sumant

Hi, can you send me that airplane game???
my ID: [email protected]

Sushant,
you can use DOE to identify the critical X’s. if you have all the factors with all the levels then go for full factorial or OA method. get the X’s which is affecting output by looking at ANOVA.

Stan,
There are two types of FMEA. one is Design FMEA and another is Process FMEA.
Yes, they have to do some FMEA in conceptual phase, that is Design FMEA. once process is ready and it is ready for production, you should do Process FMEA.

yes Tan, that is possible. If population follows normal dist and sample of 30 follows non-normal, then your sampling may be incorrect. It is not random. if it is random, then defineite it will follow normal.
Othercase, it is 100% true. averages of sampling distribution will always follow Normal, whether population follow normal or…[Read more]

Vikram,
Always use FMEA before going for full production. Develop FMEA in new process map and find out where process can go worng. then find out action items, for each step, where process can fail. Basically FMEA is a preventive tool, so always use it before full fledged implementation

Hi Csaba,
can you send me the same in this ID: [email protected]
Thanks in advance.

Hi Sudhindra,
I can help you in this project. I don’t have any such projects, but I have implemented experience.
Have you found out process capability. How it is? tell me in which phase you are or the project status. I can help you how to proceed further. If you know the factors which is affecting the output, then you can use DOE to solve this…[Read more]

Tan,
At least you need 30 data points to check normality. yes, data points should be randomly choosen. then you can check for narmility by using some normality test. you can go for anderson darling test in MINITAB.
if you are taking means of sampling distribution, it will always follow normal distribution. Sampling distribution can follow…[Read more]

Nitesh,
CMM is Capability Maturity Model. There are 5 levels in CMM. Basically CMM is a certification like ISO, but CMM is more focused towards IT, where as ISO is very generic and applicale to all industries. Unlike Six Sigma, There is certification available and registrar bodies are there for CMM. CMM will gives you process framework, measures…[Read more]

Hi Javier,
first thing is instead of collecting claims per day, you should measure claims arrival pattern in terms of hour or half an hour interval. Then it will work. once you have arrival pattern, then get the turn around time required to process each claim, so that you can get time required to process the claims in that interval. Then…[Read more]

Hi Nar,
sample size is not only one factor for getting non normal data. Thare are lot of reasons. e.g when ever you have one sided limit (only USL or LSL), maximum chances you will get non normal data. sample size does matter, but if you have more than 30 samples, it is good enough to show normality.
Thanks

Hi,
if you have 48 samples data, then you can enter it in MINITAB and do Z test as here SD as benn provided. you should use t-test if Population SD is unknown. otherwise generate 48 random samples with mean 59500 and SD of 5000. and do Z test for significance. you will get p value of more than 0.05 which means you need to accept null hypothesis.…[Read more]

Hey, any company can go beyond six sigma level. There are lot of companies operating more than six sigma level. but there is nothing called as fourteen sigma.

Hey, you can use some scientific sampling method. like for attribute data, you can use MIL STD 414D. it will give you no of calls per day you need to monitor based on the population(no of calls handle per day). there are 3 factors you need to identify. you can use all 3 standard levels of sampling….
hope this clear….

Hi Raj,
-Ve Z value means your MEAN is more than USL. but in new process shows good improvement thats why you got +ve Z value.
In any time if you are getting -Ve Z value just take it as 0. that means you are at 50% defect rate.
ok… let me know if you need further clarification
Thanks
Sumant

Indian Statistical Institute is one of the best consultant for prviding training in Quality tools,Six Sigma techniques etc…
In India ISI having 10 center(generally in big cities). If you want I can give you addresses.