Rakesh Naik 1

Presenting a summary of the critical elements and corresponding actions for building a Lean Six Sigma infrastructure in an organisation: Phase 1: Creating visibility and awareness on the key concepts of quality and six sigma Start with the WB and YB training sessions with the operations staff and assisting them to complete their respective projects. Also share regular learning snippets on the 7 QC tools via internal communications emails to create more awareness on quality and continuous improvement. Phase 2: Highlighting the achievements of the WB and YB projects Showcase the achievements of these projects in top management forums. Getting a buy in from them to carry out similar activities in all their departments. Also present case studies demonstrating cost savings and quality improvements achieved by other organizations through Lean Six Sigma initiatives. Phase 3: Assemble a team of lean six sigma trained staff Create a team and train them in cross functional improvement projects. Focus on reducing defects or improve on efficiency. Phase 4: Get involved in setting up KPI’s Implement Data Collection systems that are appropriate to assess the key performance indicators. Forefront the aspects of setting up the KPI’s for different departments of the organisation. Phase 5: Develop a continuous improvement culture Establish a recognition program for teams or individuals who achieve significant improvements through Lean Six Sigma efforts and implementation of key quality concepts. Phase 6: Share knowledge and best practices Create a platform for regular sharing of best practices and knowledge on quality and continuous improvement happenings across industries. Also create a platform to allow the employees to log in their kaizen ideas.

A Cause-Effect Matrix is a tool used to prioritise potential causes of a specific problem. In contrast to the Cause and Effect Diagram, it uses quantitative logic to asses the priority of a cause. Ratings are given to the causes in terms of Severity, Occurrence and Impact, similar to the FMEA tool. Basis the simple calculation of Severity multiplied by Occurrence, the priority is assessed considering the impact. Sharing below an example from one of my previous organisations. Problem Statement: Customers complained of long waiting queue for chat response. Potential Causes were identified and ratings provided post discussion with impacted users. Below matrix was created with ratings on a scale of 1 to 5 (1 being low and 5 being high). Potential Causes Severity Occurrence Impact Priority (Severity x Occurrence) Process Complexity 3 2 3 6 Lack of Supervision 4 2 4 8 Inadequate Manpower 5 4 5 20 Lack of training 4 3 4 12 System Slowness 4 4 4 16 We focussed our efforts towards inadequate manpower and system slowness as these factors had the most significant impact on long waiting queues. Differences between these Cause-Effect Matrix and Cause and Effect Diagram is indicated below. Cause-Effect Matrix Cause and Effect Diagram (Fish Bone) Tabular view Visual, typically fishbone shaped diagram Quantitative Approach Qualitative Approach Primary objective to prioritise causes Primary objective to explore potential causes

As the term indicates, PQPR matrix involves planning on how efficiently we can extract the best out of the processes to get the best product without any waste. This is management tool that provides a structure to map out the relationship between products, production processes and quantities involved in each step. It helps operations by enhancing visibility, reducing waste and improving efficiency in operations. Sharing below an example of manufacturing unit making chocolates. Here it clearly allows the unit to understand and plan on: 1. Exact quantities for each type of chocolate 2. Optimize resource allocation 3. Reduce the risk of overproducing or underproducing 4. Storage units Chocolate Bar Type Production Processes Quantities Produced per Batch Milk Chocolate Cocoa Bean Roasting 1000 lbs of cocoa beans Cocoa Bean Grinding 800 lbs of cocoa nibs Chocolate Tempering and Molding 500 bars Wrapping and Packaging 500 bars Dark Chocolate Cocoa Bean Roasting 1200 lbs of cocoa beans Cocoa Bean Grinding 900 lbs of cocoa nibs Chocolate Tempering and Molding 600 bars Wrapping and Packaging 600 bars White Chocolate Cocoa Butter Processing 800 lbs of cocoa butter Mixing and Conching 600 lbs of chocolate mixture Chocolate Tempering and Molding 400 bars Wrapping and Packaging 400 bars

As the word suggests, A Multi-Generational Product Plan (MGPP) is an approach used by organizations around product development ensuring the evolution of a product over multiple generations or versions. Outlining some examples below. 1. Smartphones: Brands like Samsung, Nothing, 1plus, Apple, etc keep innovating and bring new features to their flagship phones. 2. Operating Systems: Apple’s IOS and Microsoft Windows comes up with updated features and security measures within their respective operating systems 3. Automobiles: Car manufacturers normally come up with newer models of the same variant of cars. Eg. Toyoto Fortuner has 7 variants, Tata Harrier has 24 variants, etc Similarities between these approaches is indicated below. 1. Customer Centric: Both approaches look at gathering customer feedback or looking at market trends 2. Data Driven: Both approaches look at collecting data for an informed decision 3. End Goal: Both approaches try to secure business objectives Differences shown in the below table Multi-Generational Product Plan (MGPP) DMADV Evolution of product over a period of time Designing or redesigning specific prodcuts / processes Long Time ongoing strategy Project based specific endpoint

A mixture design is a type of DOE where the goal of the experiments is to achieve the perfect blend of ingredients which will enhance the product. It looks at identifying the optimal factor settings that will allow to achieve the best product specification. The below table shows the advantages of Mixture Designs over Traditional Methods in DOE Advantages Mixture Diagram Traditional Methods Modelling and Optimization Figures out the best combination of mixtures Cannot handle mixtures Resource Planning Require less experiments, thereby reducing the cost of resources Often needs more testing, hence expensive Interactions and Constraints Accounts for rules of interaction and checks the constraints May not account for these especially when the ingredients/components are combined Stated below are some examples where mixture designs are prevalent. Pharma Industry: To optimize formulation of chemicals, this design is regularly used. Researchers try to find the optimal combination of ingredients to achieve desired properties or maximize efficacy while minimizing side effects. FMCG Industry: To achieve the correct blend of taste, smell and texture, mixture designs are used. Having the correct proportions of ingredients ensures that the outcome is achieved. Paints and Coatings Industry: Researchers use mixture designs to determine the ideal combination of raw materials to achieve final product where the characteristics such as adhesion, durability, gloss, and durability is tested.

Wherever the Box Cox transformation may not be adequately sufficient, the below reasons lead to usage of Johnson Transformation. Situation Johnson Transformation When the data exhibits both skewness (asymmetry) and kurtosis (peakedness) Allows for more flexible adjustments to shape of the distribution and can handle a wider range of distributional shapes Non normal Distributions Can handle asymentric and multi modal distributions Extreme values or outliers Robust against outliers When the variance of the residuals is unequal over a range of measured values Can achieve more homogeneous variances List of statistical analyses that may have limitations within Johnson Transformation: 1. Since the transformed values may not have direct interpretations in the original scale, obtaining exact p value may not be possible or can be tricky. 2. Constructing accurate confidence intervals for some parameters may be difficult due to the same reason as above. 3. Mann-Whitney U test or Kruskal-Wallis test do not assume normality. Since transformation may change the shape, it may violate the assumptions, thereby leading to incorrect validity of the tests. 4. While analyzing contingency tables, we assume categorical data under Chi-square test of independence. Johnson transformation may lead to compromising the assumptions of the tests. 5. Kaplan-Meier estimates or Cox proportional hazards models are commonly used for time-to-event data. Since these tests assume proportional hazards or data, it cannot be relied post transformation.

An Accountability Diagram is similar to a RACI matrix. It defines the relationships between individuals or roles within a team or organization based on their areas of responsibility and accountability. Visual representation helps in identifying the individuals easily. An Organization Hierarchy Chart displays the reporting relationships within the organization. On the other hand the Accountability Diagram focuses on specific responsibilities and accountability of individuals or roles. This is the major difference. An Accountability Diagram can be used to get the team to move through the stages of team formation quickly, by virtue of the below: Provides Clarity: The diagram helps in telling individuals what their tasks are and what they have to do at what specific times. Eliminates Ambiguity: It helps in not carrying out any activity that is not supposed to be done by a specific individual. Fosters Trust: Since each individual understands each others responsibility, there is trust. It also encourages co-operation. Effective Communication: Each individual can easily identify the appropriate person to approach for information, clarification, or feedback, streamlining communication and ensuring that information flows smoothly. Reduces Confusion: The Accountability Diagram specifies decision-making and escalation matrix within the team. Since it indicates the direct responsible people or people who have authority to decide on important matters, it reduces confusion and simplifies approach to these people. Improves team bonding: Since each team member knows each others strengths and weaknesses, they can assist each other, thereby improving team bond.

Robust Design is an engineering concept. It is used to develop reliable products or processes that are closest to the customer specifications and needs. This concept looks at reducing variation to the maximum. It also looks at covering changes in any aspect of the product or process like sudden change of dimensions or requirements by the customer or other environmental factors. List of some of the tools used under this concept: Design of Experiments (DOE): DOE is a statistical tool that studies the relationship between input and output variables. This ensures that minimum run of experiments are done to achieve the output that matches the customers specifications. Quality Function Deployment (QFD): QFD is a method that helps translate customer requirements into specific product or process characteristics. It ensures that the design reflects customer needs and preferences, thereby enhancing customer satisfaction. House of quality is the basic design tool of QFD. Failure Mode and Effects Analysis (FMEA): FMEA is a structured approach to identify and address potential failures in a process or product. This single point of failure analysis tool helps to assess the severity, occurrence and detection of failure modes and rank them to apply countermeasures on priority. Robust design principles can be applied to various industries to improve the reliability and performance of products and processes. Some examples of products developed using these principles are indicated below: High End Cars: Motor car manufacturers develop cars that are reliable and perform consistently under different road and weather conditions. While designing these cars a lot of customer requirements (implied or not implied) are tested against these principles/tools. It ensures customers comfort and safety along with other features. Smartphones: Customer dynamics are changing everyday when it comes to smartphones. Manufacturers thrive on the same and ensure that the principles of robust design are adopted to create the best in class phones with a lot of features. Robotics in hospitals: Robots performing surgery is a complex and dangerous proposition giving that a life is in hands of a machine. However the deployment of such robots go through an extensive experimental methodology before being delivered. Pharma Industry: Medicines manufactured in the pharmaceutical companies go through a lot of experimentation and research before being sent to the final consumers.

Difference: 2-Level Design Planket Burman Design Small Number of factors Large number of factors More experimental runs Fewer experimental runs Includes all possible factors Does not consider all possible factors Benefits: 1. Since it enables large factors through lesser number of experiments, there is a lot of cost and effort reduction possible. 2. Helps in identifying and narrowing down a list of significant factors. Limitations: 1. Cannot separate main effects from interactions. 2. Does not provide estimates of interaction effects between factors. Real-world examples: FMCG and Pharma Industry can use the design to screen a large number of factors like contents, packaging, etc to enhance their products.

Problem definition tree Big y is the goal that is affiliated to the customer requirement. Eg. Improvement in CSAT score, NPS improvement. Project y are small projects that help achieve the big y. Eg. Customer connect rate, Resolution response TAT, etc Rules to make the split from big y to project y: 1. Assess the core problem (business goal) post agreement from leadership 2. Write causes in negative form 3. Break them into manageable chunks 4. Validate with existing process and data 5. Weigh the possibilities of success in the existing state 6. If the existing state is not feasible, then break it down further. A black belt should be able to arrive at the project Y's simply by comparing the outcomes within each level of drilldown in terms of the effort required and the results expected.

Feature creep is the tendency to add features to your product that might derail the core usage or essence of the product. This is undesirable for the below reasons: 1. More features more complications: the end product may end up having so many features that it might get complicated for the end user. 2. More features more issues: the number of issues increase which may lead to employing of additional resources to resolve them 3. More features more cost: the cost of the features may over burden the basic cost of the core product. How to overcome feature creep? 1. Assess the customer needs a) what is the basic requirement of the customer? b) will the feature be extremely useful? c) collect customer feedback to assess what are their requirements d) ensure you set a scope for features to be added 2. Assess the business needs. a) analyse what features your product wishes to attend to in a basic sense. b) consider how much time and effort will be required to create the feature c) assess how much impact would the feature have on the product workflow d) is it going to impact your profits? e) understand what features would ensure that ensure your grip on the market

Water Spider (or Mizusumashi) is one of the keys to successful Lean manufacturing. Mizusumashi is a Japanese term for water spider. These species of spiders have an ability to work between water and air. In manufacturing, it is about the ability to move quickly between production and logistics to all spots whenever required. It helps by eliminating the waste of transportation mostly. This is majorly achieved through 1. Sorting into small quantities of big orders 2. Usage of any device that eases the movement through picking these small quantities 3. Create a pattern of delivery rather than randomly issuing out orders.