How Jinmyung Precision Drastically Reduced Tryout Cost Thanks to Robustness Analyses
In this two part series Chanho Lee, Technical Manager AutoForm Korea, shares his experience in winning over Korean Tier 1 companies who invited AutoForm to examine problematic parts, still unresolved in tryout. In this post we reveal the powerful robustness analysis project in pictures that won the customers over.
In 2017 AutoForm Korea started a project to support several Tier 1 tool makers. The ongoing project allows the AutoForm team to demonstrate how to reduce the number of tryout loops by doing robustness analysis.
The Jinmyung Precision experience
Speaking of one of the tool makers Chanho said ‘After our first 2017 meeting we suggested Jinmyung Precision allow our team from AutoForm Korea to perform a Sigma study on their critical part to showcase the differences of single simulations vs a robustness analysis. They handed over the simulation files for finished engineering jobs waiting to enter tryout. This is when we successfully found dangers lurking under the waters that their simulations had not yet found.’
Fig. 1: Comparison between successful engineering result gained by single simulations (top row) vs Real Tryout (mid row) & Robustness result (bottom row) (HCT780X 1.75t, 72 realizations, CE+, 12 CPUs, 2.5hrs)
‘Jinmyung Precision had successfully finished their engineering phase, yet at the first tryout cracks occurred in three locations. Even though single simulation results could not identify critical areas our robustness analysis – taking into account unintended variation of process conditions in real production – revealed that CpK (process capability) was very bad at exactly those same points’ said Chanho. Impressed by those results Jinmyung then introduced AutoForm-Sigmaplus over several months to specifically reduce the number of tryouts and gain significant cost savings.’
Using Sigma reminds me of the great Dave Brailsford, the GM and coach of Britain’s cycling team, who said “If you break down everything you could think of that goes into riding a bike, and then improve it by 1%, those small gains would add up to remarkable improvement.” Of course they won the Tour de France title five times.’ In the same way it’s the quick simulation that was done in only 2.5 hours that amounted to massive savings later.’
Read part 2 here, with more robustness analysis pictures from another Tier 1 supplier from Korea.