Knowledge Graph-Based Matching of Product Requirements and Manufacturing Capabilities

Institut

Institut für Werkzeugmaschinen und Betriebswissenschaften (TUM-ED)

Typ

Bachelorarbeit / Semesterarbeit / Masterarbeit /

Inhalt

 

Beschreibung

Initial Situation

Automated manufacturing planning requires matching product-specific production requirements with the capabilities of available manufacturing resources. Conventional approaches frequently rely on predefined attributes and explicit matching rules, which can limit their ability to represent complex relationships between products, processes, and resources. Knowledge graphs offer an alternative for representing such relationships and supporting more flexible matching procedures. A learning factory with different product variants, predefined assembly operations, and available workstations provides a practical environment for investigating and validating knowledge graph-based matching approaches.

Objective

The objective of this work is to investigate the suitability of knowledge graphs for representing and matching product requirements and manufacturing capabilities using a learning factory as an application case. Existing approaches to ontology-based manufacturing knowledge representation, graph-based reasoning, and semantic matching will be analyzed. Based on these findings, a prototype knowledge graph will be developed to represent the products, required assembly operations, workstation capabilities, and relevant constraints of the learning factory. Different matching strategies will be implemented to identify suitable workstations for individual assembly operations. The proposed approach will be evaluated against conventional attribute-based matching using different products and assembly requirements. The evaluation will focus on matching accuracy, explainability, computational effort, and the ability to represent complex relationships between products, processes, and resources.

Voraussetzungen

Interest in knowledge representation, production engineering, and artificial intelligence. Good Python skills are advantageous. Experience with ontologies, graph databases, or semantic technologies is beneficial. Independent and scientifically structured working skills are expected.

Möglicher Beginn

sofort

Kontakt

Jannik Stöver
Raum: 1325
jannik.stoeveriwb.tum.de