Mayank Gupta

This was a tricky one as most people use the normal pareto chart only. The answers provided by Mohan PB and Vastupal are very close. The best answer selected is from Vastupal basis the detailed explanation provided for both Pareto and Weighted Pareto along with an example.

Dear Ransingh A very good question and the link shared by VK will help you visualize how CLT works. I want to highlight a common misconception about Central Limit Theorem. It is probably one of the most misunderstood concepts in Lean Six Sigma. Most of the people assume that if they have a large sample size (read greater than 30), then the data set follows normal distribution. This is far from truth. Irrespective of the sample size, the sample will always follow the distribution of the original data set. So if the original data set is Not Normal, then the sample (be it size 1 or 2 or 10 or 30 or 100 or however big) will also be Not Normal. Then where does CLT apply? CLT applies on the distribution of the sample means or sample sums i.e. if i pick up multiple samples from the Not Normal data set, calculate either the sum or the mean of all the samples and plot them on a histogram, then it will follow a Normal distribution. For e.g. consider a roll of a single dice. Possible values are 1,2,3,4,5,6 each having the same probability. A common misconception would be if i roll the dice multiple times (say 6000) times, I will get a normal distribution. This is not true. Roll of a dice follows a Uniform distribution and hence if you roll it 6000 times, it is likely that 1 through 6 will occur 1000 times each. However, what happens if 2 dice are rolled and sum of each roll is noted. The possible values are 2,3,4,5,6,7,8,9,10,11 and 12. Here however the probability is not the same. Prob. of getting 2 = 1/36 (only 1 combination will give 2) Prob. of getting 3 = 2/36 (2 combinations will give us 3) Prob. of getting 4 = 3/36 (3 combinations will give us 4) and so on...... 7 has the maximum probability (6/36) of occurrence while 2 and 12 have the least (1/36). Now, if I roll the 2 dice for 6000 times and plot the sums of each roll on a histogram, the plot will start resembling a normal distribution because of the variation in the probabilities of each number. Here if you notice closely, 1. The original distribution is Not Normal 2. Taking 2 data points from the original data set will give me a sample (equivalent to rolling of 2 dice). Then for each sample, the sum is being calculated and plotted 3. CLT is being applied on the sum and not on the individual data points The same is evident in the animation link shared by VK. So let's be aware of the misuse of this theorem and apply it correctly. P.S. there are multiple online sources where you can also find the mathematical proof of the the Central Limit Theorem.

Hi Prakash Please review the formula for Cp and Cpk calculations (covered in GB session). From the data, you will calculate mean and standard deviation. In addition to the data, you will need the specification limits. Thank you!

Dear Prakash How are weekly quality scores calculated? Provide information on what goes into the Numerator and Denominator. In my understanding the quality metric (for a KPO industry) is usually Discrete in nature. That means we either calculate defectives or defects. Secondly how is your quality metric being reported out (Defective % or DPU or DPMO). Once you have the answer to these, you can use the Sigma Calculator (shared during the GB session) for Discrete Data and get the Z values. We don't use Median for Sigma calculation. Instead we use Mean. Mean, Standard Deviation and Specification Limits (LSL and/or USL) are used in case we have continuous data (something that is measured like time, length etc.). In some organizations, quality % is considered as continuous. If that is the case, then your approach can be used. Use the mean quality score for last week, standard deviation in quality scores for last week and LSL as 98% (in quality metric, we usually have a LSL or a minimum quality as USL is 100%). Glad that you are implementing the learning. Good luck with publishing the Z scores.

You could use the template from the below IEEE link https://www.ieee.li/tmc/traditionalhoq.xlt

Hi Veeranaga It is like jumping grade 5 and going straight to grade 6. There are some who can cope up with the jump and then there are those who want to take it one step at a time. Similarly, going directly for Black Belt is possible, but it will involve extra effort from one (brush up the basic knowledge and then do a lot of self study in order to understand the GB concepts as GB concepts will not be covered again in a Black belt course). So even though doable, i do not recommend it as I believe the foundation needs to be strong and GB is the foundation for BB and MBB.

Hi Senthil You may or may not split the process into multiple modules. The decision depends on the following parameters 1. Metric that has to be improved 2. Scope of the project (should include the modules which directly work or affect the metric) 3. Level of the project (GB or BB). Usually BB projects are more complex and cross functional in nature

Hi Senthil Green Belt is a prerequisite for Black Belt training. Green belt forms the basis for mode advanced topics that get covered in Black Belt.

Hi Bheemannadora - you may use Kano model to capture the customer requirements / stakeholder needs (VOC). You may also use it to identify the need for new machinery/tools/processes in case there is a gap in the current product features and customer needs. However, you will also need to use a tool (like QFD, CTQ drill down or NGT) to arrive at specific process / product parameters (what we also call engineering characteristics) that fulfill the customer needs

Kano Model - is a product development theory which categorizes customer requirements with respect to product features into below five segments 1. Basic (must be quality) - as the name suggests, these are the basic expectations of a customer. If present, the customers are neutral, but if absent, it would lead to customer dissatisfaction 2. Performance (one dimensional quality) - features that result in satisfaction if present and dissatisfaction if absent. These are the stated needs of the customer 3. Excitement (attractive quality) - features that result in satisfaction if present and neutral if absent. These are the delighters for the customers. These needs are usually unspoken and are the differentiators for competing products 4. Indifferent - features that the customer is indifferent to as their presence or absence neither causes satisfaction nor dissatisfaction to the customer 5. Reverse - features that result in dissatisfaction if present and satisfaction if absent. These are the features that should not be present in the product Kano model is used to assess the product features on a periodic basis as the customer demands keep on changing. E.g. something that is an 'Exciting' feature today, will eventually become a 'Basic' feature tomorrow Nominal Group Technique (NGT) - is a problem solving tool that is used to generate and evaluate possible solutions. It is called 'Nominal' as it limits the interaction between participants in the initial stages as they are expected to write the possible solutions on a piece of paper. Once all solutions are written, the ideas are collected followed by a discussion on the evaluation of ideas. Usually voting is done to select the best solution. Sometimes multiple rounds of voting may be required. Using the KANO Model, we will get to know the customer requirements with respect to product features. The required features could either be present or absent. Then, NGT could be used in following ways 1. If the feature is present: to identify solutions to keep the variations and cost of the feature to minimal 2. If the feature is absent: to identify how the internal processes need to be modified / strengthened or what new processes have to be designed to be able to provide the new features

There was an accident in the recently concluded German Grand Prix where a wheel did not go correctly and it flew and hit a cameraman. Given that the wheels need to be replaced during the flash pit stops (close to 3 seconds), it needs to be done with 100% accuracy and precision. On the face of it, it might appear that it was a manual error by the mechanic (the wheel gun slipped from his hand), but in the fast paced life of F1 safety of crew and drivers is of prime importance. This lead to the identification of a design fault on the wheel gun and Red Bull team will be revising it to ensure that such an accident doesn't occur again. An excellent example of how an RCA should be performed. Full article below: http://timesofindia.indiatimes.com/sports/racing/top-stories/F1-Red-Bull-change-pitstop-procedures/articleshow/21022022.cms

Unfortunate but it's true.....every industry has an excessive focus on automations......even in service industry there are examples where automations have not yielded the desired benefits...

The difference lies in the definition iteself. DPMO (Defects per Million Opportunities) talk about the defects while DPPM (Defectives parts per Million) talks about the defectives. Take an example: A car manufacturer produces 1000 cars with each car having 50 checkpoints. Hence Units - 1000 Opportunities - 1000*50 = 50000 Assume, 10 cars are defective while the total number of defects are 250 (i.e. these 10 cars cumulatively failed on 250 checkppoints). Hence Defectives - 10 Defects - 250 DPMO = 250/50000*1000000 = 5000 DPPM = 10/1000*1000000 = 10000 Typically in a service industry, Customer is more concerned with DPPM. For a service provider it makes more sense to look at DPMO. Even if they start working with DPPM, they will eventually have to drill down to the defects that are rendering the service as defective. Not sure if the same logic applies to manufacturing, but even though DPPM is an excellent metric to track the performance of the manufacturing unit, for any improvements you still need to look at DPMO.

Hi I'm looking for a list of metrics that will be applicable in a BPO / back office operations environment. Does anyone have it ready? Pls share with me at mrgupta.mayank@gmail.com thanks Mayank Gupta

Background - There are 15 different case types that we get in one process. The total volume for the process is summation of the volumes of these individual case types. A project requires me to check whether the volume mix (over the 15 case types) has been consistent or not as compared to a baselined volume mix. What test can be used?