shalu1512

Hi Everybody, In an Out Bound Calling Center, where the Calling agent is required to pitch for selling the products to a customer over the phone, how should the productivity be defined? Should it be either of the following or a mix or something else: 1. no. of calls made in a shift per caller 2. no. of leads generated per day 3. ratio of no. appointments sought by field force/ total leads generated 4. Final Conversion rate 5. Revenue generated on conversion 6. Total Talk time/ day Your views invited, Shalini

Although Manian has left very little for others to pitch-in yet I would still sneak in and say that the organisations exist due to customers and through them they make money which is there ultimate objective so keeping customers happy and satisfied would act as a lead measure to control and achieve the lag measure of making money. Its proved that for any improvement journey if the thrust is on lead indicator, the lag indicator is taken care of to certain extent. So in short it needs to be established what is lead and lag measure for an organisation and then take a call.

Hi Everyone, My take on this would be that in order to establish and prove my capabilities as a Six Sigma professional I need to do some back ground work: 1. First and foremost the Vision/Mission/Short terms goals of the organisation need to be understood. 2. Then look at the existing dashboard of the business and see how relevant and appropriate the metrics are in order to achieve the stated Vision/ Mission. Identify whether the critical pain areas are getting addressed in some form or the other. 3. If there exists some mechanism to address this what is the scope of further bettering it in terms of methodology being followed/ the benefits achieved and the time frame to achieve the results. 4. With this homework I would carry out an alignment session with stakeholders for showing the big picture, showcase the benefits of following a structured approach like SS Vs unstructured approach (adhoc/ fire fight), sharing the prospective of capability building in this regard at various meaningful stages. 5. Target some quick wins using Six Sigma tools and attain some credibility in the system. After this I guess the way forward would be much easier. May now someone validate this hypothesis pse? Ho: The theory is theoretical and won't be able to prove the capability and Ha : The above theory will help in demonstrating the capability Regards, Shalini

OR the other way is to speak to Principal/ Chairman to identify the pain areas of college which they would like you to address OR look at the vision/mission statement of the college and see if the current metrics are appropriatley linked with vision/mission or the dashboard itself needs some changes... Views are welcome, Shalini

Dear Syed, Six Sigma is a philosophy that can be applied anywhere. Wherever an organization exists there has to be a team of people existing to operate and then there have to be some processes running to deliver the required results and if all of this exists than the scope of continual improvement is definitely there. Thereby scope of Six sigma or any other Quality methodology implementation! In your case you can begin with looking at various metrics which are critical for College Chairman or Principal like: no. of students passed out with distinction per year no. of placement companies visiting the college/ yr students getting placed through campus Professors/ lecturers/ students winning some national or state level college awards. Or some other metrics like: The ratio of students/ teaching staff. Absentism of students. Lib health in terms of visitors/day. Other infrastructure facilities like canteen, various labs...its equipment etc. Attrition of staff Based on above-suggested metrics, select one or two as project Y to improve upon, look at the data pattern, study its 'As is' process, scope it out accordingly and kick start the journey ... I hope it gives you some drift. Happy & swift take off, Shalini

Just did some search and found out about predicted R sq which goes like this: Predicted R-squared is used in regression analysis to indicate how well the model predicts responses for new observations, whereas R-squared indicates how well the model fits your data. Predicted R-squared can prevent overfitting the model and can be more useful than adjusted R-squared for comparing models because it is calculated using observations not included in model estimation. Overfitting refers to models that appear to explain the relationship between the predictor and response variables for the data set used for model calculation but fail to provide valid predictions for new observations. Predicted R-squared ranges between 0 and 100% and is calculated from the PRESS statistic. Larger values of predicted R-squared suggest models of greater predictive ability. For example, you work for a financial consulting firm and are developing a model to predict future market conditions. The model you settle on looks promising because it has an R-squared of 87%. However, when you calculate the predicted R-squared you see that it drops to 52%. This may indicate an overfitted model and suggests that your model will not predict new observations nearly as well as it fits your existing data. So, with this above explaination, can I infer that it is always wise to check "precicted r sq" value than a Adj R Sq?

well let me make an atempt, The assupmtions of OLS regression are:- 1. The model is lenear 2. The data is random 3. The expected value of error is 0 4.The residuals are independetly distributed and same varaince. Coming to 4h question: R sq gives a measure of how well the regression line fitting the real data. It lies between 0 and 1. 1 indicates the best model, able to predict the trend. whereas Adjusted R Sq is a modification of R sq. Unlike R Sq, the adjusted R Sq increases only if the new term improves the model more than would be expected by chance. It will always be less than or equal to R sq. Thats all from me..... Over to you now! R Sq S (Regds resumed , Shalini)

Hi Shantanu, Seeing these questions i feel i have still not earned my Black Belt. All the above (except 4th lto certain extent) looks unfamiliar and out of syllabus. Request you to open the suspense now by answering these questions pse. Rgds, Shalini

Dear Senthil, How is the check done on domain URL? if its done manually then there still remains the possibility of errors but if its done by the system and the system warns/ stops the user from further actioning then probably its a warning mistake proofing. I have recently intoduced mistake proofing in a system where the operators were capturing the information from one system and typing it in some other system as this 2nd system doesn't allow pasting of copied data and thus there were occurances of making typo errors. As a part of mistake proofing now the 2nd system is made to talk to the 1st system so that the data can be pulled out from 1 and be auto filled in 2nd and remain deactivated, there by not only reducing the no. of typo errors zero but also the overall TAT of processing a query.

so is it safe to infer that any process having only common cause variation is accurate but not precise? and also an accuarte but not presise process will have a good Cpk but poor Ppk? 'M I linking too many parameters at the same time? request you to clear this fog pse !

Hi Kiran, 1. Would like to understand what is the objective of doing above hypothesis? why the length of stem is critical? is it the sales/ reveneue that would be generated or which plot is more fertile or which fertilizer has a better yield? 2. well, if your hypothesis is to check the impact of fertilizer type on stem length then the conclusion (seeing p value =0.03 with significance level .05) should be " Fertilizer type has an impact on stem length" and not "the mean length of stems has a variation". 3. Yes, in order to have a narrower Confidence Interval, the sample size needs to be increased but then we really need to know the criticality of judging a process capability based on sampling. a. if decisioninig needs to be precise and can't afford error then we definitely need statistically calculated sample. Moreover in space shuttle or nuclear reactor kind of cases the overall poulation itself is finite and there sample would be very close to population (ofcourse in simulation world) b. But if risk of error can be taken then a more practical way of sampling can be taken which makes a business sense views are welcome ..... Have a great day!

Thanks SJ. will go through it and revert in case of further queries. regds

Good evening everyone, Conceptually i'm clear on the difference between accuracy and precision. However, I wonder in which all situations a process can be accurate but not precise? Is it equivalent to state that if a process is running within the control limits but still not touching the target? ie a process has a good Cp Short term but poor long term? Some pratical examples/ scenario from service industry are welcome. Regards, Shalini

Dear All, I'm keen in knowing if anyone has applied DOE in Banking sector? and how? knowledge sharing would be highly appreciated. Regards, Shalini

Dear SJ, Request you to provide some more insight on sequential test method. What is it all about? in terms of its Application/benifits/ disadvantages. Is it something similar in concept of using Frcational Factorial DOE rather than a full factorial DOE in Improve phase?

Dear Arun Maruthi Selvan Ramasamy , Kindly refer the response posted by VK dated Sep 8th and thereafter, on a very similar query. Although No. of errors per drawing can have "n" values so treating it as continous will be ok however as numerator and Denominator are discrete in natutre so the ratio of the two will still remain discrete. The data that passes the normality doesn't necessarily mean that its a continous. The p value is more than 0.05 as you have n possibilities of the above defined ratio ie the two different tests that you have done suggests that you are actually treating a discrete data as continous. Thats perfectly ok! Your 3rd query's - for continous data use parametric tests and for discrete perform non parametric tests. I hope 'm making scense.... regards

.....measure defectives from DPMO? because I'm fully aligned with what Kiarn has pointed out that a defective product can have many defects! and by assuming defect = defective by some people (as mentioned by SJ) then there is also a risk of deflating the sigma value of the process! The process might not be that poor as sigma value w'ld indicate. Coming back to my query: For eg if DPMO = 20000 (known) and the process is worked on defectives then is the question framed correctly or its being assumed that in absence of defects, DPMO can also stand for Defectives Per Million Opportunity? Tks, Shalini

To my mind, even if the Financial metric depends on external market factors, there is no harm in taking it as a project metric if the business demands that! However in such a case, rather than taking revenue enhancement/ generated as project metric, cost saved can be taken as project Y. Going by a similar example as Suresh has given - If no project is done, assume the revenue increases by 2% and cost by 1% (under stable market condition) If the project is done to improve process Cpk, the increase in revenue observed by 3% and cost by 3% (under dynamic mkt condition) Then there would be a 3rd condition (under dynamic mkt condition) If no project is done, the revenue increases by 3% and probably cost by 5% So, benifit post project completion would be the on the cost saving by 4% rather than margin earned on revenue generated. Request views pse! Shalini

SJ, Does that mean, if only defective is known, it is assumed that defect per defective = 1?

Hi Suresh, In above case is it advisable to see Ppk also? as it can throw light on Process capability over a long term? ie whether the process change will bring in the benefits over a long term?

Shukriya! Now 'm crytal clear.

Hi Rajendra, Let me pitch in here:It all depends on how you are looking at the data? 1. If data on attrition is taken in isolation ie just the count of pepole attrited its discrete but as it doesn't make much sense, we need to see this data over something. It could be over a period of time which make it continous but if the same data is seen over the total no. of people hired then its discrete in nature. VK seek your comments pse.

Hi Anish, Wanted to know how is calculated FCR% in your call center? is it on just per day basis or ratio of total calls resolved per day on total no. of calls received / day? depending on this FCR data can either be treated as continous or discrete. Another point that i wanted to understand is why you need to relate FCR with Sr Management Escalantions? as all NFCR must not be getting converted into escalations! May be FCR% can be related to RC% (Repeat Call) or response from a customer in a survey conducted on a sample basis.