aalto1 untyped-item.component.html

Fault prediction and visualization

Loading...
Thumbnail Image

URL

Journal Title

Journal ISSN

Volume Title

Perustieteiden korkeakoulu | Master's thesis

Department

Mcode

SCI3020

Language

en

Pages

31+2

Series

Abstract

This project is designed to develop a fault prediction tool for the planning team to allocate resources better and for managers to control product quality. I have proposed and developed an interactive data visualisation prototype on Power BI to analyse data about fault prediction for 5G hardware. Compared to the existing solution with scattered information distributed on different platforms, the proposed prototype can integrate all the content in Power BI. Many manual data collection tasks can now be done automatically and can sync regularly. The prototype could collect and analyse historical fault data and make future predictions. The current application applies all the algorithms in Excel, while the proposed prototype writes them in Data Analysis Expression (DAX) in a more organised way. At the same time, a user-friendly interface with flexible model configuration is designed for users with different levels of expertise. Additionally, the evaluation of the final prototype was conducted in the form of heuristic evaluation and user-based evaluation. Critical insights were gathered during the process, which gave excellent guidance for future development.

Description

Supervisor

Fdili-Alaoui, Sarah

Thesis advisor

Chen, Tao

Other note

Citation

Endorsement

Review

Supplemented By

Referenced By