University of Twente Student Theses

Login

Analysis for a possible implementation of a machine learning algorithm in the packaging process

Yoe, P.C.A (2018) Analysis for a possible implementation of a machine learning algorithm in the packaging process.

[img] PDF
1MB
Abstract:The technology of machine learning is based in the principle to recognise pattern and make decision according to the previous experiences. These algorithms are trained to be highly accurate in their prediction of the pattern. For this reason, many companies are utilising them. For examples, the medical field which can predict the flow of patients and the automotive industry uses machine learning algorithms for autonomous driving. The development of a machine learning algorithm is complicated. Most algorithm are trained on certain data features, which is the base for the algorithm to recognise the pattern. Though, these features are unknown and are often derived from human decision making. The degree of impact of features on the algorithm can be obtained by analysation and exploration. This research aims to explore the possibility to enhance digital product development in the field of computer-aided-design (CAD). The packaging team decides about the severity of clashes between CAD parts. This operation should be entrusted to the machine learning algorithm by discovering the right features.
Item Type:Essay (Master)
Faculty:ET: Engineering Technology
Subject:54 computer science
Programme:Industrial Design Engineering MSc (66955)
Link to this item:https://purl.utwente.nl/essays/75934
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page