Multivariate Adaptive Downsampling Algorithm for Industry 4.0 Data Visualization
Authors:
Date: 01.01.2022
Abstract
Following the Industry 4.0 paradigm, many industrial companies capture high volume of time series data from their industrial processes. A common task to generate value from this data is to visualize and analyze it. However, regular visualization approaches of this data require specialized hardware. Thus, downsampling algorithms as M4 are required to create a simplified view of the original data, which requires less computation power to be visualized while keeping as much information as possible from the original data. Although industrial processes involve synchronized variables that should be visualized together for their analysis, existing downsampling algorithms tackle visualization of univariate data series. Moreover, most of these algorithms require regularly sampled intervals, while data from many industrial processes does not fulfil this condition. This paper addresses these issues. The paper proposes an adaptive extension of the M4 algorithm suitable for multivariate datasets. The algorithm has been successfully tested with synthetic multivariate time series and commodity hardware, validating its suitability for the visualization and analysis of time series from industrial processes.
BIB_text
title = {Multivariate Adaptive Downsampling Algorithm for Industry 4.0 Data Visualization},
pages = {588-597},
keywds = {
Visual analytics downsampling M4 Multivariate time series
}
abstract = {
Following the Industry 4.0 paradigm, many industrial companies capture high volume of time series data from their industrial processes. A common task to generate value from this data is to visualize and analyze it. However, regular visualization approaches of this data require specialized hardware. Thus, downsampling algorithms as M4 are required to create a simplified view of the original data, which requires less computation power to be visualized while keeping as much information as possible from the original data. Although industrial processes involve synchronized variables that should be visualized together for their analysis, existing downsampling algorithms tackle visualization of univariate data series. Moreover, most of these algorithms require regularly sampled intervals, while data from many industrial processes does not fulfil this condition. This paper addresses these issues. The paper proposes an adaptive extension of the M4 algorithm suitable for multivariate datasets. The algorithm has been successfully tested with synthetic multivariate time series and commodity hardware, validating its suitability for the visualization and analysis of time series from industrial processes.
}
date = {2022-01-01},
}
