Visualization of Multidimensional Design and Optimization Data Using Cloud Visualization

Abstract

Some of the first attempts at using visualization methods to aid decisions in design and optimization are found in [1]. More recent advances in computer visualization and Virtual Reality (VR) [2][3][4] are allowing designers and scientists to interact and manipulate vast amounts of data. Until these innovations, computers were solely relied on to interpret results and compute answers based on programs written. It is now possible to interact with these large datasets even while they are being used in running analyses [5][6][7][8][9][10][11]. Users have the ability to compress large amounts of data into a visual format, to investigate trends and relationships that could not be seen otherwise, and then make informed decisions regarding a product or process design.As our ability to generate more and more data for increasingly large engineering models improves, the need for methods for managing that data becomes greater. Information management from a decision-making perspective involves being able to capture and represent significant information to a designer so that they can make effective and efficient decisions. However, most visualization techniques used in engineering, such as graphs and charts, are limited to twodimensional representations and at most three-dimensional representations. In this paper, we present a new visualization technique to capture and represent engineering information in a multidimensional context. The new technique, Cloud Visualization, is based upon representing sets of points as clouds in both the design and performance spaces. The technique is applicable to both single and multiobjective optimization problems and the relevant issues with each type of problem are discussed. A multiobjective case study is presented to demonstrate the application and usefulness of the Cloud Visualization techniques.

A Work-Centered Visual Analytics Model to Support Engineering Design with Interactive Visualization and Data-Mining

“…However, these systems hardly provide in-depth data analysis capacities for better understanding of multi-dimensional, unorganized data sets. Cloud Visualization [13] and BrickViz [14] explore methods for grouping and aggregating data to reduce users' cognitive workload; however, neither approach is supported by formal clustering techniques. Chiu and his colleagues [15,16] provide a novel means for aggregating hyper-dimensional data and visualizing Pareto fronts that span n-dimensions and steering subsequent searches [17].…”

EVOLVE: A Visualization Tool for Multi-objective Optimization Featuring Linked View of Explanatory Variables and Objective Functions

“…1 is an example of the highlighting result after the selection of blue rectangular region of interest. This kind of linked view mechanisms have not been well applied to visualization of multi-objective optimization in the previous studies [3,4].…”

“…Eddy et al [4] presented "Cloud Visualization" which applies a scatterplot for visualization of Pareto solutions. Obayashi et al [5] presented a technique applying the Self Organization Map (SOM) to cluster and visualize Pareto solutions.…”

Virtual reality for assembly methods prototyping: a review

“…VR technology has evolved to a new level of sophistication during the last two decades. VR has changed the ways scientists and engineers look at computers for performing mathematical simulations, data visualization, and decision making [2][3][4][5]. VR technology combines multiple human-computer interfaces to provide various sensations (visual, haptic, auditory, etc.)…”