ENBIS-18 in Nancy

2 – 25 September 2018; Ecoles des Mines, Nancy (France) Abstract submission: 20 December 2017 – 4 June 2018

My abstracts

 

The following abstracts have been accepted for this event:

  • Using Risk Metrics for Performance Evaluation of Control Charts

    Authors: Christian Weiß (Department of Mathematics and Statistics, Helmut Schmidt University Hamburg), Murat Caner Testik (Hacettepe University, Department of Industrial Engineering)
    Primary area of focus / application: Process
    Keywords: Value at risk, Tail conditional expectation, Expected shortfall, Run length performance, Phase-I analysis, Guaranteed conditional ARL performance
    Submitted at 21-Mar-2018 13:23 by Christian Weiß
    Accepted (view paper)
    3-Sep-2018 15:10 Using Risk Metrics for Performance Evaluation of Control Charts
    Control charts are commonly evaluated in terms of their average run length (ARL). However, since run length distributions are typically strongly skewed, the ARL gives a very limited impression about the actual run length performance. In this study, it is proposed to evaluate a control chart's performance using risk metrics, specifically the value at risk and the tail conditional expectation. When a control chart is evaluated for an in-control performance, the risk is an early occurrence of a false alarm, whereas in an out-of-control state, there is a risk of a delayed detection. For these situations, risk metric computations are derived and exemplified for diverse types of control charts. It is demonstrated that the use of such risk metrics leads to important new insights into a control chart's performance. Finally, we also show how risk metrics can be used to analyze the estimation uncertainty in evaluating a control chart's performance if the design parameters rely on a Phase-I analysis.
  • Detecting Deviating Data Cells

    Authors: Peter Rousseeuw (University of Leuven), Wannes Van den Bossche (University of Leuven)
    Primary area of focus / application: Other: invited to ASQ session
    Secondary area of focus / application: Process
    Keywords: Cellwise outlier, Missing values, Multivariate data, Process control, Robust estimation, Rowwise outlier
    Submitted at 21-Mar-2018 15:56 by Peter Rousseeuw
    Accepted (view paper)
    4-Sep-2018 15:50 Detecting Deviating Data Cells
    A multivariate dataset consists of n cases in d dimensions, and is often stored in an n by d data matrix. It is well-known that real data may contain outliers. Depending on the situation, outliers may be (a) undesirable errors which can adversely affect the data analysis, or (b) valuable nuggets of unexpected information. In statistics and data analysis the word outlier usually refers to a row of the data matrix, and the methods to detect such outliers only work when at least half the rows are clean. But often many rows have a few contaminated cell values, which may not be visible by looking at each variable (column) separately. We propose the first method to detect deviating data cells in a multivariate sample which takes the correlations between the variables into account. It has no restriction on the number of clean rows, and can deal with high dimensions. Other advantages are that it provides predicted values of the outlying cells, while imputing missing values at the same time. We illustrate the method on several real data sets, where it uncovers more structure than found by purely columnwise methods or purely rowwise methods. The proposed method can help to diagnose why a certain row is outlying, e.g. in process control. It may also serve as an initial step for estimating multivariate location and scatter matrices.
  • Show the Consequences of the Traditional ISO Method for Non-Normal Ppk versus an Alternative Proposal

    Authors: Didier Dumont (SKF France S.A.), Bryan Dodson (SKF Group Six Sigma), Rene Klerx (SKF Group Six Sigma)
    Primary area of focus / application: Quality
    Secondary area of focus / application: Six Sigma
    Keywords: Process capability, Non-normal distribution, Percentile method, Z-score
    Submitted at 23-Mar-2018 11:45 by Rene Klerx
    Accepted
    4-Sep-2018 10:30 Show the Consequences of the Traditional ISO Method for Non-Normal Ppk versus an Alternative Proposal
    Ppk is a widely measure for process capability, and is regarded as a measure of non-conformance. For example, a Ppk of 1.33 represents 33 parts per million not conforming to requirements given a one sided specification. When non-normal distributions are encountered, the ISO method for computing Ppk with non-normal distributions does not allow a direct transformation between Ppk and non conformance. When using popular software, the calculation method is not always understood, and in some cases, different methods are used in different modules of the same software package This presentation explores alternative methods, including the ISO method, for computing Ppk for non normal distributions along with advantages and disadvantages.
  • A Combined Shewhart-CUSUM Chart with Switching Limit

    Authors: Sebastian Ottenstreuer (Helmut Schmidt University), Christian H. Weiß (Helmut Schmidt University), Sven Knoth (Helmut Schmidt University)
    Primary area of focus / application: Process
    Keywords: Statistical process control, Shewhart charts, CUSUM charts, Run length performance, Switching limit
    Submitted at 23-Mar-2018 15:59 by Sebastian Ottenstreuer
    Accepted
    4-Sep-2018 12:20 A Combined Shewhart-CUSUM Chart with Switching Limit
    The common Shewhart-CUSUM chart deploys an additional Shewhart limit to expand a single CUSUM chart by triggering quick alarms for large changes in the parameter of interest. Here, we utilize this supplementary limit to initiate the CUSUM accumulation. That is, we switch between an accumulation and a silent phase. The new switching limit's value resides between the reference value of the CUSUM chart and the usual Shewhart limit. Thus, for the case that the CUSUM statistic is equal to zero, a further observation has to be more substantial than this new limit to engage the summing process. We demonstrate the setup and analyze the new combination for independent Poisson distributed data and a more involved time series model with Poisson marginals, the Poisson INAR(1). Moreover, we also test the novel chart's robustness against hypothetical misspecification such as undetected overdispersion or autocorrelation. It turns out that this kind of combination features patterns between a pure CUSUM and a stand-alone Shewhart chart and, hence, constitutes a solid alternative to both single charts as well as to the ordinary Shewhart-CUSUM. Finally, in the context of possible extensions, a real data set from semiconductor industry with apparently overdispersed counts is considered and the application to Gaussian variables data is briefly discussed.
  • Nested Mutual Information Designs

    Authors: Mona Abtini (Ecole centrale de Lyon)
    Primary area of focus / application: Design and analysis of experiments
    Secondary area of focus / application: Design and analysis of experiments
    Keywords: Kriging, Space-filling-design, Mutual information criterion, Nested desings
    Submitted at 23-Mar-2018 16:04 by Mona ABTINI
    Accepted
    In recent years, computer simulation models are increasingly used to study complex phenomena. Such problems usually rely on very large sophisticated simulation codes that are very expensive in computing time (sometimes several days). The exploitation of these codes becomes a problem, especially when the objective requires a significant number of evaluations of the code. In practice, the code is replaced by global approximation models, often called metamodels, most commonly a Gaussian Process (kriging) adjusted to a design of experiments, i.e. on observations of the model
    output obtained on a small number of simulations. Space-Filling-Designs which have the design points evenly spread over the entire feasible input region, are the most used designs. However, if X_N is space-filling-design of N points, there is no guarantee that the n (1 <= n <= N) first points
    constitute a space-filling-design. In practice, however, we may have to stop the simulations before the
    full realization of design. The aim of this presentation is therefore to propose a new methodology of construction of sequential space-filling-designs (nested desings) of experiments X_n for any n between 1 and N that are all adapted to kriging prediction. We introduce a method to sequentially generate designs based on information criteria, particularly the "Mutual Information criterion (MI)". This
    method ensures the quality of the designs generated for all values of n, 1 <= n <= N. A key difficulty of this method is that the time needed to generate a MI-sequential design in high-dimension case is very lagre. To address this issue a particular implementation, which calculates the determinant of a given matrix by partitioning it into blocks. This implementation allows a significant reduction
    of the computational cost of MI-sequential designs, has been proposed.
  • Applied Linearity Analysis

    Authors: Thomas Wagner (Infineon Technologies)
    Primary area of focus / application: Metrology & measurement systems analysis
    Secondary area of focus / application: Quality
    Keywords: Gage study, Repeatability, Reproducibility, Linearity, Relative linearity, Decision matrix
    Submitted at 23-Mar-2018 16:30 by Thomas Wagner
    Accepted (view paper)
    5-Sep-2018 10:50 Applied Linearity Analysis
    Measurement system analysis is well established in the industry in order to measure, improve and ensure sufficient measurement quality. Mostly repeatability and reproducibility indexes are analyzed in regular studies. Additional linearity analysis studies are performed less frequently.

    With the introduced new approach a frequent linearity evaluation is possible by using the existing gage studies. The basic idea is to define the relative linearity rather than an absolute linearity. New key performance indicators are derived and the advantages highlighted in order to improve the measurement quality further more. Finally a decision matrix will be presented to handle the different measurement problems in practice.