Systematic Analysis of Engineering Change Request Data

Ívar Örn Arnarsson​, Doctoral student at Product Development IMS, defends his doctoral thesis "Systematic Analysis of Engineering Change Request Data - Applying Data Mining Tools to Gain New Fact-Based Insights​".



Ívar will defend the thesis (online) on May 29, 13:00. A popular science summary is given below. For more information, see the links at the bottom of the page.

O​​pponent: Christopher McMahon, Technical University of Denmark​
Examiner: Johan Malmqvist, IMS




Systematic Analysis of Engineering Change Request Data - Applying Data Mining Tools to Gain New Fact-Based Insights

Engineering changes are common in industry as they are opportunities to improve, enhance, or adapt a product. They driver for a change can be e.g. related to quality, safety, changes in external circumstances or regulation. These engineering changes often referred as Engineering Change Requests (ECRs) are largely generated through product development projects and are often stored in database while worked and later for some form of knowledge management purpose. 
Despite ECR being captured and stored it is often cumbersome for product developers to identify historical ECRs due to the vast amount of them. Historical ECRs might contain valuable knowledge relevant to a current design and it is often wondered if the ECR content might be analyzed in a new way insightful way. The content of ECR data must contain information permitting identification of the types of errors and changes made, including part title, part name, part number, problem description, root cause, solution and test results. 
This thesis primarily focuses on ECR data in combination with three components necessary to perform data mining and data analytics: exploring and collecting ECR data, collecting domain knowledge about ECR information needs, and applying mathematical tools for solution design and testing. 
Results show a list of engineering information needs related to ECRs, examples of visualizations based on unstructured data, industrial case study where complex product development processes are modeled using the Markov chain Design Structure Matrix, and studies that investigate how advanced searches based on natural language processing techniques and clustering within engineering databases.

Read more
​Ívar Örn Arnarsson on Linkedin​

Published: Tue 06 Oct 2020.