Anirudh Kund

Effectiveness means the ability to produce a desired result that is; getting things done right regardless of time and cost; whereas efficiency means doing the things right in a best possible or optimised way. Example 1: If illumination intensity 600 lux is the requirements in a study room , you have several option to get the desired results, one may use CFL or LED light. In this case CFL is effective but not efficient as it consumes more energy for same light intensity compared to energy efficient LEAD lights. Example 2: In a beverage manufacturing unit ,day production target is to produce 10000 cases and if you use process A ,it produces 13000 with 5% deviations where as Process B produces 9000 with zero deviations. In this case Process B is efficient as it is cost effective due to zero waste but not effective as it is not meeting the required results.

How can specification limits be decided for characteristics of an innovative new product with which customer neither has an experience nor an expectation? Assume that the company creating such an innovative product does not want to expose it to the market or customers before launch. Asking foundational consumer questions is a good starting point when translating a concept into specific product attributes and guardrails. The objectives of this research are to Establish the absolute must-have product attributes in order to deliver on consumers’ expectations set by this concept. Create clear product and package guardrails, within which further product development should occur. Prioritize potential process or formulation options by obtaining consumer feedback on early product or package prototypes, and Explore with consumers any potential areas for further concept optimization. Though we know that New Product development passes through 5 stages like Idea Concept, Initial Feasibility, Development, Scale Up, and Lunch & learn. However in the above condition the Organization wants to lunch an innovative product but does not want to expose to the market or customer before lunch. In such scenario, following questionnaires may help in Idea Concept phase … a. What is the target consumer experience and emotional cues? b. What are the key "must have" attributes? Is there a priority? c. What are the key "delight" attributes that should be included if feasible? d. What are the key differentiator attributes that uniquely drive marketplace differentiation? f. What will make this product most usable for the consumer? e. 5 Senses Attributes: Sight / appearance h. 5 Senses Attributes: Smell / aroma i. 5 Senses Attributes: Sound j. 5 Senses Attributes: Taste / flavor k. 5 Senses Attributes: Texture / mouth-feel l. What are the conditions under which this is likely to be consumed/used j. Regulatory considerations for product or ingredients – Domestic/International K. Product manufacturing platform L. Can we use existing equipment, or modified or new equipment is needed? m. Packaging requirements

Every product or services passes through multi steps before coming out as an output. In each and every step value gets added and the efficiency of every step may differ. If a process consists of multi steps, any non-conformances/ defects in any steps will have impact on the final output. If you would like to find out the quality of the entire process or the probability of defect free output of the process is called rolled throughput yield (RTY). RTY is the proportion of conforming/defect free units that results from a series of process steps. Mathematically, it is the product of the yields from each process step. Consider a process is having 3 steps and each steps yield is as follow… Step A = 90% Step B = 91% Step C = 99 % The exact yield of the process will be 81.081 % (It is the product of each step yield) If the RTY is more, the process is more efficient in terms of producing defect free products. As per the question, whether the process having 100% RTY can be considered as inefficient. Since 100% is the highest rolled throughput yield, hence the process is more efficient.

What is Failure Mode Effect Analysis (FMEA)? The FMEA is an team oriented analytical methodology which aims for identifying and correcting potential failures in a systematic and structured manner The FMEA is already implemented in an early phase of the product engineering process and tries to identify imaginable/possible failures and implements corresponding corrective actions. The FMEA aims for reducing the risk of possible failures in the best possible way. The method records the generated knowledge to the advantage of the organization and integrates the lessons learned into future tasks and scopes. The FMEA is living documents that must be reviewed and updated whenever the process has been modified There are three basic cases for which FMEA process is to be applied…. New designs, New Technology or New Process Modification to existing design or process Use of an existing design or process in a new environment, location or application usage on the existing design or process. In Six sigma DMAIC methodology, Initially, FMEA is done in Measure phase and provides recommended actions for the team to minimize risk to the customer. Its revision continues up to the Analyze and Improve Phase to ensure that the evaluation criteria and cause/effect relationships are updated with data-driven conclusions. During the Control phase, the FMEA needs to be updated to reflect the final state of improved project. The information from FMEA is then summarized in the control plan documents Process of Carrying out FMEA Review the process and list the Process steps and key Process inputs – can be derived from process maps, cause and effect diagram, brainstorming sessions or existing process data. Brainstorm all the expected potential failure modes for each key process input – can be described in physical/technical terms ,not as a symptoms noted by the customer List the potential cause for this failure mode – May be several causes for such failure, here consider the operating conditions, usage or in-service, and possible combinations as potential causes. List the controls that are existing in place to detect the cause of the failure mode. Rate the severity of the failure effect the customer experiences in a Scale of 1 to 10 (1 being no effect and 10 being the Hazardous without warning) Determine how often the cause of the failure mode occurs and rate the occurrence in a scale 1 to 10 (1 being unlikely to 10 being almost inevitable) Determine the effectiveness of the current controls to detect the cause of the failure and rate the detection criteria in a scale of 1 to 10 (1 being very effective detection and 10 being no detection.) Calculate the risk priority number(RPN)by multiplying Severity, Occurrence and detection together. Note : Risk priority number (RPN) = (probability of occurrence) x (severity ranking) x (detection ranking).Highest RPN would be 1000 and lowest would be 1 , lowest RPN is lower risk. However if the severity rating is 9 or 10 ,efforts should first focus on these key process inputs to ensure detection is at least 1 or 2 and occurrence is also low. Establish the action plan which can help to minimize the risk to customer by using the RPN value –recommended action can only impact the detections or Occurrence. Severity will remain same until and unless the product/service or information is used for a different intent. Identify the person responsible to complete the recommended action with target date Once the action taken, list the specific actions taken, along with the actual completion date Calculate or reevaluate the resulting RPN, based on the completed action. After finishing the above-mentioned steps, sort the RPN numbers and identify most critical issues and where to focus first For understanding clarity, tried to describe the rating scale elaborately as below, which can be used as reference for FMEA RPN calculation … Severity Rating Occurrence Rating Detection Rating No effect -None (Unable to realize a failure has occurred.) 1 Very Low <1 /1500000 (Failure is unlikely) 1 Almost certain (Defect is obvious and can be kept from affecting the customer) 1 Annoyance- Very Minor (Very Minor on product or system performance. Minor defects noticed by discriminating customers. No disruption to product line. No effect to performances. Inconvenience to administration of customer) 2 Low- 1/150000 (Relatively low failures) 2 Very high (Very high chance of detection. All units are automatically inspected.) 2 Annoyance - Minor (Minor on product or system performance. Minor defects noticed by some customers) 3 Low- 1/15000 (Relatively low failures. Isolated failure associated with similar process,) 3 High (High Chance of detection. SPC is 100% inspection surrounding out of conditions) 3 Annoyance - Very Low (Small impact on product performance. The product does not require repair. Minor defects noticed by most customers. Major disruption to production line. Significant delays in delivery to end customer. No effect on performance) 4 Moderate -1/2000 (Occasional failures. Generally associated with processes similar to previous processes which have experienced occasional failure, but not in major proportions) 4 Moderate high (SPC is used with an immediate reaction to out of) 4 Loss or degradation of secondary function -Low (Moderate impact on product performance - The product requires repair. Reduced secondary function performance.) 5 Moderate -1/400 (Occasional failures. Occasional failures. Generally associated with processes similar to previous processes which have experienced occasional failure, but not in major proportions) 5 Moderate (Process is monitored and manually inspected) 5 Loss or degradation of secondary function -Moderate (Product performance is degraded. Comfort or convenience function may not operate. Loss of secondary function performance. Major disruption to prod line. Customer product may have to be reworked. Customer product operable. End use experiences some dissatisfaction.) 6 Moderate - 1/80 (Occasional failures. Occasional failures. Generally associated with processes similar to previous processes which have experienced occasional failure, but not in major proportions.) 6 Low (Manual inspection with mistake proofing modification) 6 Loss or degradation of Primary function -High (Product performance is severely affected but functions. Reduced primary function performance. A portion (<100%) of customer product may have to be scrapped. Customer product operable but at a reduced level of performance. End user dissatisfied.) 7 High- 1/20 (Repeated Failures. Generally associated with processes similar to previous processes that have often failed) 7 Very Low (All units are manually inspected) 7 Loss or degradation of Primary function - Very High (Product is inoperable with loss of primary function. The system is inoperable./ Loss of primary function. 100% of customer product may have to be scrapped. Customer product inoperable, loss of primary function. End user very dissatisfied.) 8 High- 1/8 (Repeated Failures. Repeated Failures. Generally associated with processes similar to previous processes that have often failed.) 8 Remote (Units are systematically sampled and inspected) 8 Failure to meet safety/regulation - Hazardous with warning (Failure involves hazardous outcomes and/or noncompliance with govt. regulations or standards) 9 Very high -1/3 (Failure is almost inevitable) 9 Very Remote (Occasional units are checked for defects) 9 Failure to meet safety/regulation - Hazardous without warning (Failure is hazardous, and occurs without warning. It suspends operation of the system and/or involves noncompliance with govt regulations) 10 Very high >1/2 (Failure is almost inevitable) 10 Almost impossible (Defects caused by failure is not detectable) 10 As we know, the latter the failure is detected, the higher will be the corrective cost and customer dissatisfaction. If FMEA is done in the early stage with right way it will help in reducing the risk of possible failures in the best possible way. While FMEA as a preventive tool is highly beneficial for the organization, but it has certain limitations too. Limitations: Team Formation: Depends on how the interdisciplinary team is formed and what is their expertise on subject knowledge (product, process) and FMEA methods – If team is good, output will be good. Team members Involvement: Whatever the subject matter expert team may be, If the team is not actively involved, a superficial look at the process will miss many failure modes. FMEA Moderator: if the moderator is not expert, competent, target orientated and biased towards certain opinion, it may derail from its objectives. Prioritization: The initial output of an FMEA is the prioritizing of failure modes based on their risk priority numbers. Mostly Severity and occurrence ratings are often difficult for individuals or teams to estimate. Time factor: This process takes time for discussion/brainstorming session Rating Scale Customization: Generic rating scale may create confusion; creation of meaningful rating scale plays a vital role in rating. Unknown Failure mode: Certain failures which is outside of the team experience, may be left in discussion and documentation. FMEA Timing: FMEA not initiated on early stage or design stage, may miss certain design /process verification Target Customer: Customer is both internal and external, but if the FMEA is targeted considering only external customers, internal process failure impacts may be missed from consideration. Product/Process: Every product /process is not similar and different from others and FMEA should be carried individually not coping & pasting from one to another. Detection Control: Assuming detection controls are more effective while in real they are not Action on Recommended failure mode: FMEA is a prioritization tool. It doesn’t eliminate failure modes or effects by itself. Organization need to apply mistake-proofing tools to eliminate the root causes of failure modes, especially true with failure modes that have very severe effects. FMEA/Control Plan Update: It is a live document, needs regular review and update if a new potential failure mode is identified, that should be added to the FMEA and control plans developed for that. The control plan tells people how to react when a failure mode occurs, if FMEA is not tied to control plan, people would not be aware how to react in case of failure. Note : FMEA Worksheet template is attached for better understanding FMEA-template.xls Rating Template.docx

Increasing competition and decreasing customer loyalty have led to the emergence of concepts that focus on the nurturing of relationships to customers. If the customer is satisfied and retained with you then, it will help you to improve your business profitability. Generally companies generate better results when they manage their customer base in order to identify, acquire, satisfy and retain profitable customers. Maintaining the customer base aimlessly will not be helpful for the organization. Not All customers are of equal importance and may not be worth on retaining at all. Some of the customers may not be helpful in the organization profitability - those are high cost-to-serve, Debtors, payment defaulters, not loyal - switching frequently between suppliers. The goal must be a better management of customer relationships across the different points of customer contact. Customer relationship management is all about retaining customers, capturing customer lifetime value, maximizing new business opportunities, and sustaining profitability Type of CRM characteristic 1. Strategic CRM : It is a core customer-centric business strategy that aims at winning and keeping profitable customers 2.Operational CRM : It focuses on the automation of customer-facing processes such as selling, marketing and customer service 3.Analytical CRM : It focuses on the intelligent mining of customer-related data for Strategic or tactical purposes 4.Collaborative CRM: Collaborative CRM applies technology across organizational boundaries with a view to optimizing company, partner and customer value. Misunderstanding of CRM: 1. CRM is database marketing: In Database marketing data used to build and exploit high quality customer databases for marketing purposes. As CRM software generates such database ,it sometimes misunderstood as database marketing, CRM is much wider in scope than database marketing, it is based on strategic, operational and collaborative approach. 2. CRM is a marketing process: CRM software applications are used for many marketing activities like market segmentation, customer acquisition, customer retention and customer development, which creates misunderstanding on CRM as a marketing process. However CRM can be used for sales and service, HR can use customer preference data to recruit & train frontline staffs. Hence CRM is more than marketing process. 3. CRM is an IT issue : Most CRM implementations require the deployment of IT solutions, some people thinks it is a IT initiative and hence IT issue. IT is an enabler which helps to gather the insights about the customer. An overarching goal of many CRM projects is the development of relationships with, and retention of, highly valued customers. This may involves behavioral changes ,where IT may not play any role. 4. CRM is about loyalty schemes : Loyalty schemes are commonplace in many industries, such as air miles, payback points etc. CRM may help the organization become more effective at customer communication and offer development. This loyalist scheme may help to retain the customer but all CRM implementations are not linked to loyalist scheme. e Let’s discuss on CRM in different Industries for different professionals with certain considerations ,which may not be limited to this list. Hotel/Hospitality: In Hotel industry the best in class hospitality attracts the customer, quick response in customer service adds happy customer too. CRM consideration would be 1. To identify frequent customers who books premium accommodation (High Profit) 2. Customers/Guest room preference on food or room type etc – smoking or non-smoking, standard or de- luxe. 3. Option on online booking and interaction facility 4. Helps to run the promotional scheme on weekend break to customers who have indicated their complete delight in previous customer satisfaction surveys. 5. For Immediate assistance, Car hire service, 6. Wish the guest on their birthday/anniversary etc. who had stayed in the hotel 7. Customer feedback surveys Telecom: In telecom industries looking at tough competition, it is required to design and develop customer centric strategies to grab the share and to sustain in long run. CRM consideration would be 1. Competitive analysis 2. Error free order entry /Billing 3. Value based customer service 4. Proactive churn mitigation 5. Encouraging prepaid customer to move to post paid 6. Customer-friendly help desk and call center Banking Availability of several banking service provider, customer used to switch one to another based on competitiveness. Retaining them to increase profitability is the most warranted. CRM consideration would be 1. Value added cross selling of other products like loan, FD etc. 2. Mobile banking 3. Online money transfer 4. Online statements 5. Toll free customer service 6. Assigning relationship manager for privileged customers 7. ATM facility 8. SMS on deposit/transfer/withdrawal Insurance: To maintain competitive edge and viability, insurance companies are focusing intently on delivering superior customer service by claim settle ratio ,attractive offers etc… CRM consideration would be 1. No claims discount structure 2. Offer of other insurance at discounted rates 3. Bonus on second anniversary of policy 4. Tips of value/other product offers 5. Offers at highly attractive terms for customer 6. Offer alternative payment mechanism Airlines: Airlines success heavily depends on its ability to intelligently manage sales, marketing and service processes and to draw mutual advantages from understanding of airline’s customers CRM consideration would be 1. The behavior of employees 2. Web Check in facility 3. Flight in time or Punctuality 4. Proper baggage handling 5. Seat occupancy Automobile: Success in automobile industries depends on understanding purchase cycle intimacy and analysis of ownership experience with a Web enabled systems making easy to capture crucial customer insights in the midst of the buying process CRM consideration would be 1. Purchase experience 2. Workshop service/ on road service 3. Vehicle performance 4. Dealer business performance 5. Behavior of the sales and service staff 6. Reminder /follow up call for service FMCG: FMCG companies have relied heavily on brand strength to generate and secure customer demand that translates into shareholder value. CRM consideration would be 1. Setting up national-level call centers to address consumer complaints/feedback 2. Launching targeted catalogues 3. Web based consumer research 4. Unique features offered by the specific brand 5. Coverage of national distribution 6. Dominance of market leadership If you look at the competitive approach of the organization due to customer awareness and expectation, then it obvious that there is no strategic alternative for enterprises rather than CRM - a relentless customer centric approach in order to achieve current, future and lifetime profitability by creating customers for life. Hence CRM is the core business strategy that integrates internal processes and functions, and external networks, to create and deliver value to targeted customers at a profit. It is based on high quality customer related data and enabled by information technology.

Before discussing the lead time vs cycle time differences and confusions, I would prefer to use few examples to make it pretty clear. Most of us are aware about the online shopping platform. Consider you place an order for “T-Shirt “in an online shopping portal today and as per the portal, the expected date of delivery is after 4 days from the date of placing the order. The moment you place the order, time starts for your order. Here you have to wait for 4 days from the date of placing the order to get the delivery. The waiting time, what customer sees, is called lead time. Lead time is the period between a new task’s appearance in your workflow and its final departure from the system. When you placed the order for “T-Shirt”, company got a demand to make the “T-Shirt” production. Based on demand company started the manufacturing process to prepare the “T-Shirt”. Cycle time clock starts when work begins for the manufacturing of the placed order, and ends when the inputs transformed to final output by the value added process. It is more into the, mechanical measurement of process capability. Average cycle time is the average time taken to produce a lot /batch/item from the process starts to end. In other word, Cycle Time = Operating Hours / items or batch produced. For example, assume the manufacturing plant operates for 16 hours per day and produced 64 T-Shirts each day. The Cycle Time is 16 / 64 = 1/4th of an hour i.e. 15 minutes. In this example, while making a “T –Shirt” takes only 15 minutes but the customer gets his order delivered after 4 days of placing the order. If you look at the workflow of lead time in this example, you will find valuable information about the workflow, some of the task is taking more time, it may be dispatch time or delivery time compared to cycle time. Whichever the bottleneck in this process can be identified and eliminated or reduced to acceptable limit to improve the lead time efficiency. Confusion: Cycle time is the time taken to complete the process, normally production process. What is cycle time is not often clear. This is because of confusion between cycle time and lead time. Cycle time is not the overall time that producing a production order takes; it is the actual time that the production works being performed. Product issues from store/warehouse or product release in system cannot be cycle time, it’s lead time. However it is not being described using lead time. People in manufacturing plant talks about cycle time, which refers to specific operation within the manufacturing process and People in supply chain managements talk about overall production cycle time, they also refer to the time of other non-production process in the supply chain, but in practice, it is not. Lead time is closely used to describe these timings. Such timings creates mostly confusion – simply waiting time what customer sees, is lead time and the time taken to complete the specific operation of manufacturing is called cycle time.

Taking a note of the question if automation is out of scope and process deals with high workforce, then Perfection in execution will require everyone’s involvement to deliver his/her part to perfection. Here Key to success is cooperative team work, which ensures win –win situation for all. If people are motivated they can do wonders. There was time, when people were only thinking that by continuous inspection (reactive approach), we can deliver the right quality to the customer. When time changed, the view also changed and people started believing in the process control (Preventive/Proactive approach) rather than inspection and the concept of six sigma or lean six sigma evolved. Defects are really tied to current conditions and best of the time, which is ignored today, if improvements are made, will be seen as a defect tomorrow. Technically saying zero defects is not possible; Hence in Six Sigma language, the definition of zero defects is 3.4 defects per million opportunities (DPMO) means allowing for a 1.5-sigma process shift, I mean which is almost near to Zero defect, the task which was looking a herculean task few years ago, is now become a part of organization culture. Setting culture is always top driven initiative. If top management is committed they can set the goal and ensures that everyone in the organization is aligned to this common goal. Simply putting the zero defect target with an audacious goal and will not help, if people are not involved even if whatever the level of automation the organization have. To run this automation process, involvement of people is also required. Without workforce nothing is possible. Hence to achieve the goal, it is required to win the heart for the workforce. Here communication is a very important factor. It is required to communicate the Plan, objectives, requirements and improvement opportunities etc. It can be achieved by creating awareness in the team by imparting required training, visual displays and monitoring, tracking and displaying the results on day to day basis and also creating an appreciation culture by recognizing and rewarding for the efforts people put in. If people are encouraged, appreciated, then their involvement will be increased and thereby helping the organization to achieve the goal. As we know even battle can be won with less ammunition if the army is motivated. Ordinary workforce with extra ordinary process can create wonder and set the benchmarking performances. While workforce is expected to follow instructions, at the same time, they can provide suggestions for the process improvements which help in improving productivity and quality. So, it is always better to keep them involve in the productivity goals and zero defect initiatives. Even in my experience I have seen most of the Kaizens in the process industries, are comes from the ground level workforce. In my opinion, zero defects or making “first time right and every time right” is a pursuit towards perfection. But the goal will create inner desires amongst the workforce to drive them forward to a point that is acceptable under even the most stringent metrics. While this goal creation is top management initiative but making it success requires workforce involvements.

Ishikawa Diagram or Fishbone diagram is also called Cause and Effect Diagram, it is used to identify, explore and graphically display, in increasing details, important possible causes related to a problem or condition to discover its root cause. The fishbone diagram gives a comprehensive list of possible causes to identify the root cause of the problem. The fishbone diagram uses a brainstorming technique to collect the causes and come up with a kind of mind map which shows you all identified causes graphically. Fishbone diagram can be used in any industry and it can be customized based on the cause and its effect. However some popular customized Ishikawa diagrams used in deferent Industries are below…… Manufacturing Industries uses 6 Ms. factor (cause) . • Machine • Method • Material • Manpower • Measurement (Inspection) • Milieu (Mother Nature – Environment) • Management • Maintenance The first six were populated by Toyota, and later on two more “Ms” were added to the list. Marketing industries uses 7Ps factor (cause) • Product • Price • Place • Promotion • People • Positioning • Packaging Service industries uses 5Ss factor (Cause) • Surroundings • Suppliers • Systems • Skills • Safety How to do it: Select the most appropriate Cause & effect format Generate the cause needed to build a cause & effect diagram, using either Brainstorming without previous preparation Check sheets based on data collected by team members before the meeting Construct the cause & effect diagram Place the problems statement in a box on the right-hand side of the writing surface Draw major cause categories or steps in the production or service process. Connect them to the backbone of the fishbone chart. If any bone is becoming too bulky, try to split it into two or three branches. Place the brainstormed or data-based cause in the appropriate category. Ask repeatedly of each cause listed on the bones.(For deeper cause ,continue to push for deeper understanding, but know when to stop. A rule of thumb is to stop questioning when cause is controlled by more than one level of management removed from the group. Otherwise, the process could become an exercise of frustration.) Interpret or test for root cause by one or more of the following Look for cause that appear repeatedly within or across major cause categories Select through either an unstructured consensus process or one that is structured ,such as Nominal Group Technique or Multi-voting Gather data through check sheets or other formats to determine the relative frequencies of the different cause. While it helps in finding bottlenecks in the process, identify ways to improve the process and the root cause of the problem but sometimes it gets misused. Fishbone diagrams are misused in the following ways. If the participant is less experienced, less involved and not more knowledgeable, the diagram will be very neat and clean and might not be able to identify the root cause of the problem. When the discussion/brainstorming session is not controlled properly it may deviate from its objective. Voting down the causes, may not be an effective way of identifying causes but the diagram follows opinion based methodology rather than evidence, this process involves a democratic way of selecting the cause. Sometimes it happens that the most obvious cause turns out to be minor and the cause thought to be a minor one was causing the issue. As the fishbone diagram Follow the divergent approach, it forces the team to consider all possible causes of a problem instead of focusing on the most obvious one. Ishikawa Diagram or Fishbone diagram.docx

Segmentation has traditionally been used by marketers to get an up close view of the market. It is based on the principle of identifying variables that predict characteristics and behavior, using a mix of quantitative and qualitative approaches to categorize customers with specific characteristics. Identifying these variables and collecting the data is considered to be time consuming and tedious. However, applying the principles of segmentation can offer deep insight about problems at hand and even help come up with potential solutions.

Quality is a make-or-break issue for most businesses. Companies with higher and more consistent quality do better over time. But this performance comes at a cost. Cost of the quality is not only the cost incurred from producing and fixing defects but also from ensuring that good products are made in the first time right and every time right. Cost of Quality Type 1.Cost of Good Quality 2.Cost of Bad Qulaity Sub Type 1.1 Prevention Cost 1.2 Appraisal Cost 2.1 Internal Failure 2.2 External Failure Definition Cost incurred to prevent or avoid quality problems /Non conformance Cost incurred for inspection/ measuring & monitoring activities. Costs are incurred for remedial measure to correct defects inside the facility before the product or service being delivered to customer. Costs incurred for remedial measure after the detection of defects by customers Example 1. Quality planning 1. Checking and testing purchased goods and services 1.Scrap (cost of product that cannot be reworked or reused) 1. Complaints Handling 2.Supplier evaluation 2.In-process and final inspection/test 2.Scrap disposal (cost of getting rid of product that cannot be reworked or reused) 2.Repairing goods and redoing services 3. New product review 3.Rework (costs of correcting quality issues on existing product) 3.Warranties 4.Error proofing 3. Field testing 4.Rework inspection (cost of inspecting a product after rework) 4.Customers’ bad will 5.Capability evaluations 4. Product, process or service audits 5.Additional material procurement (cost to replace defective or missing material) 5.Losses due to sales reductions 6.Quality improvement team meetings 6.Variability in product quality (cost of product give-away and mislabeling) 6.Environmental costs 7.Quality improvement projects 5. Calibration of measuring and test equipment 7.Downgrading (cost of lower price point of product with quality issue) 7. Product Recall/Withdrawal 8.Quality education and training 8.Supplier rework (costs attributed to supplier defects) 8. Regulatory fines and penalties Prevention is better than cure. It is always better to invest in controlling the failure rather than rectifying the non-conformity. Companies that are quality leaders spend significant resources in prevention and appraisal costs. The more a company invests in implementing strong quality methods and “good quality costs” on the front end, the more likely it is to avoid the costs of poor quality on the back end which helps in increasing customer loyalty and the resulting revenue growth. Striking a delicate balance between the cost of good quality and the cost of poor quality is an ideal .If the cost of quality is at high end of the range, you are not as efficient as your competitors. If the cost of quality is below the range, you can probably increase spending on customer satisfaction and still being cheaper than your competitors. As long as decrease in failure costs is greater than the corresponding increase in control costs, you can continue increasing its efforts to prevent or detect nonconforming units. Eventually, a point is reached at which any additional increase in this effort costs more than the corresponding reduction in failure costs. This point represents the minimum level of total quality costs. It is the optimal balance between control costs and failure costs.

Hypothesis testing helps an Organization: 1. Determine if making a change to a process input (x) significantly changes the output (y) of the process 2. Spastically determine if there are differences between two or more process outputs. Hypothesis testing assists in using samples data to make decisions about population parameters such as average, standard deviations and proportions. Testing a hypothesis using statistical methods is equivalent to making an educated guess based on the probabilities associated with being correct. When an organization makes a decision based on a statistical test of a hypothesis, it can never know for sure whether the decision is right or wrong, because of sampling variation. Regardless how many times the same population is sampled, it will never result in the same sample mean, sample standard deviation, or sample proportion. The real question is whether the differences observed are the result of changes in the population, or the result of sampling variation. Statistical tests are used because they have designed to minimize the number of times an organization can make the wrong decision. There are two basic types of errors that can be made in a statistical test of a hypothesis: 1. A conclusion that the population has changed when in fact it has not 2. A conclusion that the population has not changed when in fact it has The first error is referred to as a type I error. The second error is referred to as Type II error. The probability associated with making a type I error is called alpha (α) or the α risk. The probability of making a Type II error is called beta (β) or the β risk Let’s consider the example of a prosecuting attorney trying a case in a court of law. The objective of the prosecuting attorney is to collect and present enough evidence to prove beyond a reasonable doubt that a defendant is guilty. If the attorney has not done so, then the jury will assume that not enough evidence has been presented to prove guilty; therefore they will conclude the defendant is not guilty in the absence of enough evidence. H0 is true Truly not guilty H1 is true Truly guilty Accept null hypothesis Acquittal Right decision (probability = 1 - α) Wrong decision - fail to reject the null when it is false (probability = β) Type II Error Reject null hypothesis Conviction Wrong decision - rejecting the null when it is true (probability = α) Type I Error Right decision (probability = 1 - β) If the α risk is 0.05 ,any determination from a statistical test that the population has changed runs a 5% risk that it really has not changed. There is a 1 – α ,or 0.95 ,confidence that the right decision was made in stating the population has changed. If the β risk is 0.10, any determination from a statistical test that there is no change in the population runs a 10% risk that there really may have been a change. There would be a 1- β or 0.90 , “Power of the test “ , which is the ability of the test to detect a change in population. A 5% α risk and a 10 % β risk are typical thresholds for the risk one should be willing to take when making decisions utilizing statistical tests. Based upon the consequence of making a wrong decision, it is up to the decision maker to determine the risk he or she wants to establish for any given test, in particular the α risk. β risk ,on the other hand ,is usually determined by the following : 1. δ : The difference the organization wants to detect between the two population parameters. Holding all other factors constant ,as the δ increases, the β decreases. 2. σ : The average (pooled) standard deviation of the two populations. Holding all other factors constant, as the σ decreases, the β decreases. 3. n: The number of samples in each data set. Holding all other factors constant ,as the n increases, the β decreases 4. α : The alpha risk or decision criteria ,holding all other factors constant ,as the α decreases, , the β increases p- Value How does an organization know if a new population parameter is different from an old Population parameter? Conceptually, all hypothesis tests are the same in that a signal (δ) – to noise (σ) ratio is calculated (δ/ σ) based on the before and after data. The ratio is converted into a probability, called the P-Value, which is compared to the decision criteria, the α risk. Comparing the P value (which is actual α of the test) to decision criteria (the stated α risk) will help determine whether to state the system has or has not changed. Unfortunately, a decision in a hypothesis testing is never conclusively be defined as a correct decision. All the hypothesis test can do is minimizing the risk of making a wrong decision. Step to conduct Hypothesis Testing: 1. Define the research hypothesis for the study. 2. Explain how you are going to operationalize (that is, measure or operationally define) what you are studying and set out the variables to be studied. 3. Set out the null and alternative hypothesis (or more than one hypothesis; in other words, a number of hypotheses). 4. Set the significance level. 5. Make a one- or two-tailed prediction. 6. Determine whether the distribution that you are studying is normal (this has implications for the types of statistical tests that you can run on your data). 7. Select an appropriate statistical test based on the variables you have defined and whether the distribution is normal or not. 8. Run the statistical tests on your data and interpret the output. 9. Reject or fail to reject the null hypothesis. Rejecting or failing to reject the null hypothesis Let's return finally to the question of whether we reject or fail to reject the null hypothesis. If our statistical analysis shows that the p-value is less than or equal to the level of significance which is a cut-off point that we defined, and then we reject the null hypothesis and accept the alternative hypothesis. Alternatively, if the significance level is above the cut-off value, we fail to reject the null hypothesis and cannot accept the alternative hypothesis. A common misconception is that statistical hypothesis tests are designed to select the more likely of two hypotheses. Instead, a test will remain with the null hypothesis until there is enough evidence (data) to support the alternative hypothesis.

The 8D process was created during the Second World War by U.S. government, referring to it as Military Standard 1520: “Corrective action and disposition system for nonconforming material”. It was later applied by the Ford Motor Company in the 1960's and 1970's. 8D has become a standard in the automotive industries that require a structured problem solving process, which is used to identify, correct and eliminate problems on fast reaction to customer complaints. It is a discrete process - Start to Finish - A reactive approach that tends to only surface when correcting a problem that has already occurred. In mid of 1980s, applications of the Six Sigma methods enabled many organizations to sustain their competitiveness by integrating their knowledge of the process with statistics, engineering and project management. Motorola was the first company who launched a Six Sigma project using DMAIC methodology in the mid-1980s. Initially Six Sigma was applied in manufacturing but today it is accepted in healthcare, finance and service. Six Sigma/DMAIC is a project-driven management approach intended to improve products, services and processes by reducing defects. It is a business strategy that focuses on improving customer requirements, business systems, productivity and financial performance. Utilizing analytical tools to measure quality and eliminate variances in processes allows to producing near perfect products and services that will satisfy customers. It is continuous Improvement Process. Both are strong methods for solving problems. Both provide a consistent, structured approach, and both provide a common language so project status can be easily communicated throughout an organization. PDCA 8D DMAIC PLAN 1.Identify the problem Define 2.Use a team approach/form an 8D team 3.Describe the problem 4.Interim containment PLAN 5.Define the root cause(s) Measure Analyse DO 6.Develop solution(s) Improve 7.Implement the solution(s) CHECK 8.Prevent recurrence Control ACT 9.Congratulate the team DMAIC structure does not speak about interim containment actions , where as 8D structure particularly mentions containment as a separate step. The interim containment actions are especially relevant if you act reactively, and if your customer is already affected by the problem you are trying to solve. Comparison of Scope: SCOPE 8D DMAIC Provides Structure Yes Yes Provides containment action evaluation Yes No Provides concepts and tools No Yes Data driven No Yes ISO standards available No Yes Another important difference is the applied tools and their link to the models. While 8D only offer a structure, DMAIC offer a complete toolbox for each phase. The tools offered in the DMAIC structure is a mix of concepts and statistical tools for e.g. analysis and optimization. DMAIC is not only serving as a structure, but is often part of a data driven culture and mindset and can be used as a tool for facilitating the change to become a fact orientated company. Based on importance and urgency both the methods can be used. High Important & Urgnet - "8D Reactive Problem Solving " Important ,Not Urgent -" A proactive Improvement Process (DMAIC,Kaizen etc)" Important Low Not Important and Urgent - "Reprioritize work" Not Important and Not Urgent -"Get a new Job" High Low Urgent So, to the question regarding the use of either 8D or DMAIC, the best answer would be; use the structure requested by the customer .Organization like Ford is very particular about the 8D Problem solving methodology, If the customer asks to use 8D methodology, to satisfy him ,it is best to go with 8D .If customer has no specific request, the DMAIC maybe be the preferred structure as it also combines the tools and data driven mindset.

A control chart illustrates the dynamic performance of the process and helps us to 1. Know the historical trend or behavior of a process; 2. Monitor a process for stability; 3. Detect changes from a previously stable pattern of variation; 4. Signal the need for the adjustment of a process; 5. Detect special causes of variation. Control chart consists A central line - The center line is the horizontal reference line on a control chart that is the average value of the charted quality characteristic. Use the center line to observe how the process performs compared to the average. If a process is in control, the points will vary randomly around the center line. An upper line for Upper Control Limit (UCL)– Upper Control limit is the upper horizontal lines. The UCL is based on the random variation in the process with 3 standard deviations above the center line. A lower line for Lower Control Limit (LCL)– Lower Control limit is the lower horizontal lines. The LCL is based on the random variation in the process with 3 standard deviations below the center line. For defect data both c-chart and u-charts are used. The c chart is used to monitor the number of occurrences of an event. It requires that the opportunity for events remains the same from observation to observation while the u chart is used in the same situations as the c chart when the opportunity for events varies from observation to observation Both UCL & LCL are used to judge whether a process is in control or out of control. Control limits are calculated from the data and it is the voice of the process –how the process is capable of producing. If the LCL comes out negative in calculation, then there is no lower control limit and LCL is considered to be Zero. We know that defects cannot be less than zero (that is negative). Therefore, even though mathematically (in case) we get negative value (lower control limits) - logically we have to take zero.

Process sigma or sigma level is a measure of process capability - the higher the process sigma, the more capable the process is. Simply, the process sigma indicates how many standard deviations (“Sigma”) can fit inside the gap between the process average and the nearest specification limit Statistically a six sigma process means 2 defects per billion opportunities however popularly accepted definition of a six sigma process is one in which there are about 3.4 defects per million opportunities which is almost negligible in number and considered a near-zero defect process but it corresponds to 4.5 Sigma level. It indicates that the Six Sigma process has a short-term process sigma of 6, and a long-term process sigma of 4.5 with mean-shift of 1.5 standard deviations (sigma) over the long run. However stable any process is, over an extended period of time (Long Term), variation happens due to several sources of variation i.e. different shifts, different suppliers, different lots of material, etc. That additional variation is generally going to make a process appear less stable than a short run. Based on years of process and data collection, Motorola determined that processes vary and drift over time – and called it as Long-Term Dynamic Mean Variation and factored correction of 1.5 standard deviations (sigma) in short term sigma level rather than waiting for years of long time observation for the sigma level calculation. Using 1.5 sigma as a standard deviation gives us a strong advantage in improving quality not only in industrial process and designs, but in commercial processes as well. It allows us to design products and services that are relatively impervious, or ‘robust,’ to natural, unavoidable sources of variation in processes, components, and materials.” Conclusion: Six Sigma process capabilities is always reported in short-term sigma and the Long-term sigma is determined by subtracting 1.5 sigma from our short-term sigma calculation to account for the process shift that is known to occur over time.

Q. What is the need to identify outliers in a data-set? Outliers are extreme values or abnormal distance from other values in a random sample from a Population, in simple term the outlier is the extreme data point which is distinctly stand outs or drastically deviates from the given norms or the rest of the data. For example in the following scores 26, 27, 2, 33, 90,34,29,35 Both 2 and 90 are lies outside of the most of the other values in a set of data. 2 is much smaller and 90 is much larger compared to the other value in data set. Hence here both 2 & 90 are the outliers. Need of Outliers identification: The outliers are as important as other measures of central tendency and variability and its identification is vital for data analysis Outliers due to data entry errors (human errors) or Measurement errors (instrument errors) or Sampling errors (extracting or mixing data from wrong or various sources) distort the picture of the data we obtain using descriptive statistics and data visualization. When our goal is to understand the data, it is often worthwhile to disregard outliers. Outliers due to variability in measurements or experimental errors (data extraction or execution errors), play havoc with many machine learning algorithms and statistical models. When our goal is to predict, our models are often improved by ignoring outliers. Outliers due to novelty (not an error, natural) can be exactly what we want to learn about, especially for tasks like anomaly detection. Omitting outliers from the data set, significant changes in the conclusions drawn from the study may result. Because of this, knowing how to calculate and assess outliers is important for ensuring proper understanding of statistical data. Q. What are the methods and approaches that are useful for identifying outliers? There is some guidance which helps in a great way to start questioning about which points in the data should be treated as outliers. However none of these methods will deliver the objective truth about which of a dataset’s observations are outliers, simply because there is no objective way of knowing whether something is truly an outlier or an honest-to-goodness data point your model should account for. It is a subjective decision, depending on the goals of the analysis. Approaches for detecting Outliers. Outlier Analysis classifies Outlier detection models in following groups: 1. Extreme Value Analysis: This is the most basic form of outlier detection and only good for 1-dimension data. In these types of analysis, it is assumed that values which are too large or too small are outliers. Z-test and Student’s t-test are examples of these statistical methods. These are good heuristics for initial analysis of data but they don’t have much value in multivariate settings. They can be used as final steps for interpreting outputs of other outlier detection methods. 2. Probabilistic and Statistical Models: These models assume specific distributions for data. Then using the expectation-maximization(EM) methods they estimate the parameters of the model. Finally, they calculate probability of membership of each data point to calculated distribution. The points with low probability of membership are marked as outliers. 3. Linear Models: These methods model the data into a lower dimensional sub-spaces with the use of linear correlations. Then the distance of each data point to plane that fits the sub-space is being calculated. This distance is used to find outliers. PCA(Principal Component Analysis) is an example of linear models for anomaly detection. 4. Proximity-based Models: The idea with these methods is to model outliers as points which are isolated from rest of observations. Cluster analysis, density based analysis and nearest neighborhood are main approaches of this kind. 5. Information Theoretic Models: The idea of these methods is the fact that outliers increase the minimum code length to describe a data set. 6. High-Dimensional Outlier Detection: Specific methods to handle high dimensional sparse data