semiconductor, biostatistics, aeronautics, automotive, pharmaceutical, nuclear and space industry ...
Company
Ippon Innovation is a team of high level Data Scientists working in Statistics and Big Data through Artificial Intelligence for semiconductor, medical, aerospace, food industry... For example, solutions for process & yield optimization and outliers detection are developed.
Ippon Innovation, headquartered in Toulouse, France, was created by François Bergeret, PhD in statistics from the University of Toulouse, who held various positions in R&D, production and quality for 15 years with Motorola and Freescale (now NXP).
Ippon Learning also proposes training and consulting in Lean Six Sigma(Black Belt, Green Belt), Statistics (DOE, SPC, modeling, data science...), Quality Methods (FMEA, Problem Solving...), Metrology (ISO17020, ISO17025...).
For R&D, Ippon Innovation collaborates with the University and Institute of Mathematics of Toulouse, as well as with the Occitanie Region and the European Union.
As JMP business partner ( www.jmp.com), Ippon Innovation and Ippon Learning have developed a strong expertise with JMP software including: advanced scripting, consulting and training.
Some customers:
Solutions
Ippon proposes innovative solutions to improve the reliability of your products,yield, process and cost of test.
For High
Reliability GAT is in production for our semiconductor and pharmaceutical customers.
Reliability GAT is in production for our semiconductor and pharmaceutical customers.
GAT objective is to select anomalies by techniques of outlier ensembles.
GAT is designed for data of any dimension, including low sample size and high dimension (HDLSS).
GAT includes a double unsupervised algorithm to identify outlier items and can be applied in various fields.
Developed in partnership with Toulouse university, GAT is the ultimate zero defect tool.
To improve processes & yield
YETI is a software that finds root causes of issues in complex processes.
It is based on a set of statistical analyses adapted to any kind of parameters.
Yield has been improved by several percents in semiconductor manufacturing and drug quality has been improved in pharmaceutical industry with YETI.
To detect abnormal signal
CHAM is used by a major energy company
CHAM is a solution developed by ippon innovation to detect anomalies on sensors curves, process or metrology tool parameters or any weak signal.
Statistical analysis is applied on temporal curves, multiple parameters or any simple or complex signal, in particular images.
Learning
Statistics
Statistics and Modeling
Data Science
Outlier Detection
Design of Experiments
Statistical Process Control
JMP Fundamentals
Lean Six Sigma and Metrology
Green Belt and Black Belt with certification IASSC
Design For Six Sigma - DFSS
Measurement System Analysis and Metrology
Metrology norms: ISO 17020, ISO 17025
Statistics and ModelingData ScienceOutlier DetectionDesign of ExperimentsStatistical Process ControlJMP FundamentalsLean Six Sigma and Metrology
Green Belt and Black Belt with certification IASSCDesign For Six Sigma - DFSSMeasurement System Analysis and MetrologyMetrology norms: ISO 17020, ISO 17025
Ippon Learning proposes a large set of trainings,
standard or customized, in class or e-learning based.
standard or customized, in class or e-learning based.
For more information on trainings, please contact us.
NEWS
Information, news from Ippon
Previous
Next
GAT SOFTWARE: ZERO DEFECT FOR AUTOMOTIVE
Our GAT software was selected in 2026 by a major player in automotive electronics! Its multivariate algorithm detects atypical parts, also used in the space and optics sectors.
01.04.2026
The IPPON team grows !
Welcome to Charly and Lilia, who join the IPPON team and will be working in microelectronics, aluminum, and cosmetics, and welcome to Yasmine, she’ll be focused on AI and image processing!
01.04.2026
TOLERANCE CALCULATION FOR AEROSPACE
Ippon has developed a tolerance calculation solution for Gaussian and non-Gaussian variables for a major player in aerospace, using JMP and its programming language. Used by several dozen users.