Process Improvement in the Microelectronic Industry by State Space Modelling

Triantafyllopoulos, Kostas; Godolphin, J. D.; Godolphin, E. J.

doi:10.1002/qre.734

Cited by 6 publications

(11 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First we fit a local level model to the multivariate process and then we apply a univariate modified EWMA control chart to the logarithm of the Bayes' factor to monitor the dispersion of the predictive distribution of the data from the target distribution. Our model makes use of a generalization of the Shewhart-Deming model for multivariate autocorrelated processes (Deming 29 , del Castilo 3 , Triantafyllopoulos et al 30 ).…”

Section: Introductionmentioning

confidence: 99%

Multivariate Control Charts Based on Bayesian State Space Models

Triantafyllopoulos

2006

Quality & Reliability Eng

Self Cite

View full text Add to dashboard Cite

This paper develops a new multivariate control charting method for vector autocorrelated and serially correlated processes. The main idea is to propose a Bayesian multivariate local level model, which is a generalization of the Shewhart-Deming model for autocorrelated processes, in order to provide the predictive error distribution of the process and then to apply a univariate modified EWMA control chart to the logarithm of the Bayes' factors of the predictive error density versus the target error density. The resulting chart is proposed as capable to deal with both the non-normality and the autocorrelation structure of the log Bayes' factors. The new control charting scheme is general in application and it has the advantage to control simultaneously not only the process mean vector and the dispersion covariance matrix, but also the entire target distribution of the process. Two examples of London metal exchange data and of production time series data illustrate the capabilities of the new control chart.

show abstract

Section: Introductionmentioning

confidence: 99%

Multivariate Control Charts Based on Bayesian State Space Models

Triantafyllopoulos

2006

Quality & Reliability Eng

Self Cite

View full text Add to dashboard Cite

show abstract

“…The proposed methodology is a semi-conjugate Bayesian procedure and, although it involves a simulation step, it is found to be fast and accurate. The predictions produced outperform the relevant predictions of Box and Lucen˜o [4,5] and Triantafyllopoulos et al [18] returning notably lower sum of squared errors. The benefit gained from the entire distribution of the NVR is a step forward in the development of the local level model with theoretical and practical advantages.…”

Section: Introductionmentioning

confidence: 48%

“…40 or 60 time points), it seems that the Bayesian paradigm is the appropriate approach to be considered, since with the inclusion of prior information, this can lead to more accurate predictions [15]. Triantafyllopoulos et al [18] adopt a local level predictor for feedback adjustment and these authors show that the noise variance ratio (NVR) plays an important role in control. In the above paper it is shown that by manually changing the NVR a much improved adjusted process can be achieved with much lower sum of squared errors than in the usual EWMA.…”

Section: Introductionmentioning

confidence: 98%

Feedback quality adjustment with Bayesian state‐space models

Triantafyllopoulos

2006

Appl Stoch Models Bus & Ind

Self Cite

View full text Add to dashboard Cite

SUMMARYIn this paper we develop a Bayesian procedure for feedback adjustment and control of a single process. We replace the usual exponentially weighted moving average (EWMA) predictor by a predictor of a local level model. The novelty of this approach is that the noise variance ratio (NVR) of the local level model is assumed to change stochastically over time. A multiplicative time series model is used to model the evolution of the NVR and a Bayesian algorithm is developed giving the posterior and predictive distributions for both the process and the NVR. The posterior distribution of the NVR allows the modeller to judge better and evaluate the performance of the model. The proposed algorithm is semi-conjugate in the sense that it involves conjugate gamma/beta distributions as well as one step of simulation. The algorithm is fast and is found to outperform the EWMA and other methods. An example considering real data from the microelectronic industry illustrates the proposed methodology.

show abstract

“…Without accurate and timely forecasts, planning, scheduling, and procurement decisions can be seriously in error, leading to waste and inefficiency throughout the supply chain. Triantafyllopoulos et al 6 provided an excellent example of how time series analysis methods were used in improving performance in the microelectronics industry.Finally, a reminder to readers that Quality and Reliability Engineering International 2005; 21(3):221-328 was a special issue devoted to Six Sigma that explored many ideas useful for extending, strengthening, and expanding its applications. Many of the articles in that special issue are useful reading to accompany the article by Tang et al 1 .…”

mentioning

confidence: 99%

mentioning

confidence: 99%

Untitled

2007

Quality & Reliability Eng

View full text Add to dashboard Cite

A recent special issue of the Journal (Quality and Reliability Engineering International 2007; 23(1):1-154) contained selected papers from the Fourth International Conference on Quality and Reliability. A paper in this issue by Tang et al. 1 discussed the importance of enhancing and increasing the usefulness of Six Sigma by incorporating new techniques into the familiar DMAIC (Define, Measure, Analyze, Improve, Control) problem-solving framework. Many of these techniques are from the operations research/management science fields. The authors argue convincingly that Six Sigma and DMAIC cannot remain static if they are to sustain the value proposition for business and industry. I think that this is particularly true as Six Sigma moves beyond the traditional manufacturing and engineering applications into transactional and service industries. I would like to suggest that there are other techniques that need to be included in the Six Sigma practitioner's toolkit.Discrete-even simulation methods are critical to the implementation of Six Sigma in many environments. In transactional and service industries, it is almost impossible to experiment directly with the actual system, so most efforts directed towards improvement and studying the impact of system changes must involve a simulation model of the system. For example, if a Six Sigma team is investigating how to reduce waiting times in a hospital emergency room, it will be virtually impossible to make changes in the actual emergency room. However, with a properly developed and validated simulation model of the emergency room, changes in operating policy, staffing, and resource availability can be easily and efficiently investigated. A recent paper in the Journal by Abbas et al. 2 discusses how simulation can be used to optimize the design of clinical trials.Another important aspect of Six Sigma training is to include more information about engineer process control (EPC) methods (as opposed to process monitoring, such as control charts). EPC is widely used in the chemical and process industries to keep critical process outputs on target by making appropriate adjustments to process inputs, taking advantage of known relationships between the inputs and the outputs. These techniques deal with variability not by eliminating assignable causes but by transferring the variability in the output to the input. For example, we may keep the molecular weight near to the desired target by making regular adjustments to temperature via a technique such as feedback control. There has been considerable research on EPC methods and methods to integrate EPC and SPC are reasonably well known. However, there is usually little effort in Six Sigma training devoted to EPC, even for individuals who will practice Six Sigma in the chemical and process industries. Recent papers in the Journal devoted to EPC and SPC include Jaing et al. 3 , Runger et al. 4 , and Yang and Sheu 5 .Forecasting and time series methods are not usually sufficiently addressed in many Six Sigma training programs. Forecasti...

show abstract

Process Improvement in the Microelectronic Industry by State Space Modelling

Cited by 6 publications

References 12 publications

Multivariate Control Charts Based on Bayesian State Space Models

Multivariate Control Charts Based on Bayesian State Space Models

Feedback quality adjustment with Bayesian state‐space models

Untitled

Contact Info

Product

Resources

About