economic cost model using non-homogenous Poisson process to describe the occurrence of an APC adjustment and develop a long run expected cost to investigate the use of different controllers in such integrated systems. Numerical examples are presented to demonstrate the applicability of the proposed model.