HIPERREL
- /
- POC
- /
- HIPERREL
The experiment aims to improve reliability in metal and rubber parts production using Plexalytics software. HPC will be used to handle extensive computing capacities. The project aims to identify errors early, apply relevant measures, increase product reliability and reduce failure probability.
Start date: 01/06/2021
Duration in months: 18
Problem Description
Medium-sized part suppliers must identify production errors early to avoid high warranty costs and customer dissatisfaction. Reliability is crucial, but small and medium players face challenges. Automated evaluations using stochastic methods, artificial intelligence, and machine learning can help.
Goals
New services
Challenges
Analysing thousands to hundreds of thousands of parts with simultaneous recording, storage, and evaluation of relevant parameters over a long period of time requires extensive storage space and computational power.
Innovation results
The SaaS platform Plexalytics was used as a basis for the HiPerRel tool, allowing observation, calculation, and reliability improvement of metal and rubber parts. The tool runs on HPC to increase reliability and reduce failure probability, allowing for immediate adjustments to production parameters.
Business impact
HPC e Plexalytics sono integrati nel software a servizio, consentendo una rapida analisi dei modelli di affidabilità delle parti. Il risultato è una riduzione del 15% dei tassi di guasto dei pezzi a lungo termine e si prevede un aumento della base clienti del 25% entro tre anni.