ENBIS Spring Meeting 2018

4 – 6 June 2018; Florence, Italy Abstract submission: 17 November 2017 – 20 April 2018

My abstracts

 

The following abstracts have been accepted for this event:

  • Optimizing an oxygen input profile to estimate Michaelis-Menten respiration parameters

    Authors: Arno Strouwen (Department of Biosystems, KU Leuven), Bart Nicolaï (Department of Biosystems, KU Leuven), Peter Goos (Department of Biosystems, KU Leuven)
    Primary area of focus / application: Design and analysis of experiments
    Secondary area of focus / application: Modelling
    Keywords: • Respiration kinetics, • Michaelis-Menten, • Optimal design of experiments, • Dynamic optimization
    Submitted at 28-Feb-2018 23:44 by Arno Strouwen
    Accepted
    4-Jun-2018 16:40 Optimizing an oxygen input profile to estimate Michaelis-Menten respiration parameters
    Fresh fruits and vegetables are perishable products and need to be stored at appropriate temperature, oxygen and carbon dioxide conditions, after their harvest. Traditionally, this is achieved by independently storing the product at many different combinations of temperature as well as O2 and CO2 partial pressures. The optimal storage conditions are then inferred from traditional response surface modeling. This method is labor intensive, due to the larger number of experimental combinations that have to be tested.

    Many modern fruit and vegetable storage applications, such as modified atmosphere packaging (MAP) and dynamic controlled atmosphere (DCA), rely on knowledge of mass balances, transport phenomena and reaction kinetics, and use comprehensive mathematical models that describe the behavior of the product as a dynamical system with inputs (temperature, oxygen and carbon dioxide partial pressures) and outputs (respiration and fermentation rate, quality attributes). A key feature of such dynamic models is the respiration kinetics, which is generally described by a non-linear model of the Michaelis-Menten type. The shift from traditional response surface modeling towards dynamic models for estimating respiration kinetics entails major challenges for designing experiments. For example, quantifying and optimizing the information content of experiments is numerically more complex. This is due to the fact that the dynamic approach involves differential equations and non-linear parameter estimation.

    In this presentation, we optimize a time varying oxygen input profile to estimate the respiration kinetics of pear fruit. To optimize the information content produced by this oxygen profile, we apply optimal dynamic experimental design principles and present a modified coordinate-exchange algorithm to achieve this goal. Finally, we compare the optimal input profiles to several benchmark approaches.
  • Sequential Multi-Aspect Monitoring Multivariate and High-Dimensional Data

    Authors: Amitava Mukherjee (XLRI -Xavier School of Management), Niladri Chakraborty (XLRI -Xavier School of Management), Marco Marozzi (University of Venice)
    Primary area of focus / application: Process
    Secondary area of focus / application: Process
    Keywords: High-dimensional Data, Multi-Aspect Monitoring, Statistical Process Control, Nonparametric Methods
    Submitted at 2-Mar-2018 11:14 by Amitava Mukherjee
    Accepted
    5-Jun-2018 09:50 Sequential Multi-Aspect Monitoring Multivariate and High-Dimensional Data
    In this paper, we propose a class of distribution-free sequential Phase-II monitoring schemes for multivariate and high-dimensional data. Instead of traditional monitoring of location shift, we consider monitoring multiple aspects of a multivariate population. We also indicate possible modification of the proposed procedure for high-dimensional data. Proposed techniques are based on the simple concepts of ranks, certain distance measures and the permutation tests. We study the performance of the proposed procedure in finite sample situations via Monte-Carlo. We also provide an illustrative example. Proposed method is expected to be very effective in monitoring high-dimensional business processes or multiple aspects related to an item quality.
  • Complex Design Requirements in the Food Industry - Cases and Implementations Using JMP

    Authors: Bart De Ketelaere (presenter) (KU Leuven), Volker Kraft (presenter) (SAS Institute / JMP Division), Wannes Akkermans (KU Leuven), Peter Goos (KU Leuven)
    Primary area of focus / application: Design and analysis of experiments
    Keywords: DOE, JMP, optimal design, mixture design, covariates
    Submitted at 6-Mar-2018 20:32 by Volker Kraft
    Accepted
    5-Jun-2018 14:15 Special Session: JMP Session
    New product development in the food industry often is a highly complex setting that might involve batch-to-batch variation, in addition to mixture variables, process variables and observable but uncontrollable factors. Designing experiments with such a complexity requires a tailored and flexible approach that is offered by so-called optimal designs. During this joint presentation, we will show typical examples from the food industry to set the scene. Next, we will show the possibilities offered by JMP to produce such optimal designs. Finally, we will elaborate on a practical case from a food company that wishes to optimize a new product. The envisaged experiment involves a constrained mixture of seven components in addition to six covariates measuring properties of the batches used. After presenting a plausible model for this case, we will create an optimal design using the unique features of JMP and analyse results obtained from it.
  • Complex Design Requirements in the Food Industry - Cases and Implementations Using JMP

    Authors: Bart De Ketelaere (presenter) (KU Leuven), Volker Kraft (presenter) (SAS Institute / JMP Division), Wannes Akkermans (KU Leuven), Peter Goos (KU Leuven)
    Primary area of focus / application: Design and analysis of experiments
    Keywords: DOE, JMP, optimal design, mixture design, covariates
    Submitted at 6-Mar-2018 20:32 by Volker Kraft
    Accepted
    5-Jun-2018 14:15 Complex Design Requirements in the Food Industry - Cases and Implementations Using JMP
    New product development in the food industry often is a highly complex setting that might involve batch-to-batch variation, in addition to mixture variables, process variables and observable but uncontrollable factors. Designing experiments with such a complexity requires a tailored and flexible approach that is offered by so-called optimal designs. During this joint presentation, we will show typical examples from the food industry to set the scene. Next, we will show the possibilities offered by JMP to produce such optimal designs. Finally, we will elaborate on a practical case from a food company that wishes to optimize a new product. The envisaged experiment involves a constrained mixture of seven components in addition to six covariates measuring properties of the batches used. After presenting a plausible model for this case, we will create an optimal design using the unique features of JMP and analyse results obtained from it.
  • AKM2D : An Adaptive Framework for Online Sensing and Anomaly Detection

    Authors: Kamran Paynabar (School of Industrial and Systems Engineering), Hao Yan (Sch Compt Infor & Dec Sys Engr, Arizona State University), Jianjun Shi (School of Industrial and Systems Engineering)
    Primary area of focus / application: Design and analysis of experiments
    Secondary area of focus / application: Quality
    Keywords: Adaptive Kernelized Maximum-Minimum Distance,, Exploration,, Exploitation,, sequential design of experiments.
    Submitted at 8-Apr-2018 02:44 by Kamran Paynabar
    Accepted
    5-Jun-2018 16:30 AKM2D : An Adaptive Framework for Online Sensing and Anomaly Detection
    In point-based sensing systems such as coordinate measuring machines (CMM) and laser ultrasonics where complete sensing is impractical due to the high sensing time and cost, adaptive sensing through a systematic exploration is vital for online inspection and anomaly detection. Most of existing sequential sampling methodologies focus on reducing the overall fitting error for the entire sampling space. However, in many anomaly detection applications, the main goal is to accurately detect and estimate sparse anomalous regions. In this paper, we develop a novel framework named Adaptive Kernelized Maximum-Minimum Distance (AKM2D) to speed up the inspection and anomaly detection process through an intelligent sequential sampling scheme integrated with fast estimation and detection. The proposed method balances the sampling efforts between the space filling sampling (exploration) and focused sampling near the anomalous region (exploitation). The proposed methodology is validated by conducting simulations and a case study of anomaly detection in composite sheets using a guided wave test.
  • Consumers’ Preferences About Coffee: A Choice Experiment Integrated With A Guided Tasting

    Authors: Nedka Dechkova Nikiforova (Department of Statistics, Computer Science, Applications ”G.Parenti”, University of Florence), Patrizia Pinelli (Department of Statistics, Computer Science, Applications ”G.Parenti”, University of Florence)
    Primary area of focus / application: Design and analysis of experiments
    Keywords: Sustainability,, Choice Experiments,, Optimal Design,, Antioxidants in coffee,, Organoleptic characteristics,, Random Utility Models
    Submitted at 11-Apr-2018 11:40 by Nedka Dechkova Nikiforova
    Accepted
    5-Jun-2018 11:55 Consumers’ Preferences About Coffee: A Choice Experiment Integrated With A Guided Tasting
    This talk suggests an innovative approach to analyze the consumers’ preferences for coffee consumption by integrating a choice experiment with consumers’ sensory test and chemical analysis (caffeine and antioxidants evaluated by a High Performance Liquid Chromatography-HPLC method). Firstly, two types of coffee (blend Arabica-Robusta and 100% Arabica) are chosen with different organoleptic characteristics, and a guided tasting session is planned through the development of two scorecards for the organoleptic evaluation. Moreover, a Choice Experiment based on optimal design theory is also planned in order to build choice-sets aiming to: i) an efficient estimation of the attributes for the choice experiment, and ii) detection of the effect of the sensory assessment scores obtained through the guided tasting. For this purpose, a compound design criterion (Wynn, 1970; Atkinson and Bogacka, 1997; Atkinson et al., 2007) is applied for addressing the issues described above. The same choice experiment is administered in two consecutive time occasions, e.g. before and after the guided tasting session, in order to assess the role of tasting in determining the consumers’ preferences. All these elements, e.g. the attributes involved in the choice experiment, the scores obtained for each coffee through the consumers’ sensory test and the HPLC results, are analyzed through Random Utility Models. The obtained results clearly indicate that the guided tasting jointly with the information provided on the two coffees, have a relevant impact on the consumers’ preferences, and contributes to unequivocally define them, by also allowing us to better understand the consumers’ behavior.

    REFERENCES:
    1) Wynn H. P. (1970). The sequential generation of D-optimal experimental designs. The Annals of Mathematical Statistics, 41:1055-1064.
    2) Atkinson, A. C. and Bogacka, B. (1997). Compound D- and D_S- Optimum Designs for Determining the Order of a Chemical Reaction. Technometrics, 39(4):347-356.
    3) Atkinson, A. C., Donev, A.N. and Tobias R.D. (2007). Optimum Experimental Designs, with SAS. Oxford: Oxford University Press.