7 Steps to ERADICATE Defects

 

How to Define a 7Epsilon In-process Quality Improvement Project

Determine parts and defects that contribute to maximum finanical loss due to scrap and rework. Define a goal for the in-process quality improvement project and form a project team.


7 Epsilon’s 7 Steps to ERADICATE defects effectively refine Six Sigma’s Measure, Analyse and Improve phases for Continual Process Improvement as per the requirements of the ISO9001:2015 standard.

 

Knowledge Retention and Reuse

  1. Establish process knowledge [x's], [y's]
  • Capture team member’s knowledge about the process, its factors and responses as well as causal relationships codified using pictorial diagrams such as Process Maps, SIPOC Diagrams, Cause and Effect Diagrams, Factor and Process Response (FPR) diagrams.
    1. Refine process knowledge [y=f(x's)]
      Process engineers systematically research about process factors and responses to find out
    • how factors are related to responses and how they can be measured;
    • importance of factors in relation to responses.

The outcome of this phase is a written description of process factors' characteristics with respect to one or more responses. This proces helps to build a 7Epsilon organisational knowledge repository.

 

Knowledge Discovery

    1. Analyse in-process data using penalty matrix approach
    • Typically in many organisations, in-process data is routinely collected. Using the knowledge developed in the ‘Refine Process Knowledge’ step, the data on important factors that might affect process responses is captured and retrieved for analysis.
    • A penalty matrix approach is adopted to perform rootcause analysis and discover correlations among regions within the tolerance limits of factors and good and bad response values. Correlations and patterns found using the penalty matrix approach are prioritised. p-matrix software provides strength values to prioritize optimal.avoid ranges. Product specific process knowledge i.e.  product specific optimal and avoid ranges by visualizing patterns in data.
    • Traditional data analysis methods may also be used to analyse design of experiment data and/or computer simulation data

       

    1. Develop hypotheses (or potential solutions)
    • Using the knowledge developed in the ‘Refine Process Knoweldge’ step and correlations discovered in the Knowledge Discovery step, hypotheses on causation are established. These are referred to as new opportunities for in-process quality improvement.
    • New tolerance limits are proposed and a corrective action plan is outlined or a decision is taken to collect either more in-process data or conduct one or more design of experiments studies.

       

    1. Innovate using rootcause analysis and conducting confirmation trials.

Cultural of Innovation

    1. Corrective actions and update process knowledge
    • Upon successful completion of confirmation trials, the new knowledge obtained in the previous step can then be stored in tabular form and consists of a list of values with their new specifications. It must be noted that the new specification ranges are specific for a given part and process.
    • This knowledge is reused alongside the process knowledge compiled in the ‘Refine Process Knowledge’ step and becomes part of the 7Epsilon organisational knowledge repository.

       

    1. Building Aspiring Teams and Environments by monitoring performance
    • Once the new process specifications have been implemented, foundries must continually monitor the process responses to assess the effectiveness of actions taken.
    • The 7Epsilon organisational knowledge repository can also be used to train operators, supervisors and process engineers.