• Romel replied to the topic Hypothesis testing in the forum General 14 years, 4 months ago

    try to use the sample size calculator seen in the isixsigma website. You need to determine first the confidence interval, standard deviation, and precision.
    for the sampling, you need to carry this out in the 4 zones but the size may vary depending on the population per zone.

  • Romel replied to the topic T-test help in the forum General 14 years, 4 months ago

    one sided t-test is used if your change had been BETTER than before. 2 sided t-test is used to determine if something had changed.
    For the p-value, if it’s >0.05 accept the null

  • if your data is right skewed, you need to check first the randomness and how you have subgrouped the data, there maybe initial problems on these.
    you can’t go directly to SPC because using other test for outliers might conclude that your graph is out of control
    if the data cannot be transformed, you can use non-parametric tests

  • You can use detailed process mapping then highlight the critical steps where vulnerability is high.
    You can also consider FMEA and some hypothesis tests.

  • You cannot remove the opportunity despite having an error proof of that opportunity. It will always be counted as such. What will be diminished or reduced are the defects relative to that opportunity improved but not eh opportunity itself

  • There are a variety of tools to be used such as Hypothesis Testing, DOE, Regression, among others.
    BUT before you use these tools, and focus on supplier alone as the sole factor, consider also your warehousing and supplier in-transit. the fragility of the components can have high variability the moment these are moved and/or stored.

  • Romel replied to the topic DOE Process Optimization in the forum General 14 years, 4 months ago

    Using ANOVA/Parameter Estimates hekp you determine those with insignificant interaction. BUT before performing this, you need to check first if there’s lack of fit or not and the Rsq and Rsqadj (difference of the two should not exceed 0.20)
    If you are done with all of those stuff, either Minitab or JMP command will give you the optimal setting…[Read more]

  • Romel replied to the topic Cpk Ppk results in the forum General 14 years, 4 months ago

    Cpk focuses on process capability of a sustained process/product while Ppk is a process capability that focus on a product/process under design. the goal for Ppk is 1.67

  • I would suggest that you use DOE instead because multiple or simple linear regression only focuses on the interaction of X and Y while DOE focuses on the interaction of different Xs combination with Y then provide you an optimal setting later on. Interaction is very critical and not only the main effects similar to what multiple regression does.

  • You can utilize your existing product development process and integrate with DFSS to ensure that your life cycle cost will be very minimal in the implementation stage. Tendency before was that without DFSS, costs still intend to incur a lot during the implementation. Instead of DMAIC, you will now use DMADV with rpoduct development.

  • Come up with cross functional projects inclusive of vendors. Projects which deal with defect reduction/incoming quality, delivery time supplier auditing, among others. DMAIC will be done through a shared responsibility say internally, you focus on DM, then the vendors on A, then finally shared effort on IC. This will be dependent of course on how…[Read more]

  • Romel replied to the topic Comparison of Count Data in the forum General 14 years, 5 months ago

    You must use chi-square test (goodness of fit and tests for independence).

  • For the customer facing processes, identify first what customer needs (Ys – time, quality, cost) have low performance over time — use a trend chart, hypothesis tests, and correlation to prioritize
    For the high cost processes – you can use bar charts, pareto
    Then prioritize whether the problem in customer facing or high costs should be given…[Read more]

  • Romel replied to the topic FMEA for Csat Project in the forum General 14 years, 6 months ago

    If you want to have a deep dive in the CSAT project that you’re doing, combine this with Chi-square to check what among the factors contribute to CSAT then proceed to logistic regression. Once you have done these, you can readily identify what’s causing the CSAT score then come up with solutions. Before implementing your solution, that will be the…[Read more]

  • You need to identify first the margin of error and standard deviation and check if you are having the right sample size. This approach can be done two ways — either as a whole group comprising of different agents or on a per agent scenario.
    If you want to see the significance of each agent of concern irregardless of varying sample size, you can…[Read more]

  • You can use Cpk for continuous data metrics on a call center like AHT, among others. If you intend to monitor defects on calls, you can only use the accuracy or yield formula which is good outs/inputs then compare the results of the yield in the 6sigma Cpk table since the data is more on attributes. You can also use control charts on both.

  • You present the ffg.:
    Objective of the project
    Why is it important
    How will success look like
    Help Needed

  • After SIPOC, you can re-confirm the critical Xs and Ys. Afterwards come up with diagnostics and competitive metrics. You can start performing your hypothesis testing and initial process capability from that time.
    Proceed to ANALYZE, IMPROVE, and CONTROL

  • I would suggest that you just use the simple trend charts and compare the actual vs the target set. The target that you have set should be balanced with other metrics identified for the system. If there are any glaring variabilities, just use a paired t-test. Using a control chart for downtime may cause confusion since the downtime by itself may…[Read more]

  • Map the process first then come up with the appropriate activity drivers then the cost drivers, afterwards focus on the cost drivers which pertain to internal and external failures and compute for COPQ

  • Load More