)
)
)
)
)
)
)
How can we help you?
)
Marija Mijic
HR & Office Allrounder
T +43 660 383 95 46
office@craftworks.at
craftworks GmbH
Schottenfeldgasse 20/6a
1070 Vienna
Predictive Quality for
High Standard Materials
Ensured durability and product quality through anomaly detection.
:quality(80))
Success Story
reading time: 2 min
Manufacturing
Material Industries
Predictive Quality
Anomaly Detection
Industrial Software
67%
Reduction in human error
Certified
Product quality
Leveraging
Real-time quality data
Context
The industrial group that we were serving is a long-term leading chain manufacturer for solutions for forestry, lifting, snow chains, and others, an industry with very high standards for the quality and durability of the products. In order to achieve those standards, the manufacturer needed a solution for analyzing process data, identifying anomalies and establishing correlations between certain processes and errors.
Challenge
The challenge entailed the perfect alignment of process and quality data, since a lot of operations were not monitored, hence the data input fell short. The missing information made finding the right solution very difficult.
Assignment
craftworks was brought on board to tackle the challenge and provide a solution for finding anomalies in the operational data and, additionally align it with quality data.
Solution
Our solution utilized technologies like Python DS, PyTorch, and Fastai to detect unsupervised anomalies within the process data.
The unsupervised anomalies were compared with quality data, evaluated, and ultimately aligned with pertinent information.
:quality(80))