PHILIPS has recently developed a new male grooming device called OneBlade. The OneBlade is undergoing a phased worldwide introduction, started in 2016. PHILIPS is currently developing new production lines with increased capacity to meet the increased market demand. To remain competitive, there is a clear need to improve PHILIPS productivity by improving three main metrics: Time to market, Production costs and Product/component quality. An increase of component quality will have a positive effect on production costs and time to market.
The production line in scope at Philips is producing the cutting element for the OneBlade. Processes used for this are:
- Over moulding
- μAssembly of parts
- μWelding/joining of parts
- Pad Printing
- In-line testing
The main goal of QU4LITY for PHILIPS is to realize a holistic system that:
- can raise early warning signals based on early indicators and trends from process signals and dimensional CTQs that are still acceptable on component level but will lead to Fall of Rate on the finished good in the current quality framework
- can suggest feed-forward or feed-backward controls to neighbouring process stations, which might have an influence on the dimensional CTQ that is under observation
- increases OEE A and P by helping the operator to take correct process adjustment decisions,
- reduces Fall Off Rate (FOR) by learning from unknown data interactions
With these elements the next step in autonomous quality/production can be made: from 'descriptive' to 'prescriptive' quality assurance
| Country: | NL |
| Address: | Oliemolenstraat 5, Drachten 9203 |
Deviation on cycle-time, from 98% (as is) to 99% (to be)
Sintef delivered information to put “soft” part of organization also in daily management structure. Nowadays the KPI’s are hard technical related. Other topic is to use digital tools for operator- whiteboard sessions. People claim they are more digital oriented at home compared to work floor.
Stakeholder-training Logbook: No results obtained yet, as the implementation is not far enough to train stakeholders.
Improve OEE from 75% (as is) to 85% (to be)
Deviation on cycle-time, from 98% (as is) to 99% (to be)
There seems to be relationship to predict torque with use of in-line data. Needs to be more explored
An AI vision algorithm developed by TNO (WP3) seems to filter bad rated parts compared to installed algorithm. Advantage can be when product print is changing to catch-up development speed in traditional algorithm development.
Enable operators to work in a more complex environment while reducing the strain of administrative tasks and enabling easy production analytics by capturing information online instead of on paper.
Shopfloor worker (operator – technical support group): From a shopfloor perspective new job profiles, or altered job profiles should be defined, however In essence the job profiles will remain the same, while the operators and Technical Support Groups need to understand & be able to work with these new technologies. This requires some basic knowledge on the (digitalized) systems, for the operators a lot can be captured in SOP’s (Standard Operating Procedures), but the technical support staff should also have some basic knowledge on the workings and the hardware/software side of the systems in order to be able to support the shopfloor where needed.
There was a risk that other developments made within this pilot do not follow the reference architecture of IDS and thus are incompatible. This would cause that certain applications could not be deployed and run within in the proposed data space approach.
AI vision algorithm developed by TNO (WP3) seems to filter bad rated parts compared to installed algorithm. Advantage can be when product print is changing to catch-up development speed in traditional algorithm development. Test-case currently in progress.
- No items published
Quality- Fall Off Rate, from 95% (as is) to 98,5% (to be)