ENBIS-16 in Sheffield

11 – 15 September 2016; Sheffield Abstract submission: 20 March – 4 July 2016

My abstracts

 

The following abstracts have been accepted for this event:

  • Flipping the Classroom of a DoE Course

    Authors: Erik Vanhatalo (Luleå University of Technology)
    Primary area of focus / application: Education & Thinking
    Secondary area of focus / application: Design and analysis of experiments
    Keywords: Teaching statistics, Course design, Design and Analysis of Experiments, Higher education
    Submitted at 30-May-2016 18:03 by Erik Vanhatalo
    Accepted
    12-Sep-2016 10:20 Flipping the Classroom of a DoE Course
    The purpose of the presentation is to illustrate how an advanced design and analysis of experiments (DoE) course at Luleå University of Technology, Sweden, was ‘flipped’ to increase learning with less teacher workload. Historically the course used a lot of lectures and many teaching hours in the classroom. The presentation will cover motivation of the work, course design and implementation, student assignments and home experiments, analysis of student satisfaction, advantages, drawbacks and future development needs.

    The flipping of this particular course meant that blended learning was used to deliver instructional content. Students are encouraged to read course literature, watch recorded lectures (delivered e.g. on YouTube), work in pairs with pre-designed computer assignments and then come with their questions to topic-driven seminars where teacher(s) will discuss and answer questions. The course also features a two-stage laboratory assignment using simulation software where pairs of students learn sequential experimentation and response surface methodology. Students also work in groups of four to design, perform and analyze home experiments where the planning phase is given special attention. Students present their home experiments in the end of the course and to increase the ‘fun factor’ students are encouraged to make shorter videos as part of their presentations.

    The main work with ‘flipping’ the classroom was made in 2015 and two rounds of the course have now been completed with high student satisfaction. The course is given to roughly 20 students with a majority following a master’s program in industrial and management engineering.
  • The 16 Run Nonregular Two-Level Designs - Properties and Analysis

    Authors: John Tyssedal (The Norwegian University of Science and Technology)
    Primary area of focus / application: Design and analysis of experiments
    Keywords: Nonregular designs, Projection properties, Aliasing, Analysis
    Submitted at 31-May-2016 08:40 by John Tyssedal
    Accepted (view paper)
    13-Sep-2016 09:00 The 16 Run Nonregular Two-Level Designs - Properties and Analysis
    Among the two-level nonregular designs the 12 run and the 20 run Plackett-Burman designs seem to be the ones that are used the most. The two-level nonregular 16 run designs seem to be forgotten even though they have attractive projection properties and, when the number of experimental factors is above 6, represent important alternatives to the regular designs. For various numbers of experimental factors these designs have been classified according to criteria like generalized minimum aberration and minimum aliasing. However they also have some hidden aliasing structure, that have not been pointed out so far, and which have to be taken into account when their analysis is performed. In this presentation we will focus on this hidden alias structure and ways to analyze these designs.
  • Statistical Approach for Knowledge Improvement in Waterjet Machining of High Performance Ceramics

    Authors: Flaviana Tagliaferri (Fraunhofer JL IDEAS-SQUARE, Department of Industrial Engineering, University of Naples Federico II), Biagio Palumbo (Fraunhofer JL IDEAS-SQUARE, Department of Industrial Engineering, University of Naples Federico II), Matthias Putz (Fraunhofer Institute for Machine Tools and Forming Technology IWU (Chemnitz)), Markus Dittrich (Chemnitz University of Technology, Institute for Machine Tools and Production Processes IWP), Martin Dix (Chemnitz University of Technology, Institute for Machine Tools and Production Processes IWP)
    Primary area of focus / application: Design and analysis of experiments
    Secondary area of focus / application: Quality
    Keywords: Design of Experiment (DoE), Analysis Of Variance (ANOVA), Waterjet machining, High performance ceramics machining
    Submitted at 31-May-2016 13:29 by Biagio Palumbo
    Accepted
    13-Sep-2016 09:00 Statistical Approach for Knowledge Improvement in Waterjet Machining of High Performance Ceramics
    This work highlights the strategic role that a systematic and sequential approach to experimentation plays in order to get competitive advantage, technological innovation and knowledge improvement.
    An accurate pre-design (i.e. pre-experimental planning phase) is the solid basis on which a statistical approach has to be built. The investigation was based on the use of customized Pre-Design Guide Sheets. The documents provide a way to systematize the experimental planning. In fact, the pre-design guide sheets drive the research team to clearly define the objectives and scope of an experiment and to gather information needed to design it.
    The efficacy of this approach is demonstrated here by an applicative example about the Waterjet Machining of High Performance Ceramics, developed at Chemnitz University of Technology.
    The aim was to achieve a high material removal rate at an adequate surface quality. The experimental design was shared in two phases: screening and optimization phase. In the screening experimental phase a 25-1 design was adopted. In the optimization phase 6 central points and 10 axial (or star) points were added to the original design. The Face Centered (α=1) Central Composite Design (CCD) was used for fitting second order response surface regression model. The Response Surface Methodology (RSM) allowed to optimize the response variables.
  • A Location-Dispersion Index for the Analysis of Lower and Upper Bounded Variables

    Authors: Stefano Barone (University of Palermo)
    Primary area of focus / application: Modelling
    Keywords: Gini index, Concentration, Robustness, Location-dispersion index
    Submitted at 31-May-2016 13:39 by Stefano Barone
    Accepted
    13-Sep-2016 14:30 A Location-Dispersion Index for the Analysis of Lower and Upper Bounded Variables
    The concept of concentration in statistics is well known and consolidated. Concentration is a way to quantify how a generic resource is allocated between a number of subjects.
    It is well motivated in the economics field: the more resources lie in the hands of few, the more concentration is present.
    The most adopted indicator for concentration is the so called Gini index.
    The concept of concentration is quite far away from the concept of robustness, this latter having to do with the two aspects of location and dispersion taken together, and meaning closeness to a target value with smallest possible variation.
    None has so far felt the need to relate the concept of concentration to the concept of robustness.
    However, it is possible to show that in some special circumstances the two concepts become equivalent and a well-studied modification of the Gini index is a robustness index (hereby called Location-Dispersion Index, LDI).
    The “special” circumstances are actually not so rare in statistical analysis since they refer to cases in which the characteristic of interest in the statistical units is theoretically continue (even though it can be measured on a discrete scale) and lower- and upper-bounded. An example can be the “charge of a cellphone battery”.
    The LDI was previously formulated in another context, where the measurement of the characteristic of interest (the satisfaction) was made on a discrete scale (0,1,2,…,10 scale).
    In this work the author presents the LDI in the most general framework and illustrates some interesting properties.
    Insights on the collocation of the index in the current literature, and suggestions of possible exploitation and application fields will complete the work.
  • Simulating Experiments in Closed-Loop Control Systems

    Authors: Francesca Capaci (Luleå University of Technology), Erik Vanhatalo (Luleå University of Technology), Murat Kulahci (Luleå University of Technology), Bjarne Bergquist (Luleå University of Technology)
    Primary area of focus / application: Design and analysis of experiments
    Keywords: Design of Experiments, Closed-loop systems, Simulation, Tennessee Eastman process
    Submitted at 31-May-2016 14:50 by Francesca Capaci
    Accepted (view paper)
    13-Sep-2016 15:40 Simulating Experiments in Closed-Loop Control Systems
    Design of Experiments (DoE) literature extensively discusses how to properly plan, conduct and analyze experiments for process and product improvement. However, it is typically assumed that the experiments are run on processes operating in open-loop: the changes in experimental factors are directly visible in process responses and are not hidden by (automatic) feedback control. Under this assumption, DoE methods have been successfully applied in process industries such as chemical, pharmaceutical and biological industries.

    However, the increasing instrumentation, automation and interconnectedness are changing how the processes are run. Processes often involve engineering process control as in the case of closed-loop systems. The closed-loop environment adds complexity to experimentation and analysis since the experimenter must account for the control actions that may aim to keep a response variable at its set-point value. The common approach to experimental design and analysis will likely need adjustments in the presence of closed-loop controls. Careful consideration is for instance needed when the experimental factors are chosen. Moreover, the impact of the experimental factors may not be directly visible as changes in the response variables (Hild, Sanders, & Cooper, 2001). Instead other variables may need to be used as proxies for the intended response variable(s).

    The purpose of this presentation is to illustrate how experiments in closed-loop system can be planned and analyzed. A case study based on the Tennessee Eastman Process simulator run with a decentralized feedback control strategy (Matlab) (Lawrence Ricker, 1996) is discussed and presented.
  • Multicriterial Optimization of Real-World Applications and Computer Experiments

    Authors: Nikolaus Rudak (Dortmund University of Applied Sciences), Sonja Kuhnt (Dortmund University of Applied Sciences)
    Primary area of focus / application: Design and analysis of experiments
    Keywords: Multicriterial, Optimization, Regression, Kriging, Meta-model
    Submitted at 31-May-2016 14:58 by Nikolaus Rudak
    Accepted (view paper)
    13-Sep-2016 09:40 Multicriterial Optimization of Real-World Applications and Computer Experiments
    In real-world technical applications, as for example in a thermal spraying process, the quality of a product, represented by multiple properties, is usually affected by several machine parameters und some unaccounted noise. In order to understand the relationship between machine parameters and the properties of a product, the experimenter can conduct an appropriate experimental design and then fit a regression model for the unknown mean and variance of each property. Afterwards, one can find machine parameters, where the fitted mean of each product property is near a desired target value with minimal variance. We review some methods to tackle this problem, as for example by means of the JOP method (Rudak et al., 2015).
    If available, computer simulations can emulate costly and complex real-world experiments. In turn, these computer experiments can be very time consuming, e.g. for the finite element method. Therefore, usually an easy to evaluate surrogate meta-model is build based on computer experiments. This meta-model might then be used for optimization with the well-known Efficient Global Optimization (EGO) algorithm. However, if further constraints have to be respected, where the constraint function is also an output of the computer simulation, one needs a modification of the EGO algorithm. We review some methods for constrained optimization of computer experiments, e.g. Durantin et al., 2016, and present results from an application in the field of radial impeller optimization.