This use case aims to enable mass customization by creating an AI-enhanced value chain that connects consumer data back to product design, manufacturing, logistics, and production control. Data from connected products, such as shavers, is analyzed to better understand customer needs and translate them into customized product designs through MBSE, digital twins, ontologies, and AI-driven production services. The use case combines consumer analytics, recommendation engines, production optimization, agile supply-chain management, and production-data quality monitoring. Its main value is to improve customer satisfaction, product quality, manufacturing flexibility, time-to-market, and resource efficiency across the full value chain.
| # |
Technology Name |
Description |
| 1 |
Improved human motions for robot testing IPL devices |
Human motions are generated and validated for use in robot testing models. The work includes the use and translation of data from one skeleton model to another skeleton model, the generation of new data to train AI models, and the validation of the correctness of this generated data. |
| 2 |
Data analysis method |
A methodology is developed for analysing the production process data of OneBlade devices and proposing improvements. The method and conclusions are summarized in the confidential thesis by M. Smits, RUG: “Detecting Defect-Prone Paths in Manufacturing Processes Using Process and Pattern Mining – A case study of the Philips OneBlade Assembly Line.” |
| KPI # |
Related Technology |
KPI Target |
| 1 |
Improved human motions for robot testing IPL devices |
Efficiency and cost reduction by preventing extra human testing for each newly developed IPL device. Shorter time-to-market: 10–20% reduction. |
| 2 |
Data analysis method |
Improved product quality for 1 to 3 CTQs. Cost reduction: up to 10% less scrap due to bad quality. |
| KPI # |
Baseline |
| 1 |
Every robot test has to be validated by human testing. |
| 2 |
One CTQ is not capable, and scrap occurs due to bad quality. |
| KPI # |
Validation Method |
| 1 |
Evaluate the reliability of the robot testing. |
| 2 |
Measure the Cpk improvement and the scrap-level improvement. |
- Industrial process phases made more effective and precise through the integrated and tailored AI algorithms, methods, and tools.
- Improved overall productivity of the manufacturing factory by approximately 20%.
- Reduced manufacturing costs in the long term.
- Improved performance and faster time-to-market.
The use case contributes to more effective and precise industrial process phases through AI-supported robot testing and production process data analysis. The improved human-motion technology supports more reliable robot testing for IPL devices and reduces the need for additional human validation, which contributes to shorter time-to-market and lower testing effort. The data analysis method supports the identification of defect-prone paths in the OneBlade assembly process, helping to improve CTQ capability and reduce scrap caused by bad quality. Manufacturing cost reduction is therefore mainly connected to lower scrap levels, while productivity improvement is connected to more efficient production quality control and reduced process losses.
| AI Toolbox Tool |
Application in the Use Case |
| Human Pose Estimation Tool |
The Human Pose Estimation Tool was applied to represent and validate human skeletal motion data used for robot testing of IPL devices. It supported the analysis of human movements and the comparison of motion data across different skeleton models. |
| Motion Playground |
Motion Playground was used to model, visualize, and inspect motion data during the development of improved human motions for robot testing. It supported the preparation and validation of motion sequences used in AI-based test scenarios. |
| Tabular Prediction Tool |
The Tabular Prediction Tool was used to analyse structured OneBlade production process data and support the identification of process patterns related to quality issues. It contributed to quality-oriented process analysis and improvement recommendations. |
| Device-Level Quality Detection Tool |
The Device-Level Quality Detection Tool was applied to connect device-level quality problems with process- and machine-related features. It supported the analysis of defect-prone production paths and helped identify factors contributing to scrap and CTQ issues. |