Objectives: Acquire methodological and practical knowledge of Statistical Process Control (MSP/SPC) and the different types of control charts
Duration and terms: 2 days (14 hours)
Active pedagogy combining presentations, exercises and practical applications.
Each participant will be able to implement the applications in the software of his choice among Minitab, JMP or StatGraphics.
Notes: - General Introduction
- Prerequisite to Statistical Process Control (MSP/SPC): measurement capability
Notions of metrology
Sources of Variation - Common Causes (random), Special Causes (assignable)
Introduction to uncertainty studies (R&R pledge)
- Process capability
Introduction to Six Sigma
Capability index Cp and Cpk
Calculations in the case of parameters not following the Normal law
- Control charts
The principles of the control chart
Construction of a control chart: sampling mode and calculation of control limits from a reference period
Differences between control limits and specification limits
The rate of false alarms in the Gaussian case (Normal distribution)
Western Electric's rules for determining if a process is out of control
The different types of control charts
Individual type single cards
Shewhart maps (mean / dispersion) for the detection of rapid misalignments: Xbar/R, Xbar/S
CUSUM and EWMA type charts (moving averages) for the detection of slow deregulations
Cards with attributes of type P cards or C cards
Multivariate cards (T2 from Hotelling)
Control charts under non-standard assumptions
Different sources of variation
Control limits in the case of non-Gaussian parameters
Auto-correlated data
Reaction in case of out of control
The Statistical Control of Business Processes
Dedicated web site: https://www.datavalue.fr/formation-maitrise-statistique-des-processus-msp-spc