Session Program

 

  • 11 July 2017
  • 08:00AM - 10:00AM
  • Room: Santa Lucia
  • Chairs: Luigi Troiano and S. Irene Díaz

Fuzziness in Data Science for Industrial and Manufacturing Applications

Abstract - Time series in finance are characterized by upside and downside movements that can be or not part of a larger trend. Trends become more obvious when we look at data points as granules and we study the relationship between them. In this paper we discuss which role granulation can have in describing the behavior of time series. In particular we investigate granules obtained by means of Ruspini partitions and we provide some examples of application.
Abstract - The media industry is increasingly personalizing the offering of contents in attempt to better target the audience. This requires to analyze the relationships that goes established between users and content they enjoy, looking at one side to the content characteristics and on the other to the user profile, in order to find the best match between the two. In this paper we suggest to build that relationship using the Dempster-Shafer's Theory of Evidence, proposing a reference model and illustrating its properties by means of a toy example. Finally we suggest possible applications of the model for tasks that are common in the modern media industry.
Abstract - We propose the application of triangular norms in the time-frequency domain, as a general framework for implementing consistency tests between the data gathered by different non collocated GW detectors, for denoising and signal isolation purposes
Abstract - This paper presents an Ensemble Data-Driven Fuzzy Network (EDDFN) for laser welding quality prediction that is composed of a number of strategically selected Data- Driven Fuzzy Models (DDFMs). Each model is trained by an Adaptive Negative Correlation Learning approach (ANCL). A monitoring system provides quality-relevant information of the laser beam spectrum and the geometry of the melt pool. This information is used by the proposed ensemble model to asist in the prediction of the welding quality. Each DDFM is based on three conceptual components, i.e. a selection procedure of the most representative welding information, a granular comprehesion process of data and the construction of a fuzzy reasoning mechanism as a series of Radial Basis Function Neural Networks (RBF-NNs). The proposed model aims at providing a fuzzy reasoning engine that is able to preserve a good balance between transparency and accuracy while improving its prediction properties. We apply the EDDFN to a real case study in manufacturing industry for the prediction of welding quality. The corresponding results confirm that the EDDFN provides better prediction properties compared to a single DDFM with an overal prediction performance $>$78\%.
Abstract - This paper presents ELIGERE, a new open-source distributed software platform for group decision making in engineering design. It is based on the fuzzy analytical hierarchy process (fuzzy AHP), a multiple criteria decision making method used in group selection processes to rank a discrete set of alternatives with respect to some evaluation criteria. ELIGERE is built following the paradigm of distributed cyber-physical systems. It provides several features of interest in group decision making problems: a web-application where experts express their opinion on the alternatives using the natural language, a fuzzy AHP calculation module for transforming qualitative into quantitative data, a database for collecting both the experts' answers and the results of the calculations. The resulting software platform is: distributed, interactive, multi-platform, multi- language and open-source. ELIGERE is a flexible cyber-physical information system useful in various multiple criteria decision making problems: in this paper we highlight its key concepts and illustrate its potential through a case study, i.e., the optimum selection of design alternatives in a robotic product design.
Abstract - An optimal fuzzy disturbance observer-enhanced sliding mode controller (FDO- SMC) for magneto-rheological damper (MRD)-based semi-active train-car suspensions (MRD-TSs) subjected uncertainty and disturbance (UAD) whose variability rate may be high but bounded is proposed. The two main parts of the FDO-SMC are an adaptive sliding mode controller (aSMC) and an optimal fuzzy disturbance observer (oFDO). First, initial structures of the sliding mode controller (SMC) and disturbance observer (DO) are built. Adaptive update laws for the SMC and DO are then set up synchronously via Lyapunov stability analysis with a used parameter constraint mechanism. An optimal fuzzy system (oFS) is designed to implement fully the constraint mechanism so as to guarantee for the stable converging to the desired state even if the UAD variability rate increases in a given range. As a result, the aSMC and the oFDO are created from the SMC and DO. The compared simulation surveys reflected that the positive competence to stamp out and isolate vibration with the lower consumed power is the main advantage of the proposed controller.