[Amit Kumar Ojha]

Dear Pravin,
First of all RCA is a very generic term and given its wide use it can be used for both type of causes. However, I would like to understand where is it written that special causes can not be controlled or same for common cause. I suggest you are missing the whole purpose of six sigma.

0

[Amit Kumar Ojha]

@leaning:

There are 6 Standard Deviations between Mean and the nearest specification limit for a process operating at 6 sigma level.

So in case both specification limits are present, there would be [-6 to +6] 12 SD between USL and LSL.

0

[Amit Kumar Ojha]

@chris another way of doing it is to first collect all the data which can be easily fetched which would be around 40-50 out of 1000 (I am taking the data only for the current year) and then listing those 40-50 data points and randomly selecting 15-20 out of it.

But then it would not be considered as random sampling. Thats why what I did is to first select some 30-40 data points randomly from 1000 and then ignore those for which data is not easily available, which gave me 15-20 samples.

Awaiting your response!!!

0

[Amit Kumar Ojha]

@ Chris.. You are absolutely right, however I think that I did not make the problem clear enough. Let me try once more:
See we have a process wherein data (numerical) is entered into two different systems (I understand this is redundant and needs to be eliminated at first) pertaining to few entities(multinomial) which are named differently in the two systems. Now I need to analyse this data in terms of variation between two systems, distribution among entities, any patterns etc. The problem is if I list down the data and try to select randomly I end up in a situation wherein for the selected sample, I can’t fetch the details (owing to nomenclature, inconsistency in data entry etc).

Data is huge.

@Sylvain: Right now my objective is Exploratory Data Analysis wherein we need to gauge the problem so as to formulate the Business Case (with Quantification).

Hope I have elaborated the problem well enough for your further valuable inputs.

0

[Amit Kumar Ojha]

Hi Vikas,

In other words I would say that occurrence rating which you need to enter to calculate the RPN is actually the probability of occurrence (it does not imply that the particular failure must have occurred) and hence I think Shelby has given you various methods by which you may figure out the same.

0

[Amit Kumar Ojha]

Perfectly elaborated by Mike and Shelby!!!

In Software Industry, Internal Quality Audits (Functional Testing done by BA Team, Performance and Security Testing done by concerned teams) during testing phase can be valuable if those are done with the focus on customer requirements. In this aspect Audits do add value on unveiling lot of defects which could lead to bigger problems had they been gone into production.
However, the external audits (Final Inspection in software industry) done by third party who have limited knowledge about the particular domain or process may not be useful in providing quality product to the end customer.

0

[Amit Kumar Ojha]

Thanks Robert for the reply. I understood what you are suggesting me to do. But the problem we are facing is that I can not randomly choose the sample. Fetching the data for any randomly selected data is again difficult. Actually there is a system in which a lot of data is just entered and it is not organized. In order to fetch the data for any sample, lot of steps are required such as confirming on the what the attribute value implies, attribute mapping with scenario, validating the meaning / nomenclature with users etc.

Hence I think the second approach which you have mentioned is more appropriate here. I will try it once and see whether the management is convinced.

Thanks for the help!!!!

0

[Amit Kumar Ojha]

Rightly said Chris and Don Strayer. As I have been implementing Six Sigma Projects in IT industry I can give you some example of how we use it.

There can be two aspects here as mentioned below:
1. Improving the Quality of the Software or Software Development Process – In this aspect Defect Density (Defect / KLOC or Defect / Function Points etc), Estimation Accuracy, Computing Resource Utilization Effectiveness, Compliance to Best Coding Practices etc can be used as metrics.
2. Improving the Business Process being digitized – In this aspect, Cycle Time, Effort Consumed or any other process specific metric can be used.

Hope it helps !!!

0

[Amit Kumar Ojha]

Hi Prabhu,

Another line of thought could be to consider the percentage of products recalled and finding the relationship with Market Share for each brand. Here considering both as continuous data, you can use correlation/regression to find the relationship.

Hope it helps !!!

0

[Amit Kumar Ojha]

Thanks Mike.
Sincere apologies if my post had offended you. I just wanted to ask you the name of the book as I was interested. I would first go through the book mentioned in your post. Also Mike you rightly mentioned about the current trend. I have a very small 3 years of experience in Six Sigma where I have seen people using the tools just for name sake rather than thinking of six sigma as a thought process which aims at continuous improvement. I will definitely keep in mind what you said and will do my best to contribute to Six Sigma community wherever I get chance to apply it.

0

[Amit Kumar Ojha]

Dear Mike,

Thanks for enlightening me!!! I was not aware that it started as just MAIC. I would definitely want to read you book so that I can learn from your vast experience.

0

[Amit Kumar Ojha]

Hi John,

My simple response to your query would be – No it does not overlap. Mentioned below is the explanation for the same.

The purpose of Define phase is to know the problem and build consensus on taking the project ahead by getting buy-in from sponsor. Basic tools used in define phase are Business Case, Project Charter, SIPOC, VOC, Affinity Diagram, KANO model, CAP tools etc. While talking about the Measure phase; it involves establishing the baseline process capability though a validated measurement system. The tools used in Measure phase are – VOC – CTQ translation, data collection plan, CTQ components identification, MSA etc.

Hence to summarize, there is a clear distinction between Define and Measure phase in terms of their purpose, tools used and phase of the project when these are used,

All the best!!!

0

[Amit Kumar Ojha]

Hi Mo,

First and foremost, you need the exact count of the following entities for calculating DPMO:
1. No of Defects (D) (and Not Defectives) even if its in percentage you can easily use the implied value i.e. in you case if there are 60% defectives rates (assuming each defective has only one defect), no of defects you can take as 60.
2. Total no of items (N) (again based upon your scenario you can take it as 100)
3. Total no of opportunities (O) (You need to get a count for this entity)

DPMO = [D/(N*O)]*10^6
In you case = [60/(100*O)]*10^6

All the best !!!

0

[Amit Kumar Ojha]

Hi David,

Just a few important points to keep in mind at the onset:

1. First and foremost do finalize on the scope of the process. A tool (you might be well aware of) – COPIS would help you in finalizing the high level process map along with the process boundaries. It would help you in your discussion with the mentor.

2. When you are writing Business Case or Problem Statement, please do not focus on one aspect such as cycle time or defect. You need to identify the pain area and not what is causing the pain / delay etc in the process. A pain area could be something like this:

The current process requires manual rechecking at the end which taken almost 2 Hrs of storekeeper’s efforts per day impacting the productivity.

(Problem statement must have What, Who, When and By How Much)

Hope it helps..

All the best !!!!

0

[Amit Kumar Ojha]

Hi Philip,

Shelby has almost covered all the aspects. In addition I would just like to share my experience with Six Sigma projects. I have found that many people are not very much interested in the documentation part of any Six Sigma project and it is done half halfheartedly which clearly reflects in the output. It is very important to understand the purpose of proper documentation. While in your case, you may think that the main purpose is your BB evaluation, however please note that one of the most important aspects of documentation is to convert your tacit knowledge (gained while doing the project) into explicit knowledge which may help many people in applying the similar solutions and help in overall addition to the Six Sigma knowledge-base.
Hence, apart from focusing on you curriculum specific details (if there are any) please document the project in such a manner that it helps others from your experience.

All the best.

0

[Amit Kumar Ojha]

Dear James,

In general, every measurement system must be validated before you can use the same for measuring the CTQ’s. Here one of the CTQ’s you have mentioned is Seal Strength. In this case if you happen to assign this task of measuring the seal strength to 3 operators, how would you ensure that the variation in the measurement of the seal strength is not due to operators’s measurement techniques but is really the variation in the strength.

I would recommend MSA to be a done for every measurement and also if MSA has been done once for an instrument in the past, it needs to be performed again in order to ensure that the instrument is reliable enough to produce same measurement over a long period of time.

All the best.

0

[Amit Kumar Ojha]

Dear Prasenjit,

In order to understand and model the combined effect of multiple independent variables on one dependent variable, you need to perform multiple regression. There are basically two approaches i.e Forward Stepwise and Backward Stepwise.

Both these approached are available in Minitab 16.

You use Forward Stepwise when you start with one most important variable and continue adding others, whereas you use Backward Stepwise Elimination when you start with all the variables and proceed with eliminating one at each step.

Hope it helps…
All the best!!!

0

[Amit Kumar Ojha]

Hi Tracy,

Katie is right but since you asked the question, I am mentioning below how you can resolved it.
See here any lot which has more than or equal to 6 defects will be considered defective. We are concerned about the number of lots which would be accepted (non defective) in long run. Since here we are talking about defectives, the appropriate distribution would be Binomial Distribution.
No of Trials(N) = 80 , Probability of success(p) 0.1

Probability of getting a lot which is defective = P(Defects <=5) [Let r=5]
The formula for calculating the probability in binomial distribution is NCr*(p^N)*(q^N-r).

Hope it helps..If you find further difficulties, please feel free to let me know.

All the best!!!

0

[Amit Kumar Ojha]

Hi,

As per my expereince the time it takes to complete a six sigma project mostly depends upon the following factors:
1. The calrity of scope.
2. Top Management commitment – Whether top management is ready to invest the time and resources in the project and if yes, then to what extent. Much of it also depends on the criticality of the project.
3. Urgency to fix it now. If there is no urgency, top management may provide limited resources leading to delay in the closure.
4. Availability of Data.
5. Mentor Support.
6. Involvement of cross functinal teams and overall attitude towards continuous improvement and operational excellence.

Going be my experience, I would say that it generally takes 6-8 months on an average to complete a green belt (considering 2 months or pre and post data for analysis) and 12-16 months for a black belt project.
PI projects of other methodologies such as Lean / PDCA gets closed in reletively shorter duration (2-3 months on an average).

0

[Amit Kumar Ojha]

Hi Jackie,

As such you dont need to worry too much about the know-how of using a tool. As Rinkon mentioned you can find all the details in iSixSigma. However, what is more important is your expereince in selection of appropriate tool to be applied in a particular business scenario.

All the Best!!!

0

[Amit Kumar Ojha]

Thanks Robert for your valuable insights.
This discussion is very useful for people who want to know the basics of DOE…

0

[Amit Kumar Ojha]

Thanks Robert, Norbert, Shelby and JB.

I read you valuable comments. Although I have never used DOE but after carefully reading the inputs, I came to the following conclusion. (Please correct me if my understanding is wrong)

1. DOE is an active method wherein the factors are manipulated to see their impact on dependent varaible however, regression is passive wherein the available data is analysed for determining the variable which have significant impact on Y.
2. Robert what I understood from your post is that you used DOE to find the optimum value for your inputs. (One small doubt here may sound a bit silly also- Can we not use some mathematical techniques such as Linear Programming for optimization as these are relatively easier to use given the constrains and factors)

All the bset!!!

0

[Amit Kumar Ojha]

Hi Nibin,

BY write up, what exactly do you mean ? If you mean Project Charter, here are few guidelines:
1. You need to mention Business Case (Current Pain Area), Specific Problem Statement (Pain Area Quantified), Goal Statement (Improvement target), In scope and Out Scope, Team Roles and Responsibilities.
2. Only if you explain the entire business case and improvement, I can help you with estimated savings. Actual Saving depends upon your actual data. For Cost/Benefit or ROI analysis take Impacted Users*Usage Frequency*Effort Saved*CTC

3. This is Soft Saving. Hard saving is direct tangible savings such as releasing a facility / machinery etc.

Hope my response helps..
Best of Luck!!!

0

[Amit Kumar Ojha]

Hi John,

See as far as I know, there can be only one point which you need to take care about the formula for sample size calculation i.e. whether you are concerned about one tailed or two tailed test.
In this context Nigel is correct.

However, the formula which you used to calculate the sample size is not correct.

Refer Levin and Rubin – Statistic for Management for sample size calculation. This is best book I have ever known for statistics for beginners.

0

[Amit Kumar Ojha]

Hi Patrick,

As mentioned by Strayer, you are complicating the things at the initial level.
See first understand the purpose of COPIS. COPIS is not a tools for in-depth analysis of your process, the purpose of COPIS is :
1. To communicate to everyone the high level understanding of the process
2. To check if there is proper alignment / utilization of all inputs / process steps to generate the output as desired by customer. It helps in seeing if there is any gap in relating Customer,Output,Process,Input & Supplier.

Now coming to your business scenario, first and foremost find out who is/are the customers. (Remember now this tool is referred as COPIS to signify that you always need to identify your customer first and then go backwards to see what are outputs to satisfy the customers needs and how you process converts the inputs into those outputs. There can be multiple Customers, Outputs and inputs or suppliers but you need to fix the boundaries of your process.

Response to your two queries:
1. Yes your understanding about customer is correct.
2. No a trigger / event can not be an input.

Hope my response helps you.
Best of Luck!!!

0

[Amit Kumar Ojha]

Yes Tracy.

You are right. No success in this case means 0 defects and 1 success implies 1 defect.
Success on single trial of 0.1 means that if you test a sample of 100 units what is the probability of getting 10 defects.

Hope I clarified your doubt :-))

Best of Luck!!!

0

[Amit Kumar Ojha]

Hi,

by 1 proportion test for determining sample size what exactly do you imply ? Sorry I dint get that :-(

0

[Amit Kumar Ojha]

Hi John,

Please find below the solution for the given problem:
It is given that

Tolerance = 10
Alpha = 0.05 and Hence Z = 1.96
Standard Deviation corresponding to 3.4 parts per million defect rate = 6
Hence Sample Size = ((1.96*6)/10)^2 = 32.14

So you need a sample of size = 33 (Rounded Off)

Good Luck !!!

0

[Amit Kumar Ojha]

I think the basic difference between education and training is in terms of the objective and the way they are done. The fundamental purpose of education is learning and growth whereas that of training is improving the performance of the trainee through teaching him / her the skills required to perform a job perfectly.
Education encompasses training, teaching, skill building as well as learning by experience.
Hence eduction can be considered as a superset which encompasses training as one of the way of enhancing skills in a particular job.

Also training particulary focusses on practical application whereas eduction also involves teaching the theory in addition to its application.

Philosophically it can be said that education goes much beyond skill building or accumulation of knwoledge. Purpose of eduction is Enlightenment where in apart from knowledge and skills it also involves awareness about one’s duties and responsibilities towards soceity.

0

[Amit Kumar Ojha]

Norbert shared a very simple way by which you can get the answer without getting into complexity of calculating the probabilities.

In the above problem, since the calculations were not so tedious, hence you can do either way. However, in many real life scenarios the calculations are too complicated to be done manually.

Hence it is good to use minitab, the way Norbert explained.

Thanks Norbert :-))

0

[Amit Kumar Ojha]

I Agree with JB. Better use the same period before and after for showing the data in SPC chart and use simple Box Plot to compare the peformance befor and after.

0

[Amit Kumar Ojha]

Hi,

The correct answer is D(0.9662)

Here is the solution….
First of all consider this as a problem of Binomial Distribution. Our objective is to find the probability of getting 2 or more defects (thats when the lot will be rejected).
Technically we can write it as P(X>=2)
It is given in the problem that X is no. of defects, n=50, p=0.1, q=0.9

P(X>=2) => 1- [P(X=0) + P(X=1)]

P(X=0) => 50C0*(0.1^0)*(0.9^50) = 0.00515
P(X=1) => 50C1*(0.1^1)*(0.9^49) = 0.02863

Hence P(X>=2) => 1 – (0.00515 + 0.02863)
=> 0.966214

Hence Option D is correct Answer

Please let me know if you need any further clarification on this or any other such calculations.

Good Luck!!!!

0

[Amit Kumar Ojha]

ASQ is widely recognized. Also regarding the reference material, you may need to refer more than one book.

Try to find out topic wise books.
For Advanced Statistic you may use Levin & Rubin – Statistics for Management
In general The certified Six Sigma Black Belt Handbook by Kubaik is good reference material. pls note that it has few printing and calculation mistakes but the coverage is pretty good.

0

[Amit Kumar Ojha]

One more point, please do not limit the scope of the analysis by presuming the cause as having too many vendors.

Many a times I have observed that, post thorough root cause analysis, it is found that the cause which was presumed is actually not a very significant one.

I suggest let us go step by step as per DMAIC methodology.

0

[Amit Kumar Ojha]

So in this case the real pain area is delay in getting the stuffs. Now I suggest you to quantify the problem as in how much time does it take on an average and how is it negatively affecting the business in terms of causing further delay to other activities or loss of sale etc.
Once you are done with this, we are good to go with the problem statement.

Post that we can focus on other aspects such as finalizing the CTQs etc.

Goddluck !!!!

0

[Amit Kumar Ojha]

The hierarchy Yellow-Green-Black-MBB is made to indicate the increasing depth and scope of the tools and their application. If you feel that you understood the tools very well and can relate how it is applied in different situation, you may skip a level.
However before you skip pls make sure that you do not miss any basic concept so that you find it totally new in green belt.

0

[Amit Kumar Ojha]

I suggest you to first try to link the strategic objective with operation goal and also keep in mind that there many variation and tools in Six Sigma. The application of tools depends upon your objective as well as the scenario.

Good Luck !!!

0

[Amit Kumar Ojha]

Agree with Shelby’s point regarding scope finalization.
Its very important to finalize on the following points along with concurrence from your friend.
Try to think of the following points:
1. Business Problem
2. Goal
3. In scope / Out of Scope
4. Timelines
If you planning to do it phase wise than think of phase wise plan…

0

[Amit Kumar Ojha]

Start with the Value Stream Map. Try to look for the areas where most of the cost is incurred or cycle time / efforts are involved. Then try to do a root cause analysis so as to find the reason for the inefficiency. Techniques such as pareto analysis along with fishbone diagram would help you see the real pain.

Then try to come up with alternatives and use optimization techniques to re-design the process.

0

[Amit Kumar Ojha]

Hi Hmalfedel,

Can you exactly state what is the problem your friend is facing with two many vendors. Try to focus on the pain area and then start quantifying it. Having two many vendors can not be a problem statement as it does not specify the real pain area.

Hope it helps !!!

0

[Amit Kumar Ojha]

I agree with Don Strayer. In IT/Software industry DFSS is more applicable than DMAIC. DMAIC basically applies when you have a process which you want to improve however, in case of product/process design DFSS is used to pro-actively optimize the design at the initial stages in order to fulfill customer requirements.
The suitability of DFSS for agile development methodologies can be found in many books.

0

[Amit Kumar Ojha]

According to me, this problem statement would limit the scope of analysis as the cause of the problem is already mentioned.

Mentioning below few points to be taken care while formulating any problem statement in a Six Sigma project:

1. It must give the real pain area in terms of quantitative data.
2. Three aspects must be present in the problem statements – What, When, How Much.
3. Neither the problem statement should be two broad or generic nor should it be so specific as to limit the scope of analysis.
4. The cause or solution should not be implied in the problem statement. If you already know the cause or solution, there is actually no need to do a Six Sigma project.

Hope it helps!!!!!

0

[Amit Kumar Ojha]

In order to determine the takt time you need to consider the available time which in this case is limited by op2. Hence you may consider 2400 sec as available time to determine takt time.

0

[Amit Kumar Ojha]

I think you can consider ARIS as one of the Business Process Modelling software which has industry wide acceptance and excellent capability to model the business process. There are various levels available in ARIS through which you can model your entire organization starting from Organization structure to detailed process flows.

0

[Amit Kumar Ojha]

If the demand is not clear, how did you figure out the Takt Time …
I suggest you to use some predictive modelling technique such as Regression analysis to predict the demand and based upon that you need to determine the manpower requirement.

0

[Amit Kumar Ojha]

For the first three points you just need to input the data in any of the statitical softwares and get the descriptive statistics. By default SPSS or Minitab gives the 95% confidence interval for mean and standard deviation.
Attaching the output for your data obtained through SPSS. You may further refer Statistics for Management by Levin and Rubin in order to learn calculations.

0

[Amit Kumar Ojha]

Change in the culture is a slow process which required change in the way employees perceive their own work and its significance in achieving organizational goal.

In order to initiate the cultural change, one needs to clearly state the objective and expected benefits of the change to the top management and post getting the buy-in from them, it must be conveyed to all the employees. Various Change Acceleration Process tools such as ARMI, GRPI etc. can be employed to effectively track the change initiative.
The important part is to enable everyone in the organization to relate to the change and see it as a necessary transition to achieve the organizational goal. The resistance to change must be overcome through effective influence strategy.

0

[Amit Kumar Ojha]

The main difference between Mean Absolute Deviation (calculated by taking the absolute value of difference around mean) and standard deviation (calculated by squaring the differences and then adding them up and finally taking the Square Root) is that Standard Deviation gives more weightage to the extreme value and hence is considered a better estimation than Mean Absolute Deviation.

0

[Amit Kumar Ojha]

I agree with CFB.
The most important point in six sigma project is to clearly identify the scope of the process based upon which one can easily identify the customer.
As an example, let us take an ‘Appraisal’ process of an organization. Let is consider that the scope includes all the full time employees who have worked more than a year.
In this process the recipeint of the service (performance appraisal) in terms of increments / promotion is ‘Employee’ of the organization. Hence the customer for ‘Appraisal’ process would be the ‘Employee”

0

[Author]

Posts