iSixSigma

Logan Luo

Forum Replies Created

Forum Replies Created

Viewing 17 posts - 1 through 17 (of 17 total)
  • Author
    Posts
  • #133367

    Logan Luo
    Participant
    #88575

    Logan Luo
    Participant

    Hi Sam,
     
    Yes, it has been a long time. I did not hear anything from Bill, but I think I have found the way.
     
    I believe no matter what kinds of SS tools are being used, they must serve a basic criterion – driving improvement. For large corporations, they may have more resources to make improvements in large scale, but for small companies, they can still use simple SS tools to drive improvement in small scale. It doesn’t matter what kind of tool is being used, what matter the most is making improvements.
     
    Recently, I read an article argued that Six Sigma is fading away as “…as companies find that the technique often achieves less than expected…” and I believe lots of companies are facing this problem now. To me, the solution is “return to the basic” – two major functions of a quality system: monitoring performance, and driving continuous improvement.  I have wrote an article about this subject, click here to read the details.
     
    Please comment.
     
    Logan Luo

    0
    #88221

    Logan Luo
    Participant

    The setup of your monitoring system is wrong. From your example, the monitoring system does not catch any valuable data and monitor the right parameters. You should measure and record the process performance – defect/reason for losing time, not only the process output – total hours loss.  I have seen similar problems in manufacturing industry that made the SPC go nowhere.
     
    Here are my suggestions:
     
    1) Develop a process map and identify opportunity for a defect for each step.  In your case:
    Step 1: Analysis of Customer feed back & Preparation of Report      USL  8 hr   Opt: 1 (I assume you need to conduct only 1 analysis in this step)
    Step 2 Report given to Marketing Dept                                          USL  4 hr   Opt: 1
    Step 3: Reply from Marketing to Customer                                      USL  4 hr  Opt: 1
    USL- upper specification limit
     
    2) Record defect in each step. I would suggest that you use a Traveler. Here is a Traveler would look like if you had one.
    Complaint 1
    Step #                          Total Time        USL     Lost Hour         Defect QTY
    1                                  10                    8          2                      1 (reason for losing 2 hr)
    2                                  5                      4          1                      1 (reason for losing 2 hr)
    3                                  5                      4          1                      1
     
    3) Do the same thing for each complaints or random selected complaints
     
    4) Now your can calculate DPMO, Yield and Sigma Level to represent your process performance. You can calculate these measurements in daily, weekly, monthly or annual basis.
    Complaint 1 Performance:
    DPMO= Total Defect/Total Opp =(3/3) *10^6= 1*10^6
    Yield= 1- (Total Defect/Total Opp)= 1-(9/9)=0%
    Sigma Level= -infinity Z
     
    Assume Complaint 2 and 3 are the same as Complaint 1
    Daily Performance:
    DPMO= (9/9)*10^6=1*10^6
    Yield= 1- (Total Defect/Total Opp)= 1-(9/9)=0%
    Sigma Level= -infinity Z
     
    5) Issue corrective action request to improve the weakness links.
     
    I have made lots of assumptions in your case in order to demonstrate what are going wrong in your monitoring system. I have a computer program, which will do all the works mentioned above. Send me an email at [email protected] if you want to try it or talk more about your case.

    0
    #86837

    Logan Luo
    Participant

    That is exactly the answer I was looking for. Now I can close the loop of my calculator. 
    In stead of looking Z value in a table, I designed this gadget which is similar to the one in this website to calculate the DPMO, Z value, Yield tp and Cpk for me .

    Thanks
     

    0
    #86115

    Logan Luo
    Participant

    Sounds like long term and short term sigma.
    I also use a short cut to calculate sigma level “process Sigma = 0.8406 + SQRT(29.37 – 2.221 * (ln(DPMO)))”. Using this short cut, I can easily calculate all these measurements (DPU, Sigma, DPMO, Yield and etc.) in my database.
    To answer Sergio’s question, I think the key is to calculate DPMO in different levels/process phases, and to have process performance indices in different levels. (E.g. work order/lot, product l, and overall level).
    Like you did in final inspection, you can calculate DPMOs in lot, product, and overall levels to represent the performance of your process, product and plant’s quality level.
    Does anyone have experience in using similar approach monitoring process performance?
     

    0
    #86064

    Logan Luo
    Participant

    1) You are using DPMO
    2) One part may have more than one Opportunities
    3) You can use “DPMO= ((Total Defects) / (Total Opportunities)) * 1,000,000” to calculate DPMOs in Process Level, Work Order Level, Product Line level and Overall level.
     

    0
    #82849

    Logan Luo
    Participant

    I keep receiving emails asking for the program. Please make sure following items before sending me email.
     
    1)      It is a Microsoft Access program. Make use you have Access installed.
    2)      It is not a calculator. It is a database that keeps production data (input, defect, operational time) and calculates Yield, Sigma Level and Operative Cycle Time.
    3)      It is primarily designed for manufacturing processes to record and generate reports/charts for production performance measurements such as Yield, and Cycle Time. I don’t know if it fits to service industries (I think it can).
     
    Thanks
    Logan

    0
    #82674

    Logan Luo
    Participant

    Again, the Lean tool kit. Kanban pull system, one piece flow and takt time target on reducing inventory/WIP which is one of the 7 wastes.

    0
    #82570

    Logan Luo
    Participant

    You’d better take a look at Lean tool kit.
    Anyway, I think the best way to start is to map out the processes and start to record the operative cycle time for each process to define the weakest link.
    I have been working on a computer program, which will calculate DPMO, Yield, Sigma Level and Cycle Time. The program may help you record the operative cycle time for your processes and help you manage the Internal CAR.
    Send me an email, I will forward the program to you.
     

    0
    #81670

    Logan Luo
    Participant

    It sounds like that after you exported the file to Excel, the exported file overwrote the old Excel file which has the Sigma formula. 
    The other possibility is that after you exported data to Excel (without opening it), the Sigma field in your Excel file would still “remember” the old data. If you import the data without opening the file, you will be end up importing the “old data”. You have to open the Excel file to refresh the “memory” of the Excel file.
    You don’t have to export the data to Excel to have Sigma Level. Use following formula you can have the Sigma Level in Access. I have already done that.
    Process Sigma = 0.8406 + SQRT(29.37 – 2.221 * (ln(DPMO)))

    0
    #81619

    Logan Luo
    Participant

    Thanks Ram. Now I know the difference between these formulae.
     
    I have another question.
    In iSigma’s dictionary Yield is defined as
     
    “Yield is the percentage of a process that is free of defects.” [not the probability of process free of defect]OR”Yield is defined as a percentage of met commitments (total of defect free events) over the total number of opportunities.”
     
    And in the iSigma calculator, Total Opportunities has been considered.
    Yield= 100- defect percentage
    Defect percentage= D / Total Opportunities
     
    Can I say these formulae are more appropriate to describe Yield, and e**(-dpu) is for the calculation of the Probability of Defect Free Product? Sometimes I am confused by different interpretations from different sources for a same measurement.

    0
    #81589

    Logan Luo
    Participant

    Thanks Stan.
    What is the difference between Yield=e**-DPU and Yield=1-dpu? The formula has been used in iSixSigma Calculator is Yield=100-defect percent.
    The reason for doing all these is because I am trying to develop a general-purpose software to ease the calculation process of all Six Sigma measurements. Different people may need/like different measurements.

     

    0
    #81471

    Logan Luo
    Participant

    You will not have the average of 5 fields in this way. You have to Union all five fileds (put them in one field) before you can have the average. Some knowledge of SQL language are requried here. You can copy following code and past to SQL. (New a query –>select sql from View–>cut ans past the code–>chage the table name to the real table name)
    SELECT Hole1 as [AllHoles]FROM table nameUNION SELECT Hole2 as [AllHole]FROM table nameUNION SELECT Hole3 as [AllHoles]FROM table nameUNION SELECT Hole4 as [AllHoles]FROM table nameUNION SELECT Hole5 as [AllHoles]FROM table name;

    It seem the table design that cased the problem. You should consider redegin the table structure to following:
    Meansurement ID
    Measurement (record Hole1, or Hole2 or etc.)
    MeasureResults (record the results)
    Let me know if you still have problem. [email protected]
     

    0
    #81469

    Logan Luo
    Participant

    Can I have one copy of the Overview? [email protected]
    Thanks
     

    0
    #81460

    Logan Luo
    Participant

    Hi Bill,
    Could you please advise what kind of the tools that you have applied? I don’t think some of the SS tools can be applied in small company (specially the real SS tools); and small company doesn’t have money for all Black Belts and Green belt who actual drive the programs and project. Training is another issue in small company. How to calculate the real ROI?
    I want to lean from you how to apply SS in a small company.
    Logan Luo

    0
    #81458

    Logan Luo
    Participant

    Try these:
    Mean:
    in the Query filed type in: Mean: =avg([your field name])
    Range:
    type in: Rang: =max([your field name]-min([your field name])
    You have to eliminate other fields to have it working right.
    let me know if it works. I am working on an Access database, which can calcualte DPU, DPMO, Yield, Cycle Time and ect. If you are interested in it, I can send you one for testing when it is ready.
    Logan Luo

    0
    #81406

    Logan Luo
    Participant

    Thanks Erik.
    My email: [email protected]
    Logan

    0
Viewing 17 posts - 1 through 17 (of 17 total)