ENBIS-14 in Linz21 – 25 September 2014; Johannes Kepler University, Linz, Austria Abstract submission: 23 January – 22 June 2014
The following abstracts have been accepted for this event:
Composite Indices based on Partial Least Squares
Authors: Jisu Yoon (Georg-August-Universität Göttingen), Tatyana Krivobokova (Georg-August-Universität Göttingen), Stephan Klasen (Georg-August-Universität Göttingen), Axel Dreher (Ruprecht-Karls-Universität Heidelberg)
Primary area of focus / application: Economics
Secondary area of focus / application: Modelling
Keywords: Composite indices, PLS, PCA, Non-Metric variables, Discrete variables
Submitted at 30-Apr-2014 12:15 by Jisu Yoon
Accepted (view paper)
Statistical Quality Monitoring of Profiles and Surfaces: Current Approaches and Future Challenges
Authors: Bianca Maria Colosimo (Politecnico di Milano)
Primary area of focus / application: Process
Secondary area of focus / application: Metrology & measurement systems analysis
Keywords: SPC, Profile monitoring, Surface monitoring, Quality, Gaussian process, Control chart, Data fusion
Submitted at 30-Apr-2014 12:20 by Bianca Maria Colosimo
On the other side, metrology is facing a paradigm shift where contact systems are substituted or coupled with contactless ones, sometimes asking for multisensor data fusion for surface reconstruction.
In this scenario, the paper describes possible solutions and current challenges when i) quality inspection requires statistical models for surface reconstruction starting from noisy point clouds; ii) statistical quality monitoring is applied to reconstructed surfaces via control charting.
Appropriate models combining Gaussian processes and multivariate control charting are shown and applied to real case studies.
Quality Indices based on Choquet Integrals
Authors: Konrad Waelder (BTU Cottbus-Senftenberg), Olga Waelder (BTU Cottbus-Senftenberg)
Primary area of focus / application: Quality
Secondary area of focus / application: Business
Keywords: Quality index, Interaction, Supplier management, Choquet integral, Fuzzy measure
Submitted at 30-Apr-2014 12:59 by Konrad Waelder
Even quality indices are often based on the aggregation of indices with respect to sub criteria, e.g. an index is obtained as arithmetic mean of sub indices.
In our talk we focus on quality indices in supplier management. Obviously, supplier management is a central part of any quality management system. Especially in the context of outsourcing or lean production assessments of suppliers are of great importance.
The final assessment of a supplier – maybe supplier A, supplier B and supplier C- is based on a weighted aggregation of such criteria like qualification level of employees, indices for product quality, financial indices, delivery based indices (e.g, adherence to delivery dates), etc.
From a mathematical and logical point of view simple aggregation ignores effects like necessity of fulfillment of criteria and interaction among criteria (redundancy, synergy).
Therefore, we present in this talk a general approach for aggregating sub criteria based on the so called Choquet integral, which is defined by a corresponding fuzzy measure modelling interaction effects. We introduce the Choquet integral as a general tool for dealing with multiple criteria decision making. After a theoretical exposition giving the fundamental basis with respect to fuzzy measures practical problems corresponding to the assessment of suppliers are addressed. In particular the problem of determining useful and suitable fuzzy measures is discussed.
We close our talk with a case study from supplier management.
Open Problem Session
Authors: Christian Ritter (Universtié Catholique de Louvain)
Primary area of focus / application: Consulting
Secondary area of focus / application: Education & Thinking
Keywords: Open problems, Consultancy, Live collaboration, Crowd solutions
Submitted at 30-Apr-2014 13:06 by Christian Ritter
I propose to host an open problem session (as I had done in at ENBIS2014 in Antwerp). I would need a 60-90 minute time slot and would issue a call for open problems in August. During the session, we could treat a maximum of three problems.
The way such a session runs is as follows:
Problems are either submitted before or during the conference. Time permitting, they can even be taken from the audience of the session (done in Antwerp). The person who volunteers the problem then has 10-12 minutes to explain. After that it goes by rounds of clarification and, hopefully, useful suggestions on how to treat the project.
Approximate Bayesian Computation Design with an Application to Spatial Extremes
Authors: Markus Hainy (Johannes Kepler University)
Primary area of focus / application: Design and analysis of experiments
Keywords: Simulation-based optimal design, Approximate Bayesian computation, Extreme value distribution, Max-Stable process
Submitted at 30-Apr-2014 14:12 by Markus Hainy
The methodology is applied to find the optimal design for a network of temperature measuring stations when the main interest is in inference for extreme events. ABC methods are applied because the likelihood function of the multivariate extreme value distribution (Schlather model) is not available in closed form for dimensions greater than two.
Practical Issues in Using Nonlinear Regression Analysis for Short-Term Prediction of High Water Levels
Authors: Alessandro Di Bucchianico (Eindhoven University of Technology), Krijn Saman (Dutch Ministry of Transport, Public Works and Water Management), Jan-Rolf Hendriks (Dutch Ministry of Transport, Public Works and Water Management)
Primary area of focus / application: Modelling
Secondary area of focus / application: Process
Keywords: Short-term prediction, Water level, Nonlinear regression, Conditional linearity, Variable projection algorithm, Starting values
Submitted at 30-Apr-2014 14:48 by Alessandro Di Bucchianico
protection system consisting of dikes and dams. An important part of the Delta Plan is the Easter Scheldt Storm Surge Barrier, which is a bridge-like
construction that can be closed when water levels become too high.
When the water level is above 2.75 m, a Decision Team is physically present at the barrier to close the barrier at the right moment. In order to keep
flooding risk at acceptable levels, the barrier automatically closes and takes over full control when there is a water level prediction of 3 m
exactly. In order to avoid environmental damage, the barrier is not allowed to close at predicted water levels below 3 m. It is thus important for
the Decision Team to avoid unnecessary closures of the barrier as well avoid that the barrier takes over full control. The Decision Team has access to
several long-term water level predictions. As an addition to these existing predictions, we are developing short-time prediction models. Part of these short-time prediction models is a nonlinear regression model with a quadratic polynomial to model the trend and various sine waves to model oscillations. Straightforward use of nonlinear regression to water levels measured every 10 seconds during the past 30 minutes surprisingly failed because of singular gradients. In this talk we will show the approaches that we are using to overcome the numerical problems, including scaling of the variables involved, using the variable projection algorithm of Golub and Pereyra to exploit that the model has conditionally linear parameters and the the systematic use of different starting point for the nonlinear regression.