Knowledge graphs in egg grading process – Sanovo
Overview
| Company Name / Department | Sanovo Technology Netherlands |
| Contact Person |
G. Heinen |
| Location | Aalten |
| Optional remote work | Yes, officially 3 office, 2 at home, but for this thesis project we could agree on another distribution |
| Travel expenses (own account or reimbursed by the company) |
Incl. in compensation for home to work travel Travel for the assignment to e.g. a customer is fully reimbursed |
| Housing arranged by company | No |
| Housing expenses (how much per month, own account or subsidized by the company) | No |
| Internship compensation | €500 per month |
| Study program | |
| ESCF community | |
| Start date | tbc |
Company Description
Sanovo is a global leader of developing equipment for processing and handling of eggs. In the Netherlands the development and production of the egg grading machines takes places. The main process of this machine is to grade and pack eggs based on characteristics like weight, shell color, and quality. From simple sensors to AI and deep learning is used to measure those characteristics.
Project Description
Sanovo Technology Netherlands builds machines that grade and pack eggs through a complex, multi-stage process. Occasionally, performance issues arise at specific packaging lanes (e.g., low throughput or increased errors), but the root cause may lie earlier in the process—such as the egg batch characteristics, machine settings, or mechanical conditions. Understanding these relationships requires insight across multiple interconnected data points.
Goals of the project
The goal of this project is to explore how a knowledge graph can be used to model the grading process, machine components, historical events, and sensor data to support root cause analysis. The student will define a graph-based model of the machine (and process), ingest sample data, and develop queries or visualizations that can help to trace problems back to their source. The goal is to demonstrate how knowledge graphs can support smarter diagnostics and continuous improvement.
Deliverables
- A written report documenting the approach, implementation, and findings, including the added value of using knowledge graphs in an industrial context.
- A working knowledge graph populated with example or historical data, demonstrating the structure and relationships in the process which illustrates how performance problems at the packaging lanes can be traced to potential root causes earlier in the process.
Knowledge
- A background in data modeling, graph databases, or industrial systems is recommended.
More information: escf@tue.nl
