ENBIS-17 in Naples

9 – 14 September 2017; Naples (Italy) Abstract submission: 21 November 2016 – 10 May 2017

My abstracts


The following abstracts have been accepted for this event:

  • Examining Potential Reductions in Wind Tunnel Testing Data Requirements

    Authors: Raymond Hill (Air Force Institute of Technology), Douglas A. Dillard (Air Force Institute of Technology), Darryl K. Ahner (Air Force Institute of Technology), Douglas C. Montgomery (Arizona State University)
    Primary area of focus / application: Other: US Session and Tribute to Stephen Fienberg
    Keywords: Experimentation, Regression, RSM, Modeling
    Submitted at 25-Apr-2017 18:08 by Raymond Hill
    11-Sep-2017 12:00 Examining Potential Reductions in Wind Tunnel Testing Data Requirements
    This research explores the application of Design of Experiments (DOE) techniques in routine wind tunnel testing to reduce overall data requirements and demonstrates that significant reduction without information loss is possible. In addition, this research also shows there is a limit to how small a data set can be before a DOE design fails to remain statistically equivalent to a large data matrix created via the OFAT method. Practical applications of DOE to wind tunnel testing have significantly decreased wind-on minutes and total data volume compared to traditional tests. Thus, the DOE process can be used throughout the development of a major flight system to conserve resources. In addition, the research presents information loss differences between four distinctly different DOE designs; Covering Array, Nested Face Centered Design, I-Optimal Design, and Latin Hypercube. The information loss due to three different small sample sizes is quantified for a legacy wind tunnel test provided by Arnold Engineering Development Center, the sponsoring organization.
  • Control Charts on Dependent Streams

    Authors: Moshe Pollak (The Hebrew University of Jerusalem and Israel Central Bureau of Statistics)
    Primary area of focus / application: Process
    Secondary area of focus / application: Quality
    Keywords: Cusum, Shiryaev-Roberts, Average run length to false alarm, Exponential distribution
    Submitted at 28-Apr-2017 11:00 by moshe pollak
    Accepted (view paper)
    12-Sep-2017 19:00 Control Charts on Dependent Streams
    Consider a process that produces a series of independent vectors. A change in an underlying state may become manifest in a modification of one or more of the marginal distributions. Often, the dependence structure between coordinates is unknown, impeding surveillance based on the joint distribution. A popular approach is to construct control charts for each coordinate separately and raise an alarm the first time any (or some) of the control charts signals. The difficulty is obtaining an expression for the overall average run length to false alarm (ARL2FA).

    We argue that despite the dependence structure, when the process is in control, for large ARL's to false alarm, run lengths of parallel Cusum-type control charts are asymptotically independent. Furthermore, often, in-control run lengths are asymptotically exponentially distributed, enabling uncomplicated asymptotic expressions for the ARL2FA. We prove this assertion for certain Cusum- and Shiryaev-Roberts-type control charts and illustrate it by simulations.
  • Multiplicity-Cause Economic and Economic Statistical Design of T2-Control Charts Under Proportional Hazards Shock Model

    Authors: Mojtaba Aghajanpoor Pasha (Allameh Tabataba'i University), Mohammad Bameni Moghadam (Allameh Tabataba'i University), Nader Nematollahi (Allameh Tabataba'i University)
    Primary area of focus / application: Quality
    Secondary area of focus / application: Reliability
    Keywords: Multiple assignable causes, Increasing failure rate, Proportional hazards shock model, Expected cost per unit time, Economic statistical design
    Submitted at 1-May-2017 09:34 by Mojtaba Aghajanpoor Pasha
    13-Sep-2017 10:30 Multiplicity-Cause Economic and Economic Statistical Design of T2-Control Charts Under Proportional Hazards Shock Model
    Multivariate quality control of the producing processes is more crucial rather than considering a single characteristic from customer's point of view. On the other hand, the assumption of multiple assignable causes in economic design of control charts is more reasonable than only a single assignable cause from manufacturer's point of view. ‎The multiplicity-cause economic models of Duncan (1971) and Chen and Yang (2002) are extended in this paper ‎‎‎‎‎‎‎by employing the proportional hazards model with an arbitrary baseline lifetime distribution as the shock model of a deteriorating process with multivariate quality characteristics. ‎For illustration, a numerical study assuming the lifetime distribution in Chen (2000) with increasing failure rate that belong to the family of proportional hazards model is used to obtain the economic and economic statistical design parameters of T2-control charts, ‎i.e.‎, ‎the sample size, ‎the sampling intervals, ‎and the width coefficient of control limits. ‎Furthermore, ‎comparisons between a multiplicity-cause model and a single one are performed under the same time and cost parameters of the model.
  • Benchmarking Rater Agreement Indices: Statistical Properties and Power Analysis

    Authors: Amalia Vanacore (University of Naples Federico II), Maria Sole Pellegrino (University of Naples Federico II)
    Primary area of focus / application: Other:
    Keywords: Inter-rater agreement, Intra-rater agreement, Kappa-type indices, Monte Carlo simulation
    Submitted at 3-May-2017 11:50 by Amalia Vanacore
    12-Sep-2017 16:20 Benchmarking Rater Agreement Indices: Statistical Properties and Power Analysis
    This paper presents a critical review of some kappa-type indices proposed in the literature to measure the degree of rater agreement. Single measures of agreement provide only limited information and do not account for statistical uncertainty thus, following recommended guidelines for reporting agreement studies, we will present agreement indices including their confidence intervals. The magnitude of each estimated agreement coefficient will be related to the notion of extent of agreement by comparing the lower limit of its confidence interval against a benchmark scale. Specifically, we will explore the case of agreement among series of ratings referring to n items classified into k ordered categories by different raters (i.e., inter-rater agreement) or, equivalently, by the same rater in different occasions (i.e., intra-rater agreement).
    The reviewed indices are Gwet’s AC1 and the linear weighted variants of Scott’s Pi coefficient, Cohen’s Kappa and Brennan-Prediger statistic. In order to evaluate the statistical behavior of the reviewed indices and of a non-parametric benchmarking procedure, a Monte Carlo simulation study has been conducted for several scenarios differing from each other in sample size, rating scale dimension and agreement level. The estimate precision is evaluated in terms of relative bias, variance and coverage rate of the percentile bootstrap confidence interval, whereas the effectiveness of the benchmarking procedure is assessed in terms of statistical power.
    Simulation results suggest that the analyzed indices have satisfactory estimate precision that improves as n, k and agreement level increase and a coverage rate close to its nominal level, only for n ≥ 30; the benchmarking procedure is generally adequately powered in testing null and non-null cases of rater agreement and thus it can be suitably applied for the characterization of agreement over a small or moderate number of subjective ratings provided by one or more raters.
  • Kriging Modelization in Predicting Metal Sheet Elongation

    Authors: Valentina Calì (Università degli studi di Torino), Maria Teresa Giraudo (Università degli studi di Torino), Roberto Sofia (Amada Engineering Europe), Grazia Vicario (Politecnico di Torino)
    Primary area of focus / application: Design and analysis of experiments
    Keywords: LH designs, Kriging, Variogram, FEM, Bend deduction
    Submitted at 4-May-2017 16:16 by Valentina Calì
    11-Sep-2017 17:50 Kriging Modelization in Predicting Metal Sheet Elongation
    Modern industries increasingly replace real experiments with non-stochastic simulation models for their restrained costs and increasing reliability. The non-stochastic simulator used in this paper is the Finite Element Simulation code (FEM), a widely used numerical technique for treatment of engineering problems modelled by a system of partial differential equations defined on a time-space domain. In such a context, it is common practice to provide a metamodel, a global approximation of the FEM experiment response on the design space to capture local minima/maxima.
    We resort to the most popular metamodel, the Kriging model, applied to an industrial instance: prediction of the Bend Deduction. Metal sheet bending is a manufacturing process in which a plastic deformation of the work pieces over an axis occurs. This is a metal forming process, and similar to the other processes, bending changes the shape of the work pieces.
    The work focuses on the construction of an optimal initial design in order to achieve a good accuracy of the metamodel at an acceptable computational cost, on the theoretical study of this model and on understanding how it could be conformed to the bend deduction prediction.
    The correlation structure, mandatory in a Kriging model, has been evaluated by means of the variogram, whose refinements of its specification naturally improve the Kriging predictions. The empirical variograms for each input variable brought to light unusual behaviors. This peculiarity suggested that the bending angle could be related with the bend deduction according to two different models with a discontinuity certain in their relationship but uncertain where it is.
    Then the accuracy achieved has been evaluated using different indicators of the robustness and of the uncertainty of the leave-one-out methods.
  • Electroluminescence Image Analysis and Suspicious Areas Detection

    Authors: Evgenii Sovetkin (RWTH-Aachen), Ansgar Steland (RWTH-Aachen)
    Primary area of focus / application: Metrology & measurement systems analysis
    Secondary area of focus / application: Quality
    Keywords: Image processing, Nonparametrics, Change-point, Spatial statistics
    Submitted at 6-May-2017 11:35 by Evgenii Sovetkin
    11-Sep-2017 12:20 Electroluminescence Image Analysis and Suspicious Areas Detection
    In this work we consider several problems arising in quality control analysis of electroluminescence (EL) images of photovoltaics (PV) modules. The EL image technique is a useful tool for investigating the state of a PV module and allows us to look inside a module and to analyse the crystalline structure at high resolution. However, there is a lack of methods to employ the information provided by EL images in the analysis of large PV systems.
    We first consider several practical issues that arise in field studies, i.e.\ when images are taken under outdoor conditions and not in a lab. We discuss a new problem-specific procedure for automatic correction of rotation and perspective distortions, which to some extent employs statistical approaches such as robust regression; and a procedure for automatic detection of the module and its cell areas (by means of a modified version of the Hough Transform). Those techniques provide us with images of the PV module cells, intensity light of which are of the main interest in quality study.
    Secondly, we discuss a spatial test to screen large databases of EL image data aiming at the detection of malfunctioning cells. The spatial test statistics is based on comparing sample averages inside two regions indexed by a region location parameter. The asymptotics is established for a general class of random fields for several of regions sets.
    Lastly, we discuss simulation studies and an application of the method to the real EL image data.