Please use this identifier to cite or link to this item: http://hdl.handle.net/10125/79562

Decision Analytics in Practice: Improving Data Analytics in Pulsed Power Environments Through Diagnostic and Subsystem Clustering

File Size Format  
0182.pdf 306.34 kB Adobe PDF View/Open

Item Summary

Title:Decision Analytics in Practice: Improving Data Analytics in Pulsed Power Environments Through Diagnostic and Subsystem Clustering
Authors:Yu, Andy
Keywords:Practitioner Research Insights: Applications of Science and Technology to Real-World Innovations
clustering
experiments
pulsed power
systems
Date Issued:04 Jan 2022
Abstract:Modern day processes depend heavily on data-driven techniques that use large datasets clustered into relevant groups help them achieve higher efficiency, better utilization of the operation, and improved decision making. However, building these datasets and clustering by similar products is challenging in research environments that produce many novel and highly complex low-volume technologies. In this work, the author develops an algorithm that calculates the similarity between multiple low-volume products from a research environment using a real-world data set. The algorithm is applied to pulse power operations data, which routinely performs novel experiments for inertial confinement fusion, radiation effects, and nuclear stockpile stewardship. The author shows that the algorithm is successful in calculating similarity between experiments of varying complexity such that comparable shots can be used for further analysis. Furthermore, it has been able to identify experiments not traditionally seen as identical.
Pages/Duration:4 pages
URI:http://hdl.handle.net/10125/79562
ISBN:978-0-9981331-5-7
DOI:10.24251/HICSS.2022.230
Rights:Attribution-NonCommercial-NoDerivatives 4.0 International
https://creativecommons.org/licenses/by-nc-nd/4.0/
Appears in Collections: Practitioner Research Insights: Applications of Science and Technology to Real-World Innovations


Please email libraryada-l@lists.hawaii.edu if you need this content in ADA-compliant format.

This item is licensed under a Creative Commons License Creative Commons