Anusha Vemuri

Obviously yes, because project execution experience gives more clear picture of the fruits what we are enjoying from a good project. As some of the people are debating that six sigma is a set of tools where we can use them for solving problems in bits and pieces. But a whole Some project gives idea how to use those tools combining each other and coming to a fruitful conclusion.

Tribal knowledge: Unwritten rules which contributes to overall quality of the product in a company which are known by a group of individuals within an organization but not common to others. How to Capture Tribal Knowledge: 1. Knowledge identification 2. Identifying persons with tribal knowledge 3. Investment on training by knowledge people 4. Retaining persons with tribal knowledge and plan to transfer that to new generation. Unlocking: 1. Raise awareness of the problem, symptoms, it’s causes and outline a solution. 2. Arrange suggestion box in the companies 3. Rewarding best ideas and informations 4. Making the speak up culture popular Harnessing: 1. Capturing such information in standard formats 2. Educating the people on tribal knowledge by awareness sessions 3. Recording the captured knowledge for new employees

A group of units produced under the same set of conditions is call as rational subgrouping. It is used to represent the process with same conditions. The selection of rational subgroups should be consistent with the structure of the data from the process. What would an excellence practitioner lose if he does not utilise the concept of rational subgrouping in the pursuit of process improvement? 1. Variation in same groups cannot be identified. 2. If variation is not identified then minimizing the variation is difficult. 3. Stability of the process is questionable. 4. Successful control charts cannot be plotted 5. Differentiation of common causes and special causes are difficult 6. Process changes or process behavior cannot be studied over time 7. Monitoring of the process improvement and its efforts are not possible. 8. Cannot predict unexpected changes in a process

1. First and foremost, Sponsors has to been trained on Six Sigma as a metric and methodology. 2. As the project sponsor is responsible for project charter, he/ she should have complete process knowledge and should be proactive in approach.. 3. Sponsor should be able to fix the metrics for project based on their expectations. 4. Before handing over to process owner, sponsor is a person who acts as surrogate process owner before handover. 5. Sponsor should judge the project for correctness and approve it accordingly. 6. Sponsor should provide the right resources and appropriate timelines 7. He/She should be capable of evaluating the analysis of problem identified. 8. Every stage of improvement project D, M, A, I, C should be approved by the project sponsor 9. An understanding of the common sense, systematic and objective approach is required for the sponsor in every step of project success.

It’s a rare case where the baseline and performance after are not comparable. This may happen due to some special causes which occurs during the execution of project. Some examples are 1. Consider our project is TAT reduction of call resolution time taken. We have calculated base line by considering the data available for last 3 months. After implementing the changes proposed and calculating the performance at the end shows drastic increase in the TAT instead of decrease in the time. Reason: Base line was calculated during the festive season and when the volume is less. Performance after improvement was calculated when the volume of calls is very high nearly double the numbers. For getting better results the data collected for baseline and after the performance results should at least match. 2. For” A” product no. of complaints are 45 in 2016, we want to decrease the number of complaints and by six sigma tools DMAIC project was executed and after implementing the changes the tendency of no. of complaints receipt increased to 60. There are some special causes to be back verified in the improved process where the product is doing wrong. Measuring differences in calculating the baseline of the product also lead to performance after improvement is not comparable with performance before improvement.

zero rework is impractical or undesirable in some situations. If we take pharma production as example some cases like 1. As per the market requirement and forecast if a particular product "A" was manufactured and packed for USA and the requirement got changed after packing the product that the same product "A" is urgent in Canada. Repacking of the product as per the canadian regulations has to be done. And repacking is allowed as per GMP practices if the written procedure for repacking is followed. Then again the product has to be repacked and sent to canada to meet the market requirement. In this case zero rework is impractical because if we doesn't meet the Canada requirement we may loose the market share and the business will get effected.

As per my opinion Descriptive and prescriptive analytics areas are well covered by lean six sigma community well. We need to work more on Predictive analytics area of business analytics areas. Hope this area will be covered and well explored by business analytics rather than six sigma.

As per Kano model basic needs, performance needs, and excitement needs are three different attributes which have an impact on customer satisfaction. These three shows business impact if they are fulfilled or not fulfilled. Kano model applies to products, services, business process and also software. Basic: Basic needs are must be’s. These are customer expected requirements, should be or must be available in the product or service offered. Customer satisfaction is neutral when done well, but customer satisfaction is very dissatisfied when done poorly. These needs are the basic price of product or service offered in the market. Examples: Car should have wheels, steering, seats and doors Performance: Performance needs are like uni-directional. These are the requirements that the customers are at the top of the customer minds when making choices and evaluating options to buy a product or choose a service. The better they perform the more satisfaction they bring, same in the reverse way when they perform poor customers are dissatisfied. Examples: Car should give good mileage as per the claim. Excitement: Excitement needs are surprises which delights the customers, they are unexpected. These are customer unexpected requirements, so the presence of excitement needs delights the customer, but do not cause any dissatisfaction when missing because the customer never expected them in first place. Example: Automatic whippers on when rain falls. Opening of airbags when something hits car. Once understood for a specific product or a service, what would be your approach for putting these needs to good use? Follow these steps to put these kano model needs to good use 1. Every product should have some basic needs to enter into the market. So, first study the basic needs and design the product or service in such a way to meet the requirements of customer basic expectations. 2. Basic expectations are required to stand beside the other products in the market. List down the basic requirements which should be full fill every time and we should not be poor in implementing the basic needs. 3. If basic needs fail, total product or service fails. 4. Check for performance attributes which increases the customer satisfaction. The attributes should perform well always because if they did not which leads to decrease in customer satisfaction. 5. Performance attributes should be choosen carefully and make sure that they perform good. 6. Excitement attributes are manufacturers choice. If they increase more they will attract customers. 7. But manufacturer should be carefully in choosing that as it should not increase the price of the product and go beyond the limits which is again hit to the business.

Probability of rejecting null hypothesis when it is true is statistical significance i.e the chance of error allowed is significance level. It is denoted by greek letter “α” · P-value is used for decision making in a hypothesis testing. · If P value is less than or equal to a predetermined level of significance (α level), then null hypothesis is rejected and alternate is claimed. · If P value is greater than the α level, then null hypothesis is accepted and alternate cannot be claimed. · 0.05 is the acceptance level of Type I error, thus any p-value less than 0.05 means we reject the null hypothesis. · If α is small, incorrectly rejecting the null hypothesis chances are less. · If α is large, incorrectly rejecting null hypothesis chances are high. Thus choosing α level is very important for a data to check whether it is statistically significant or not to proceed further.

S.No Stable process Capable process 1. Process operating within specification limits Process operating with in control limits. 2. Variation is more Variation is less 3. Process is within specification limits Process is in statistical control 4. It meets customer needs It meets customer and business needs 5. First step to go for process improvement If process is stable then only process capability should be performed. 6. All stable process are not capable processes But all capable processes are stable processes 7. No prerequisites for stable process Pre-requisite of the process capability analysis is a stable process Is Process Stability supposed to be a pre-requisite for all type of processes? Process stability is required for all the quantitative data of all types of processes. In a process, every parameter or every item which gives quantitative data will have specification parameters in place. So the process should perform stable before going to capability of process. Process capability assessment should only be performed after first demonstrating process stability.

Correlation is finding a relationship between two or more sets of data. It measures the strength between the variables whether they are strong, moderate or weak and also the direction of relationship i.e positive or negative. To find the correlation between the variables they should be independent which are not impacted by changes to other variables in a process. So, independent variables shows observed variation. If the absolute value of the correlation coefficient is greater than 0.85, then there is a good relationship. Correlation does not prove the cause-effect relationship between two variables. Why do we still use it in root cause analysis? Yes, correlation is a good tool to know the relationship between the independent variables which are found or listed during the investigation. Usually after listing down the probable root causes select the independent variables and conduct the correlation which shows whether the probable root cause has positive or negative relationship with the problem. In such a way we can use the correlation tool to find the appropriate variable which is leading to the cause.

The Voice of customer (VOC) is the starting point of any project and data collection plan. How the Voice of customer is collected is also crucial for business to take it forward. Following are the few questions to be asked before collection of VOC 1. What are the elements of your business that are the most critical, from the perspective of your top or best customers? what are their relevant needs? 2. What data has been collected to understand the customer requirements? 3. How do you operationally define the defect from the perspective of the customer? under what conditions does it occur? Determining customer requirements: 1. Review customer comments and comparitive data 2. convert into terms of process performance 3. Describe actual customer requirements - write the requirements, use measurable terms, identify performance targets. Can overemphasis on VOC be detrimental to business? Explain with examples I can say obviously yes, overemphasis on VOC be determental to business. Below are some examples for this statement 1. Especially in Healthcare field, VOC is collected to a product of huge sales. VOC results shows that if the type of packing for medicine is an issue that the patient is unable to see the product clearly out because of amber colour blister packing/aluminium foil packing. By considering the VOC company cannot change the packing because it will effect the stability of product. It will impact the potency of medicine and also hits the business to huge loss. So, VOC should be considered where the quality of product is not effecting its potency. 2. This example is also from healthcare industry, One product which is used for Diabetes getting repeated complaints because of the big size of tablet. By considering the large number of complaints we cannot alter the quantity of drug to be required by the body for one single dose. And also the patients who require lesser dose can take the half tablet directly because of its bigger shape. The formulation type is also extended release, which patient can manage whole day with single tablet dose. In such a way company is managing single dosage of tablet in place of two varieties of dosageforms. Which saves huge to the company.

Continuous data: Continuous data is information that can be measured on a scale or having limit for measurement. It can be measured and broken down into smaller parts and still have meaning. Attribute data: Attribute data, also known as discrete data, are counted in whole numbers or integers. The result will always be a whole number—never a decimal fraction. It is also known as count data. Any type of data collected is for analysis Examples for confusing datasets 1. Sizes of clothes 24,28,30,32,34,36,38,40,42,44 etc. No. of clothes available in the particular size is attribute data and measuring 10 pairs of clothes and they should have size ranging from 24-32 is continuous data. 2. Age of children in group I is 4,2,8,5,6,4,7,2,3,6,4,7,5,3,5,1,2,9,7. This is attribute data. If the same data is having a limit of age group and the number falling into that is continuous data. 3. No. of links in a chain is 100,102,105,100,101,103,108,105,106,102,105,107 is attribute data. Different lengths of chains is a continuous data.

Correction: It is the step when you fix the thing that went wrong (the nonconformity) This is the immediate action to keep customers or management happy. Eg: Putting out fire when fire accident happens immediately. Corrective action: It is the step taken to prevent the reoccurrence by removing the causes of non-conformity. Eg: Analyzing what caused fire and how to prevent the fire after the occurrence. Preventive action: It is to prevent the occurrence of potential nonconformities or undesirable actions. Preventive Action have to be identified and apply the preventive action, against the potential non-conformities, risks, defects or non-compliances. Eg: Stopping fire from happening not only in that location but made applicable to similar locations and areas where the threat is more. Are there situations where both preventive action and corrective action are undesirable and correction is the only preferred action? Ideally in every situation of daily life or in work, corrective action and preventive actions are applicable and useful. The above sentence can be written in vise versa like correction of each and every situation may not be possible but we can apply corrective action and preventive actions to that. Eg., Natural disasters like volcano eruption, Cyclones, Tsunami etc creates a lot of disaster, in such situations we cannot correct the situation but we can take preventive and corrective actions to decrease the effect of calamities. Situations where we only correct it and cannot prevent that leads to repetition of the situation again and again, which is not acceptable.

Check sheet is a effective lean QC tool to collect the real time data from the accurate location. Check sheets are famous as one of the effective tools in 7 QC tools. It is sometimes referred to as a concentration diagram. Check sheets are useful in all phases of lean six sigma DMAIC phases.Check sheets should be used appropriately when the data can be observed and collected repeatedly by same person or from same locations. But now a days as the data collection and analysis of data are automated the use of check sheets are less when compared to earlier. Check sheets have now become obsolete. They have been replaced by modern day Business Process Management software. This has enables more complex data to be automatically recorded. The process now depends neither of the intelligence of the human nor on the reliability of the check sheet. Data is now automatically recorder and can be arranged in whatever manner required in a few clicks. Many software even produce the data in a ready to use graphical format enabling further convenience for the users. BPM is useful for only the data that can be collected online and also automatically. Check sheets will not get replaced by BPM for manual data collections. They can be modified and used according to the requirement. Correct usage of check sheets is important in manual data collection. BPM cannot replace the check sheets usage in lean world. But they may acquired other shape for correct usage.

1. Pull System :- Manufacturing system in which production is based on actual daily demand (sales), and where information flows from market to management. 2. Push System :- Manufacturing system in which production is based on a projected production plan and where information flows from management to the market. Pull system and push system are majorly used to maintain the inventory levels which in turn impacts the business. Pull system is always producer centric while push system is customer centric. Pull system is not always successful in every industry. Sometimes customer centric is also important for the business to grow. Examples: a) In case of medical care, sometimes push system should be implemented for essential medicines which are widely used in disaster cases occurs due to natural calamities etc, the essential medicines should be available readily with supply chain forecast to serve the needy. In this case pull system is not that successful as the patient needs the medicines urgently and if this is not forecasted leads to loss of lives. Push system is used in food supply business like hotels, restaurants etc, if the customer demands for a specific food variety and it is not ready will hits the business. Push system always requires a good forecasting system or else the inventory costs goes up which impacts business.

Analysing COPQ (cost of poor quality) is always a worthwhile for every process. Measuring quality is measuring the requirements to get the job done right the first time out. Poor quality is a form of waste that is not an integral part of a product/ service. It does not add value for shareholders and customers. Cost of poor quality is a concept that helps quantify the impact of process inefficiency and waste has on internal and external customers. It helps quantify the impact of not fixing any process and tool that touches both internal and external customers. Measuring cost of poor quality is easier to identify when it comes to processes that directly impact external customers. Most systems report out metrics around external quality issues that lead to lower customer satisfaction. COPQ increases as we move from Prevention to Rejection. The goal is to invest time and money on defect prevention to eliminate as much as possible Inspection, Internal and External failure costs. Measuring of COPQ can be divided into 4 categoroies. 1. Internal failure cost – Defects corrected internally before release to customer 2. External failure cost – Defects that reached customer 3. Appraisal cost – Measuring /monitoring activities related to quality 4. Prevention cost – Training /educating personnel to prevent errors from occurring. So, If we can control the COPQ the cost incurred on the product and production can be reduced leading to the decrease in product cost with increase in quality which satisfies the customer.

I would prefer a false alarm rather than a missed alert. Because False alarm: This is any small shift in the process behavior will be detected (that’s good) but also you will have too many points outside the limits just by chance when there is no special cause to address (that’s bad). That is called a false alarm. Missed alarm: Missed alarm is that we have already missed the special cause which is suppose to be detected. Already the loss happened, cannot be recovered. So false alarm is better than missed alarm.

S.No Central limit theorem Law of large numbers 1. Sample mean will be normally distributed Sample mean equals to population mean 2. Can be executed for less data Can be executed for very large numbers 3. It will not converge to a number, it converges to a normal distribution. Sample mean will converge to a Number 4. Gives value and distribution of sample mean for n numbers Gives value and distribution of sample mean for n numbers 5. The standard deviation of the sample means equals the standard deviation of the population Standard deviation declines as the size of population or sample increases When data is not normally distributed, the cause for non-normality should be determined and appropriate remedial actions should be taken, without that we cannot change every non normal data to normal data by using central limit theorem. 1. Too many extreme values in a data set will result in a skewed distribution. Normality of data can be achieved by cleaning the data. This involves determining measurement errors, data-entry errors and outliers, and removing them from the data for valid reasons. 2. Round-off errors or measurement devices with poor resolution can make truly continuous and normally distributed data look discrete and not normal. Data errors should be taken care to change the non normal data to normal data

Fault tree analysis: A technique by which many events that interact to produce other events can be related using simple logical relationship. It is effective in Qualitative description of potential problems, Quantitative estimates of failures frequencies and helps to prioritize issues, Limitations: Fault tree diagram construction is very tedious for large systems and also it is a back ward method which is used to think about consequences which may occur.

1. A value adding activity is one that customer is willing to pay for. - Yes this should be modified in pharmaceutical industry because all the activities during manufacturing of a product are value adding but we have to perform necessary non value adding activity to complete the manufacturing of end product 2. A value adding activity is transformational by nature. - Natural value adding activity are not common in every industry.

Does 5S actually save enough money when implemented outside manufacturing? If yes, how? Obviously the answer is yes, typically the 5S refers to Sort, Set in order, Shine, Standardise and Sustain. These terms not only applies to manufacturing facilities but to service field also. S.No 5S Application Result 1. Sort Sorting the necessary process steps or required documents in a particular process (required and not required) Required steps to execute the process gives better results 2. Set in order After identifying only, the required steps, arranging them in proper order. Promotes efficient work flow. 3. Shine Remove the unwanted data and unwanted stuff in the process with proper approvals Creates clean and understandable work flow 4. Standardize Set standards for the rearranged process Standardization leads to maintaining the process flow consistently. 5. Sustain Maintain the proper work flow after applying the above steps Good customer satisfaction without delays

Definetly in the world of product innovation, Lean six sigma creates its own place by moving forward with DFSS (Design for six sigma). DFSS is a powerful set of tools and techniques that can help you develop products and services that truly satisfy your key stakeholder needs. By adding this toolkit to your repertoire, you can demonstrate to your current or future employer that you understand how to not only reduce the bottom line costs using DMAIC but also develop robust products and services to increase top line growth using DFSS.

Kaizen Kaikaku Kakushin Continuous improvement everyday, everyone and everywhere Revolutionary change to existing situation Innovation – creating new Small incremental changes Radical change Complete departure from current situation by innovation and transformation Eg: Reducing production time, implementing 5 S Eg: Introducing a new improved part in place of heavy part in machinery Eg: Innovating new system in place of old system These are 3 essential K’s for organizational success: Kaizen, Kaikaku, Kakushin. The concepts of Kaizen, Kaikaku, and Kakushin is the right solution for any organization who wants to deliver a sustainable high-quality product or service and to ensure a steady growth in the market.

If you can't measure it, you can't manage it – its true every process and every procedure should be measurable. Dashboard visualizes the Key Performance Indicators and other strategic data of an organization at a glance. Numbers are used to track KPI’s, metrics and other key data points. "It is wrong to suppose that if you can't measure it, you can't manage it. The most important numbers are unknown and unknowable" - Dr. W. Edward Deming – Deming’s sentence doesn’t mean that we have to ignore the numbers, metrics and dashboards, they show the common causes variations. He mean to say that special cause variations also lead to process variations and CTQ’s. Considering only numbers and ignoring the special causes may not show successful process improvements or exact root cause findings.