The PROPHESY system and its components were evaluated in a real-life manufacturing environment, in Jaguar Land Rover (JLR) and Phillips factories. Within JLR and Philips, use-cases have been defined around the maintenance of cutting tools and cold forming tooling for high precision metal parts. The evaluation was performed by Philips and JLR employees (shop floor employees, engineers, IT specialists and management) and involve all aspects of the system, technical and economic.
Data is being collected from several sources (material, machine, product, process and tooling) and fed into a set of PROPHESY-ML algorithms to determine the RUL of the wear parts involved. The PROPHESY-ML algorithms are trained to detect patterns in the data and predict failures without the need for ‘feature engineering’.
The PROPHESY-ML RUL prediction is targeted to move from breakdown maintenance and preventative maintenance to predictive maintenance. The impact evaluation was performed by Philips and JLR employees (shop floor employees, engineers, IT specialists and management) and involve all aspects of the system, technical and economic, resulting in viable business cases for both the JLR and Philips PdM implementation. The techno-economic impact of the PROPHESY PdM instantiations was evaluated by assessing the status of selected KPIs, as defined for both JLR and Philips at the start of the project. Indicatively, at Philips, the MTTR has slightly increased at M36 of the project, with significant improvement of the registered OEE. At JLR achieved consistent improvements in the amount of inventory of cutting tools and a steady improvement of MTTR and Technical Machine Availability. PROPHESY has also created a valuable portfolio of cost-benefit and optimization tools that is available online through the ecosystem either in the online web form format or through offline Excel spreadsheet downloadable forms-tools (www.pdm4industry.eu). These are offered for free in the online community to any registered member of the ecosystem.
- It is challenging to run a development project in a real-life production environment due to high pressure on production output.
- Data collection from legacy systems brings some specific IT challenges.
- Data understanding and data-pre-processing are crucial before starting the data analysis.
- It was proven once more that close collaboration is needed between data scientists and process experts to get reliable results.
- The results of PROPHESY-ML application to the use cases have been reported for both demonstrators in the project’s deliverables, achieving RUL prediction with accuracy as little as 4-5% Mean Absolute Percentage error.
- No items published
- Not specified