Open Access

Free eLearning, tools, templates, downloads and much more, to deliver Lean Six Sigma results, quickly.





  • DEFINE
  • MEASURE
  • ANALYSE
  • IMPROVE
  • CONTROL
  • Project Chartering
  • Customer Focus
  • High-level Process Map
  • Detailed “Current State” Process Mapping
  • Process Data Collection
  • Baseline Current Process Performance
  • Explore all Potential Causes
  • Organise Potential Causes
  • Quantify Cause & Effect Relationship
  • Generate Solutions for root causes
  • Select Solutions
  • Implementation Planning
  • Document the Process
  • Monitor the Process
  • Sustain the Results
  • The Business Case – Cost of Poor Quality
  • Defining the Problem Statement
  • Project Charter to start on the team on the right footing
  • Stakeholder Mapping to understand the resistance to change
  • Customer Types
  • Translate Customer Needs into Critical to Quality (CTQ) requirements
  • Kano Model to prioritise needs
  • Linkage to CTQ’s to Key Output Measures Y’s
  • SIPOC Process Mapping
  • Pareto Analysis Focus on vital few
  • Flowcharting
  • Value Stream Mapping
  • Operational Definitions
  • Data types, Continuous vs Discrete
  • Data Collection Plan
  • Determine Significant Sample Size
  • Gauge R & R
  • Measurement System Analysis (MSA
  • Calculate Process yield Metrics
  • Calculate Defect Metrics Process Sigma and Z Metrics
  • Graphical Analysis
  • Descriptive Statistics to Understand Process Effectiveness vs Efficiency
  • Baseline Process Capability
  • Control Chart
  • Moments of Truth Analysis
  • Hand-off & Bottleneck Analysis
  • Day In Life Of (DILO) Studies
  • The nine elements of Waste
  • Why Root Cause Analysis
  • Cause & Effect (Fishbone) Diagram
  • Identify Common vs Special Causes of Variation
  • Understand different data distributions
  • Scatter Plot
  • Correlation & Regression
  • Hypothesis Testing
  • Apply Creative Thinking to identify potential solutions
  • Mistake Proofing (Poka Yoke)
  • Eliminating Waste
  • Baic Lean Techniques – 5S, A3, 5 Why and Morning Meetings
  • Decision / Prioritisation Matrix
  • “Future State” Process Map
  • Evaluate Benefits
  • Design of Experimentation (DOE)
  • Piloting
  • Change Management and Communication Planning
  • Implementation Project Plan
  • Develop Stantdard Operating Procedures (SOP’s)
  • RACI Charts
  • Skills Matrix
  • Statistical Process Control
  • Response Plan
  • Failure Mode Effects Analysis (FMEA) to assess Process Risk
  • Align Organisational Systems & Structures
  • Visual Management
  • Lessons Learned and Best Practice Transfer

DEFINE

Purpose

DEFINE helps us put a stake in the ground so we can start our journey.

To start any journey, we need to know:

  • where we are starting from
  • where we are going
  • how we’ll get there, and
  • how we’ll know when we’ve got there

In this first step in the DMAIC Roadmap you clearly define the problem you are working on in terms of internal and external customer requirements and the high-level process that delivers those requirements.

At this stage the project and team will be chartered. Depending on your scope, team members dedicate the required time towards the project which could vary from as little as a few hours to full time.

You will also agree project scope, goals, boundaries and constraints with your Sponsor. It is very important for you/the team to be in agreement on these parameters otherwise problems will occur later on.

Output

  • Established project objectives and a focused Problem Statement which meets strategic business objectives.
  • Team formed with Project Charter describing purpose, goals and benefits of the six sigma project.
  • Ensure resources are in place for the improvement project.
  • Process to be improved identified with a high-level SIPOC process map.
  • Customers identified and their needs and requirements understood.
  • Translated customer needs into Critical to Quality (CTQ) or output requirements.
  • Quantified gap(s) between process outputs and customer requirements.
  • Defined performance standards or measures.
  • Stakeholder resistance understood.

Tools

  • Project Charter
    • Structures for the define phase and acts as a good communication tool
  • SIPOC
    • Identification of the process and boundaries of the projects
  • CTQ Statement
    • The customer need(s) made quantifiable
  • Stakeholder Map
    • Understand the level of buy-in and alignment to Project Charter

Defining the Problem Statement

Purpose

A problem statement is a clear description of th issues and business needs that need to be addressed by a problem solving team and should be presented to them (or created by them) as a first deliverable in a DMAIC project.

A good problem statement

  • Focuses the attention of the team on specific issues that need to be addressed.
  • Gives perspective and historical background.
  • Does not contain symptoms or solutions.
  • Shows the need for improvement using quantitative evidence (where possible).
  • Helps answer ‘Why this project’
  • Sets an improvement goal.
  • Should explain why the team is needed.

If the focus or scope of the problem statement is too narrow or too large or unclear, this could lead to “scope creep”, whereby avaliable resources become unmanageable

In project management, the problem statement is part of the project charter. It lists what’s essential about the project and enables the project manager to identify the project scope as well as the project stakeholders.

Method

  1. WWWHW to help us state problems”I had six honest serving men. They taught me all I knew. Their names were: Where, What, When, Why, How and Who”

    Rudyard Kipling on Learning

    The simplest way of taking Kipling’s advice is for the team to adopt a general questioning approach to problems. Thus, when faced with an uncertain situation, ask: ‘What is the problem? When does it happen? Why should it be fixed? Where are the facts?’ and so on. It is amazing how much you can achieve just by being open-minded and not accepting the status quo.

  2. Affinity diagram to help us group issuesThe affinity diagram is a common business tool used to organise ideas and data typically used as part of brainstorming activity.People have been grouping data into groups based on natural relationships for thousands of years; however the term affinity diagram was devised by Jiro Kawakita in the 1960s and is sometimes referred to as the KJ Method.

    The best results tend to be achieved when the activity is completed by a cross-functional team, including key stakeholders.

    1. Record each idea on a card or post-it note.
    2. Look for ideas that seem to be related.
    3. Sort cards into groups until all cards have been used.

    Once the cards have been sorted into groups the team may sort large clusters into subgroups for easier management and analysis.

    This is a general tool that can be used in any phase of DMAIC to sort free form comments and information, into logical groups for review and analysis.

  3. Is/Is Not Analysis to describe what the project will and will not address.Is/Is-not analysis uses a simple two-column table to help you clarify what is a part of the problem to be solved and what is not. Is the scope a one-off big bang or incremental improvements? A web-tool? A new service to our customers?It helps to frame the characteristics about the problem, such as urgency, cost, process boundaries

    The project scope focuses attention on the boundaries and resources of the team’s process improvement effort. A scope that is too broadly defined can lead a team to propose solutions that are outside it’s influence or resource capability to implement. On the other hand, a scope that is too narrowly defined can produce solutions with limited (and often disappointing) results.

    Inappropriate Frames Can Result in Poor Decision-Making

    Problem:

    Overwhelming

    Problem:

    Blindness

    Scope Cartoon

    Results:

    • Non-decisions
    • Oversight

    Results:

    • Errors of the third kind-right answer; wrong question
    • Unforeseen threats
    • Lost opportunities

    Yours Is/Is Not Analysis should clearly answer the following questions:-

    • What process will the team focus on?
    • What are the boundaries of the process we are to improve (what is the Start Point, Stop Point?)?
    • What resources are available to the team?
    • What (if anything) is out of bounds for the team?
    • What (if any) constraints must the team work under?

    Always keep the ‘Big Picture’ in mind. Your project may be part of a larger effort, so project scope may seem small relative to the big picture. If your scope is small, you need to verify that the project is part of a larger effort to impact the customer.

    Tactical projects must link to address a broader strategic perspective

    Example:

    Although the project goal is focused on order process time, we will include the inputs to the ordering process in our scope. Therefore, the accuracy and completeness of information on the customer order form will be in scope. Our process starts with the reciept of the order request and ends when the order has been delivered and the customer has acknowledged receipt. In this project, any changes to the sales process are out of bounds.

    1. Generate and record each team member views of what is in / out of scope for the project in time, money, resources, technologies etc.
    2. Team to agree and list each item under either heading “Is” or “Is Not” (in scope for the project to address)
    3. Note any outstanding actions where the team has not managed to reach agreement, to be followed up laster.
  4. SMART to help set good objectivesIt is OK to have very general visions, aims, targets or objectives, but at some point these MUST be broken down into SMART objectives – otherwise you have no way of knowing whether you have done a good job.This is a common list used as a prompt to check and help ensure objectives are suitable to meet the initial purpose set out:
    • Specific – The problem is clearly defined in all detail and is unambiguous (agreed by all concerned)
    • Measurable – Can be measured and data is avaliable to do so within reasonable cost, time and effort.
    • Achievable – It may require effort, but may reasonably be achieved
    • Relevant – How does this fit into the ‘Big Picture’ to align with other business goals and initiatives to meet customer needs.
    • Time-bound – The problem will be addressed within an agreed timescale which is neither too short (unachievable) nor vague and open_ended.

    Specific

    This stresses the need for a specific goal which aligns to a business need rather than a more general one. This means the goal is clear and unambiguous (agreed by all concerned), and not vague. To make goals specific, they must tell a team exactly what is expected, why it is important, who’s involved, where is it going to happen and which elements are important.

    A specific goal will usually answer the five “W” questions:

    • What: What do I want to accomplish?
    • Does the problem align to On Time or On Quality improvements?
    • Why: Specific reasons, purpose or benefits of accomplishing the goal.
    • Who: Who is involved?
    • Where: Use the facts to pinpoint location, geographically, process boundaries, functions or departments?
    • Which: Identify requirements and constraints.

    Measurable

    This stresses the need for concrete criteria for measuring progress toward the attainment of the goal. The thought behind this is that if you can’t measure it, you can’t manage or improve it continuously. Measuring is key to the Lean Six Sigma approach to help a team use objectivity and facts to solve the problem and achieve the goal.

    A measurable goal will usually answer questions such as:

    • How much is the benefit
    • How will I know when it is accomplished?
    • Are the key performance indicators (KPI’s) quantifiable, process related
    • Is the data available or can it be collected in a reasonable timeframe

    Achievable

    This stresses the importance of goals that are realistic and achievable with reasonable effort. When you identify goals that are most important to you, you begin to figure out ways you can make them come true. You can develop the attitudes, abilities, skills and financial capacity to reach them.

    On the other hand, a goal that is too narrowly defined can produce solutions with limited results or unreasonable expectation and the team quickly become demotivated.

    An achievable goal will usually answer questions such as:

    • Does the goal deliver best-in-class or benchmark levels of performance?
    • How can the goal be accomplished?
    • Are we committed to stretch ourselves to deliver the objectives?
    • Can we deliver more in less time

    Relevant

    This stresses the importance of choosing goals that matter to the business and fit’s into the ‘Big Picture’ and is aligned to all other activity being performed in the organisation. A manager’s goal to “Leave the office by a certain time” may be specific, measurable, attainable, and time-bound, but lack relevance.

    Every Lean Six Sigma project needs resouces and a Sponsor (someone who visibly supports breaking down barriers) to accomplish the goal. Goals that are relevant to your job role, your team, your organisation will receive that needed support. Every Lean Six SIgma project starts with a business need. A goal that is too broadly defined can lead a team to propose solutions that are outside it’s influence or resource capability to implement.

    Relevant goals (when met) drive the team, department, and organisation forward. A goal that supports or is in alignment with other goals would be considered a relevant goal. Tactical projects must link to address a broader strategic perspective.

    A relevant goal can answer yes to these questions:

    • Does addressing this problem seem worthwhile?
    • What business need is addressed by solving this problem?
    • How will our efforts align to the ‘Big Picture’.
    • Are we the right person / team to resolve it?
    • Is this the right time to look at it?
    • Is the rest of the organisation ready for it?
    • How can the rest of the organisation benefit from what we are doing?

    Time-bound

    This stresses the importance of setting a sence of urgency to the team with a deadline, giving a target date. A commitment to a deadline helps a team focus their efforts on completion of the goal on or before the due date. Early on, your team should establish target dates for completing key tasks in the DMAIC process.

    This part of the SMART goal criteria is intended to avoid lowering the importance of goals to be met and prevent golas from being pvertaken by the business as usual activities which tend to invariably arise in a n organisation. A time-bound goal is intended to establish a sense of urgency and avoids slippage.

    A time-bound goal will usually answer the question:

    • When will the deliverables be met?
    • What can we do in five days/five weeks/five months from now?
    • What are the consequences of missing the deadline?

Downloads

Search – Define

Customer Types

Purpose

Let’s start with “Who is the Customer?”

A Customer is a recipient of an output of a process. The Customer can be internal or external.

External Customers are not only the end-users of the product or service they might well be a stage before the end users of a product or service such as sales agent or distributor.

External Customers might also include key Stakeholders such as Regulators, The Community and Employees.

Once we have identified the customers, we must understand their needs and their opinions expressed about the quality of products or services.

There may already be an established process within your business to extract this valuable information on an ongoing basis (e.g. Customer Satisfaction Survey), however, it may also be the case that the business does not capture such information in a coherent and usable format. This is often the case with internal customers. Therefore, specific techniques and deeper analysis is required to understand what the customers requirements fo rth eprocess output actually are.

Method

  1. SIPOC process mapping is a very useful technique used to identify who the internal/external customers of the process actually are. Best method is to hold a workshop with all the internal stakeholders/project team to create the SIPOC process map aligned to the scope of your problem.
  2. Every organisation talks to its customers. To gather customer need Start from what you know – or think you know. Investigate what customer data is already avaliable? Market research, Customer complaints, Business Key Performance Indicators (KPI’s), Customer call reports, Industry benchmarks, Sales reports, Customer Specification’s, Service Level Agreements etc.
  3. If you need more data than currently exists, identify which of the many methods are best for collecting this information. Surveys, Personal Visits, Questionnaires, Interviews, Focus Groups, Telephone Valls, Process Stapling, Day In Life Of (DILO) studies, Internal/External benchmarking etc.
  4. As with any activity, paln before doing.
  5. Verify Customer Requirements to ensure you capture what the customer said and agree with the customers to agree a true representation of their actual needs ;
    • Not the interpretations and internal perceptions of what we think the customer wants
    • Not a subjective , anecdotal or one-sided opinion of their words
    • Not the conclusions derived from prior analysis
    • Not the decisions made from those conclusions

Data Collection Plan

Translate Customer Needs Into Critical to Quality (CTQ) requirements

Purpose

Let’s start with “Who is the Customer?”

A Customer is a recipient of an output of a process. The Customer can be internal or external.

External Customers are not only the end-users of the product or service they might well be a stage before the end users of a product or service such as sales agent or distributor.

External Customers might also include key Stakeholders such as Regulators, The Community and Employees.

Once we have identified the customers, we must understand their needs and their opinions expressed about the quality of products or services.

There may already be an established process within your business to extract this valuable information on an ongoing basis (e.g. Customer Satisfaction Survey), however, it may also be the case that the business does not capture such information in a coherent and usable format. This is often the case with internal customers. Therefore, specific techniques and deeper analysis is required to understand what the customers requirements fo rth eprocess output actually are.

Method

  1. SIPOC process mapping is a very useful technique used to identify who the internal/external customers of the process actually are. Best method is to hold a workshop with all the internal stakeholders/project team to create the SIPOC process map aligned to the scope of your problem.
  2. Every organisation talks to its customers. To gather customer need Start from what you know – or think you know. Investigate what customer data is already avaliable? Market research, Customer complaints, Business Key Performance Indicators (KPI’s), Customer call reports, Industry benchmarks, Sales reports, Customer Specification’s, Service Level Agreements etc.
  3. If you need more data than currently exists, identify which of the many methods are best for collecting this information. Surveys, Personal Visits, Questionnaires, Interviews, Focus Groups, Telephone Valls, Process Stapling, Day In Life Of (DILO) studies, Internal/External benchmarking etc.
  4. As with any activity, paln before doing.
  5. Verify Customer Requirements to ensure you capture what the customer said and agree with the customers to agree a true representation of their actual needs ;
    • Not the interpretations and internal perceptions of what we think the customer wants
    • Not a subjective , anecdotal or one-sided opinion of their words
    • Not the conclusions derived from prior analysis
    • Not the decisions made from those conclusions

Data Collection Plan

Downloads

Search – Define

Project Charter to start the team on the right footing

Purpose

A Project Charter is probably the most important deliverable necessary to get a process improvement team started off on the right footing.

A Lean Six Sigma Project Charter:

  • Clarifies the expected result(s) of the project.
  • Explains how the process is broken or not performing to requirements.
  • Keeps the team focused on the Cost of Poor Quality (COPQ) to include scrap, rework, wasted time etc.
  • Keeps the team aligned with organisational priorities.
  • Highlights measures which will be used to track performance of the project results.
  • Transfers the project from the Sponsor to the Process Improvement team .

The principle of ‘customer value’ is fundamental to applying the Lean Six Sigma problem solving approach and the project charter addresses key questions such as;

  • Does it actually add value? The answer to this question requires the organization to measure and understand who the customers are, how they define value and how this is delivered by the process being investigated.
  • Is the value being added important? Will the customer value this improvement and how does this deliver financial benefits for the business.
  • Can we quantify the non-value add? How much can we gain and is this realistic.
  • Does it create a value advantage? Does removing the waste offer opportunities for competitive differentiation.

Method

The project charter template contains eight major elements:

  • Executive Summary- around 40-60 words that outlines the focus of the project.
  • Business Case- how this project drive business initiatives and goals.
  • Problem & Goal Statements- what is wrong and what is the improvement objective.
  • Project Scope- the boundaries and resources of the team’s process improvement effort.
  • Project Organisation- detail how the team will be structured and how it will interact.
  • Key Milestones- target dates for completing key tasks in the DMAIC roadmap.
  • Communication Plan – how will the team communicate internally and externally during the project.
  • Project Assumptions- state any judgements or assumptions made.

Usually, a first draft of the charter is prepared by the Project Lead and the project Sponsor at the start of the Define phase. Then, the improvement team reworks the charter until it is something that they can feel comfortable owning. At this point, the charter goes back to the Sponsor for his or her approval
The Improvement Team and Sponsor should continue to further clarify the charter as work progresses through the subsequent DMAIC phases.

It is important that the charter reflects the best understanding of the improvement project at all times.

Downloads

Search – Define

SIPOC Process Mapping

Purpose

We want to think about the business organization and business activities as a Process.

Processes are the means by which value is generated and delivered within the business to meet and exceed our customers expectations. Processes are the lifeblood of every organisation.

Processes are

  • Usually invisible
  • Frequently intangible
  • Often unrecognisable
  • Always the foundation for identifying, analysing, designing and sustaining performance improvement within the business

Processes are frequently not understood . You cannot improve what you do not understand.

Describing work as a process;

  • Applies to all nature of functional or departmental activities across any type of manufacturing, service or transactional business environment.
  • Applies to all work, whether repetitive in nature or ‘one-of-a-kind.’

Having a high level view of a process helps:

  • Document a process as it moves from suppliers’ parts to customers’ product(s).
  • Identify high level business activities aligned to the Problem Statement.
  • Define project boundaries (starting and ending points) and the project scope.
  • Describe where to collect data.
  • Process Owners to use the right metrics and verify them continuously with the customers and suppliers.

Process Elements

  • Supplier: Whoever provides the inputs to your process
  • Input: Materials, Resources and data required to execute your process
  • Process: A process is a collection of activities that takes one or more kinds of input and creates output that is of value to the customer
  • Output: The tangible products or services that result from the process. Equivalent to a customer need
  • Customer: Whoever receives the outputs of your process – internal or external
  • Requirements: A specific characrteristic of the output that will determine the extent to which customer satisfaction is achieved
  • Boundary: The limits of a particular process, usually identified by the inputs and outputs, that separate what is outside the process – where it starts and stops

Process Diagram

Method

  1. Identify the primary 5-7 process steps using brainstorming or other techniques, including the Start and End point.
  2. For each step, identify primary Outputs and the Customers who receive them. Remember that Customers can be both external and internal.
  3. For each step, identify the primary Inputs needed to perform the activity to produce the Output. Remember that the Output from one process step can be an Input in a subsequent step.
  4. For each step, identify the Supplier or source of the needed Inputs. Remember that Suppliers can be both external and internal.
  5. Remember that the Customer of one Process step can be a Supplier of Inputs in a subsequent step.
  6. Verify your ‘finished’ SIPOC map with the people who are involved in the operation of the process.

Downloads

Search – Define

Stakeholder Mapping to understand the resistance to change

Purpose

Lean Six Sigma provides a framework for the management and improvement of our activities and processes

BUT

The project itself needs stakeholder acceptance to be able to start and progress successfully.

Stakeholder Analysis helps the project team identify key stakeholders and assesses the gaps between how they currently feel about the change initiative and the level of support they need to exhibit for the change to have a good chance of success

Method

  1. During the Define/Measure phase as you begin to understand which processes are impacted, list individual stakeholders i.e. those who have a vested interest in the project and its outcome.
  2. Discuss where each is currently at with regard to the proposed change initiative in terms of their understanding for the need to change and alignment to the general consensus on how this will be achieved.
  3. Discuss where each stakeholder needs to be
    • who must move
    • how far
    • in which direction
  4. Analyse the gaps.
  5. Begin to form influencing strategies and communication actions to close the gaps.
  6. Continually review and update stakeholder positioning as this will change during the course of the project.

The Business Case – Cost of Poor Quality

Purpose

The Business Case Links the Project to the strategic priorities of the business, answering:

  • How will this project drive business initiatives and goals?
  • How will this project impact the customer (internal or external)?
  • Why is it important to do it now?
  • Why is it a priority?
  • What are the consequences of not doing it now?
  • What are the expected financial benefits (increased revenues and/or cost reductions)

Financial benefits for a Lean Six Sigma project are estimated early in the DMAIC roadmap for two reasons:

  • – To help focus the team’s effort on achieving a tangible goal and realising results.

Method

A good business case report, which brings confidence and accountability into the field of making investment decisions, is a compilation of all information collected during enterprise analysis and the business case process. The key purpose is to provide evidence and justification for continuing with the investment proposition. Here is a recommended structure:

  • Preface
  • Table of Contents
  • Executive Briefing
  • Recommendation
  • Summary of Results
  • Decision to be Taken
  • Introduction
  • Business Drivers
  • Scope
  • Financial Metrics
  • Analysis
  • Assumptions
  • Cash Flow Statement (NPV)
  • Costs
  • Benefits
  • Risk
  • Strategic Options
  • Opportunity Costs
  • Conclusion, Recommendation, and Next Steps

Where Do Financial Benefits Come From?

Project benefits can be divided into two categories, Hard benefits and Soft benefits. Hard benefits fall mainly into two categories:-

  • Incremental Revenues
    • Increased Volume
    • Increased capability
    • Improved product quality or service
    • Improved commercial processes
    • Increased price (for higher quality products & services)
  • Decreased Costs
    • Reduction in rework
    • Labour efficiency
    • Reduced operating expenses
    • Reduced plant and equipment depreciation or lease expenses
    • Reduction in non-value add

Your business case should be backed up with supporting data. Soft benefits however are more difficult, if not impossible to quantify. Theses should be tracked, but separately from Hard financial benefits. Soft benefits include:-

  • Cost avoidance, where no auditable trail to operating margins exists
  • Customer retention
    • Avoiding losing future sales/use of services
    • Reducing customer churn
    • Reducing the cost of customer acquisition by differentiating from our competitors through superior products and services
  • Improved stakeholder satisfaction
    • Employees
    • Customers

Soft benefits are important, but are often difficult to quantify

  • In Define, the estimated financial benefit is based on best available data and assumptions about where incremental benefits will come from. Subsequent phases of DMAIC will help to refine the estimate.
  • In Measure, you will take measures of the process’ performance. These will serve as a baseline, against which any incremental improvement will be measured.
  • In Analyse, you will determine root causes of the problem and assess the assumptions used in estimating benefits in Define. The estimate will be revised to incorporate new data.
  • In Improve, you will select the best solution to the problem and complete a full cost/benefit analysis which incorporates the cost of implementing your solution, to estimate the actual bottom line benefit of the improvement.
  • In Control, actual benefits will be measured and reported.

Downloads

Search – Define

Kano Model to proritise needs

Purpose

The Kano Model of customer satisfaction classifies product attributes based on how they are perceived by customers and their effect on customer satisfaction. These classifications are useful for prioritizing customer needs as they indicate when good is good enough, and when more is better.

Having insight into which customer requirements fall into which quality dimensions can improve focusing on the “vital few” CTQ’s (Critical to Quality).

Method

The Kano Model divides product or service attributes into three categories: threshold, performance,
and excitement. A competitive 
product meets basic attributes, maximizes performances attributes, and includes as many “excitement” attributes as possible at a cost the market can bear.

Kano Model

Kano Model Chart

Threshold Attributes

Threshold (or basic) attributes are the 
expected attributes or “musts have’s” of a
 product or service requirement and do not provide an
opportunity for product differentiation.
 Increasing the performance of these 
attributes provides diminishing returns 
in terms of customer satisfaction,
 however the absence or poor 
performance of these attributes results in 
extreme customer dissatisfaction. An example of a threshold attribute would be brakes on a car.

Performance Attributes

Performance or “need to have” attributes are those for which more is generally better, and will improve customer satisfaction. Conversely, an absent or weak performance attribute reduces customer satisfaction. Of the customers needs expressed, most will fall into the category of performance attributes.

The price for which the customer is willing to pay is closely tied to performance attributes. For example, customers would be willing to pay more for a car that provides them with better fuel economy.

Excitement Attributes

Excitement or “Delighter” attributes are unspoken and unexpected by customers but can result in high levels of customer satisfaction, however their absence does not lead to dissatisfaction. Excitement attributes often satisfy latent needs – real needs of which customers are currently unaware. This can provide a competitive advantage. Although they have followed the typical evolution to a performance then a threshold attribute, cup holders were initially excitement attributes.

Linkage of CTQ’s to Key Output Measures, Y’s

Purpose

CTQ stands for Critical To Quality. It represents the “vital few” attributes which have the most impact on quality that are defined by the customer. It may include the upper and lower specification limits or any other factors related to the product or service. It usually must be interpreted from a qualitative customer statement to an actionable quantitative business specification.

To put it in layman terms CTQs are what the customer expects of a product or service…the spoken needs of the customer. The customer may often express this in plain english and it is up to us to convert them to measurable terms.

Method

  1. The Voice of Customer’s could be in any form and can often contain implied solutions or anecdotal evidence. We must translate them first into clear Statements of Need.
  2. Then we can identify the characteristics of our product or service that will tell us whether we have met that need.
  3. The VOC data collected to help categorise needs and specific requirements. We are aiming to translate what we hear form the Customer (VoC) into something we can measure and control in our process (Output Characteristic).
  4. Once we have identified the output characteristic, we need to identify what we can measure in our process that will tell us whether we are meeting the need or not. With the CTQ we have a target value and often limits (tolerances).
  5. The CTQ allows us to measure the performance of the process through cycle time analysis, process capability and process control techniques in subsequent phases of the DMAIC methodology.
  6. Once we have identified all our CTQs, we have the definitive list of things that are important to the customer – we have the metrics against which we assess ongoing performance.

CTQ Chart

MEASURE

Purpose

MEASUREMENT is at the heart of the Journey

If we don’t know where we are, or where we’ve been…

  • What’s happened in the past?
  • What’s meant to happen now?
  • What really happens now?
  • When is it happening?

And we need to understand our performance to eliminate guesswork and assumptions and assess how well it is working.

The measure step of the DMAIC Roadmap helps you to get a clear picture of the process.

Processes are the means by which value is generated and delivered within the business. Processes are the lifeblood of every organisation. Processes are usually invisible, frequently intangible, commonly unrecognisable and often not well understood. You cannot improve what you do not understand.

The current baseline performance is understood in terms of degree of variation in the process, capability to perform against the required customer level and the current cycle time of the process.

In this step you should develop three types of measures: output, process and input measures. Output measures determine how well the customer requirements are met, Process measures indicate how well the process is performing and Input measures assess supplier quality.

Data collection begins once you have determined what you will measure. Collecting quantitative and qualitative data is important and helps pinpoint how the process is operating which gives useful insight on where to focus your analysis.

Output

  • Current process mapped in detail.
  • Process metrics identified.
  • Key Measurements Agreed and linked to CTQs.
  • Identification and Definition of Defects in Process.
  • Data Collection Plan.
  • Process Variation Displayed.
  • Baseline Measure of Process Capability and Process Sigma Level.
  • Baseline Cycle and Lead Time in the process.
  • Identify current level of Process Risk.
  • Cost of Poor Quality (COPQ)/ benefit estimate.

Tools

  • Flowcharting
    • Building on SIPOC thinking but adding more “levels” of detail
  • Process Metrics
    • How good is the data and what metrics do we use to measure current performance?
    • Data Collection Plan
  • Control Charts
    • What is the “nature” of the process and what is the current performance and capability?
    • Value Stream Mapping
    • Failure Mode Effects Analysis
    • Measurement System Analysis

Calculate Process Yield Metrics

Purpose

Traditional Yield / Final Yield – how many units come out the end of the process from input units.

The traditional definition of yield is also referred to as Final Yield. In the traditional sense, the term “yield” is often used to refer to the number of “good” units produced. Used in this manner, “yield” is expressed as a percentage of the total number of units produced.

Final Yield Calculations

In the example shown above, 100 units enter into a three-step process.

At the end of this process, 70 units are counted as “good” with 30 units scrapped for having defects. Applying the traditional definition of yield to this three-step process produces a yield calculation of 70%.

Traditional final yield fails to take into account the time and money required to rework defective units. This is the “Hidden Factory.”

Time and resources expended through rework is referred to as the Hidden Factory, because the rework loop diminishes overall process capacity.

Why do we care about the hidden factory?

  • waste opportunity
  • unmeasured cost traditionally
  • reworked units more likely to be problems in the future

Throughput Yield accounts for this lost output by only counting those units that were not scrapped and did not require rework as “good” or defect-free units.

Throughput Yield Calculations

In the example shown above, 100 units enter into a one-step process. In this process, 10 units were lost as scrap and 10 units needed to be reworked and placed back into the process. Applying Throughput Yield to this process produces a yield of 80%.

Method

Rolled Throughput Yield

  • FY – Final Yield is units output vs. units required as input (only accounting for scrap).
  • TPY – yield for the step including measurement of reworked units.
  • RTY – Rolled throughput yield – the percentage of units passing though all process steps without being reworked or scrapped along the way.
  • Rolled Throughput Yield (RTY) is a more accurate representation of overall process output, because it accounts for the impact of both scrap and rework on the production of “good” or defect-free units.
  • Rolled Throughput Yield is calculated by multiplying TPY for each process step.

Downloads

Search – Measure

Data Collection Plan

Purpose

Data collection is the process of gathering and measuring information on process variables of interest, in an established systematic fashion that enables us to answer stated questions, test hypotheses and evaluate outcomes. Data collection and analysis must not start until you have defined a data collection plan

  • what you will measure
  • who will be involved
  • how

The goal for all data collection is to capture quality evidence that then translates to rich data analysis and allows the building of a convincing and credible answer to questions that have been posed.

A formal data collection plan is necessary as it ensures that data gathered is both defined and accurate and that subsequent decisions and findings are valid. The process provides both a baseline from which to measure and in certain cases a target on what to improve.

Consequences from improperly collected data include:

  • Inability to answer questions accurately.
  • Inability to repeat and validate the analysis.

Method

  1. Identify what detailed information is still needed – qualitative measurements are as valid as quantitative measurements
  2. Develop operational definitions of the measurements
  3. Identify how to measure – data sources and methods
  4. Prepare a data collection and sampling plan
  5. Design surveys, interviews, questionnaires, workshops
  6. Implement and refine data collection as required

Downloads

Search – Measure

Data Types, Continuous vs Discrete

Purpose

Generically, there are different types of data & it is important to recognise what data we have & put it in the right category. We need to have an understanding of variation in data and to do this, we need to understand the different types of data.

There are 2 main types of data –continuous (measured) and discrete (counted). Continuous data is also referred to as Variable data and Discrete data is also referred to as Attribute or Countable data.

Attribute data is black or white, on or off, right or wrong.

The more continuous the data, the more it will tell you about your process.

  • We need to be able to recognise the two different types of data
  • We need to understand that continuous data is usually more useful:
    • It is more common; any time we measure using SI (International System) of units or any kind of scale we are generating continuous data.
    • Usually it gives us more information about process performance and it allows us to do more sophisticated analyses.
    • It usually has a Normal Distribution (but not always).

Method

  • Continuous data is measured on a continuous scale.
    • Examples include weight, height, time, temperature, pressure, etc.
    • In theory there is an infinite possibility of values (even if we choose to round the value to the nearest unit)
      e.g. time taken = 229.5689773… seconds.
  • Attribute data
    • Each item is classified as belonging to one of a number of categories. Usually there are just two categories, e.g. correct and incorrect, defective and non-defective, in or out of specification
    • Nominal – Unordered, named categories.
    • (e.g. SIC categories: manufacturing, mining, etc., job titles: engineers, scientists, etc)
    • Ordinal – Ordered categories where the distance between the categories is unknown (e.g. poor, fair, good, excellent).
  • Countable data
    • Each data item is the count of the number of faults, accidents, events, etc for a given time or quantity, e.g. number of defects, errors, etc

Determine Significant Sample Size

Purpose

This is an area of statistics on it’s own – entire books are written on the subject. Our common sense tells us that the nearer the sample size is to the population size then the more accurate an indicator of the population it will be. Data is obtained using samples because we seldom know the entire population. Population statistics are desired, but often not available.

  • Population is every possible observation (census)
  • Samples are subsets of populations

For all purposes, the required sample size depends on:

  • How big a difference we need to distinguish
  • How much risk we’re willing to accept
  • How precise we need to be

We need a much larger sample to be 99% confident that a true average is within 1 unit of our sample average than to be 75% confident that that the true average is within 100 units of the sample average!

Imagine a needle in a hay stack. How many handfuls of hay do I need to take to find the needle? Suppose I’m looking for an elephant in a haystack, how many handfuls.

Samples from a normal distribution can be used to ‘infer’ or approximate population parameters. The sample must be chosen randomly in order to be supported statistically as a valid representation of the entire population.

For business decisions

  • You will take a relatively small sample of a population.
  • You will assume your sample to be representative of that population.
  • You will project the findings in the sample to the population.
  • You will seek other confirming evidence.

Method

Decisions on sample size are a balance between an acceptable amount of uncertainty (due to a small sample) and cost (the larger the sample the higher the cost). The researcher must be ready to abandon the study if the budget will not support an acceptable sample size. The decision based on faulty data will be no better than one based on no data.

There are Sample Size Calculators to work out a statistically significant sample size needed

Check the web (search for “Sample Size Calculator”)

Downloads

Search – Measure

Flowcharting

Purpose

The development of a flowchart is a team effort and it will educate everyone on the reality of the situation and typically how little value added activities exist in the current process.

Typically, there are obvious improvements that will be become evident while the mapping is being done.

It will not be perfect the first time through, add more detail as it evolves. It is important to remain with the Start and Finish boundaries specific to the SIPOC generated in the DEFINE phase.

The team needs to begin identifying the value added steps versus the non-value added steps and time. This realization is clearer during Value Stream Mapping (VSM) which will quantify the amount of value and non-value add time spent in the process.

Method

Simple steps to draw a Process Flowchart;

There are only a few basic activities within any process and these can be described using the following basic symbols:

Flowchart Symbols

  1. These four symbols are capable of describing any process, so to start with keeps it simple and easy to follow.
  2. Draw the main flow in a straight line. Left-to-right or top-to-bottom.
  3. Keep to one page from start box to stop box. But this may mean one large page.
  4. Consider the normal course of actions and business activities specific to the scope of your process, aligned to the SIPOC generated in the Define phase. Start with the first key activity and represent it using the appropriate symbol.
    • the starting ‘trigger’ – uses the start/stop box
    • a process step – use the process step / operation box
    • a decision (or branch point) – use the decision box
    • Process loop, alternate path or rework activity – use a combination of process step boxes and a decision box to determine until the activity ends and joins to its original start point
    • the end ‘trigger’ – uses the start/stop box
  5. Continue to consider the next activity and add it using the appropriate box.
  6. Don’t worry at this stage if the boxes won’t fit on a page, this can be sorted later.
  7. Keep the ‘normal’ route in a straight path starting at the start box at the top of the page and ending with the stop box at the bottom. Also keep the alternate routes in parallel straight lines (i.e. following a decision box) to the left or right of the main line and avoid crossing flow lines.
  8. Any action and subsequent process flow required following failure to pass a ‘gate’ must be shown.

Gauge R & R

Purpose

Before you…

  • Make adjustments
  • Implement solutions
  • Run an experiment
  • Perform a statistical analysis

You should…

  • Validate your measurement systems
  • Validate data and data collection systems

A Gauge is a measuring device. Often an employee forms part of the measuring device.

Gauge Repeatability & Reproducibility tests are often referred to as Gauge R & R.

Gauge Repeatability is about the inherent precision and consistency of the gauge itself.

Gauge Reproducibility is about whether we would expect inconsistency in gauge measurements due to different people using the gauge.

  • If you know the Standard Deviation of your Gauge System you can decide if you have a good gauge?
  • Can the gauge discriminate between Good and Bad Parts?
  • Can the gauge detect process variation, the ultimate aim?

Method

Test for Repeatability

The variation when one person repeatedly measures the same unit with the same measuring equipment.

  • For Continuous data
    • Calculate variation in terms of measurement units (standard deviation, etc.)
  • For Discrete data
    • Count number of times the same result is achieved for a given point (% correct)

Test for Reproducibility

The variation when two or more people measure the same unit with the same measuring equipment

  • For Continuous data
    • Calculate the difference between two people in terms of measurement units
  • For Discrete data
    • Calculate the difference in number of times each person achieved a given result (% difference)

Downloads

Search – Measure

Measurement System Analysis (MSA)

Purpose

Before you…

  • Make adjustments
  • Implement solutions
  • Run an experiment
  • Perform a statistical analysis

You should…

  • Validate your measurement systems
  • Validate data and data collection systems

MSA is a scientific and objective method of analyzing the validity of a measurement system

  • A “tool” which quantifies:
    1. Equipment Variation
    2. Appraiser (Operator) Variation
    3. The Total Variation of a Measurement System
  • MSA is NOT just Calibration
  • MSA is NOT just gauge Repeatability & Reproducibility (R&R)

Measurement System Analysis is often a “project within a project”

MSA quantifies a major source of variation, Measurement Error.

Method

Components of Measurement Error

  • Resolution/Discrimination – Smallest detectable increment between two measured values
  • Accuracy (bias effects) – Difference between the average value of all measurements of a sample and the true value for that sample
  • Linearity – A measure of any change in accuracy or precision over the range of instrument capability
  • Stability (consistency) – distribution of measurements that remains constant and predictable over time for both the mean and standard deviation
  • Repeatability-test-retest (Precision)- Variability inherent in the measurement system under constant conditions
  • Reproducibility (Precision)- Variability among measurements made under different conditions (e.g. different operators, measuring devices, etc.)

Each component of measurement error can contribute to variation, causing wrong decisions to be made.

Operational Definitions

Purpose

An operational definition is a clear, concise description of a measurement and the process by which it is to be collected.

  • What: Must have specific and concrete criteria
  • How: Must have a method to measure criteria
  • Must be useful to both the business and the customer (the “end-to-end” concept)

The purpose of the Operational Definition is;

  • To remove ambiguity
    • Everyone has a consistent understanding
  • To provide consistency and predictability
    • Defines when and how to take the measurement to minimise ability to ‘doctor’ the results
    • Makes sure that no matter who does the measuring, the results should be the same

Method

To define a metric, you must include two specific elements:

  • Operational Definition: Unambiguous and precise
  • Unit of measurement

An Outside-In measurement perspective

The Operational Definition should reflect how the ‘customer’ sees the process. This may expose issues which may lie outside the scope of the process, but which can impact process effectiveness and efficiency – and so must be factored into your analysis and solution

Scale of scrutiny

  • Measure one scale/level ‘smaller’ than the one your customer uses
  • For example, assume the metric is focused on turn-around time
    • If your customer measures (expects) in terms of days, your measure should track performance in terms of hours
    • If your customer measures (expects) in terms of hours, you should measure in minutes
  • Use a scale of scrutiny that allows you to identify and assess performance variation

Pareto Analysis, Focus on vital few

Purpose

Vilfredo Pareto was a 19th century Italian economist who noted that most of the country’s wealth was owned by a small proportion of the population.

The use of this principle was popularised in 20th century quality control by the Quality Guru, Joseph Juran, who called it the Pareto Principle.

It is also widely known as the ’80/20 rule or the law of the vital few, and is true in many different situations.

The Pareto principle states that, for many events, roughly 80% of the effects come from 20% of the causes (the vital few). For example:

  • 80% of a company’s profits come from 20% of its customers
  • 80% of a company’s complaints come from 20% of its customers
  • 80% of a company’s profits come from 20% of the time its staff spend
  • 80% of a company’s sales come from 20% of its products
  • 80% of the Output of any business is completed by 20% of the Workforce

Method

Simple steps to draw a Pareto Chart

  1. Identify what you are measuring – eg. Type of customer enquiry.
  2. Identify measure of importance.
    • Number of items is a common and easy measure to use.
  3. If you have not already done so, measure it.
    • ideally, aim for 30 or more measurements. Always aiming for collecting as much data points as possible.
  4. Sort the measures into order.
    • If there are a number of small items, bunch them together into a ‘miscellaneous’ category.
  5. Plot the bars in order of size.
    • but always put the ‘miscellaneous’ bar at the end.

Pareto Chart

Downloads

Search – Measure

Value Stream Mapping

Purpose

Toyota have been benefiting from Value Stream Mapping since 1940’s.

The use of Value Stream Mapping (VSM) has been attributed to the cause of much of the
success. Developed during the work conducted by Taiichi Ohno at Toyota in the 1960’s and 70‘s, at its basic level VSM is a systematic methodology to identify wasted time and actions in a process.

Value Stream Mapping is a key business tool utilised throughout the world, in many organisations to strategically plan and focus on continuous improvement.

In Taiichi Ohno’s words –

“All we are doing is looking at the time line from the moment the
customer gives us an order to the point when we collect the cash. And we are reducing that time line by removing the non – value-added wastes.”

(Ohno, 1988)

Key Concepts

It is useful to explain the meaning of several key concepts used in VSM . These are:

What is meant by a process, what waste is, what is meant by ‘flow’, what constitutes value-adding, along with what is needless non-value-adding and what is necessary non-value-adding.

A process is a series of activity steps that move inventory from one step to the next to transform it into the intended output. The output could be a physical item or a service.

A process can be any type or size and cover any period of time. Each step in a process also consists of processes within the step. VSM is used to investigate processes to identify improvement opportunities lying in their wastefulness and lack of fluidity.

A Value Stream Map highlights the concept of “cycle time” which provides a dynamic picture of the value stream.

Waste is one of the seven wastes identified by Toyota.

These are

  • Overproduction: Producing items for which there are no orders.
  • Waiting Time: Employees standing about.Inventory or Work-in-Progress (WIP) at stand still.
  • Unnecessary Transport: Moving material or information unnecessarily or long distances.
  • Over-processing: Using more steps to produce a product than necessary.
  • Excess Inventory: Retaining unnecessary inventory between process steps.
  • Unnecessary Movement: Any wasted motion by employees or machines.
  • Defect: Making incorrect product

Method

  • Use standard symbols to ensure common language and tractability.
  • Trust nobody – check everything. Be there. Can’t be done remotely, you must get down on the shop floor and immerse yourself in the Value Stream.
  • Team can be Senior, Junior or mixed. Best is six to eight people who understand the end to end process.
  • Value Stream Mapping can be a very intrusive process and, therefore, it can often be hard to get top level buy in. This is because it can often be seen as a threat, therefore, you must communicate what and why; gather a broad cross-section of the business to generate the map; then get people to look at the map, review and enhance.

Calculate Defect Metrics Process Sigma and Z Metrics

Purpose

Graphical analysis is commonly used to visually display the relationship between two or more variables. There are many graphical techniques such as Bar Chart, Dot Plot, Pareto Chart, Control Chart, Histogram, Run Chart, Scatter Plot.

The shape of a graphical plot can reveal a great deal about a process, giving a picture of the
variation, highlighting unusual values and indicating the probability of particular values occurring.

The latter three techniques are further explained below.

Method

  • To look at a static display of process variation – use the Histogram to understand the shape, central tendency and variation of the data without any dynamic time element.

    It gives a picture of the extent of variation, highlights unusual areas and indicates the probability of particular values occurring and can also be used to compare a process to requirements, if the specification lines are drawn on the histogram

    Histogram

  • A Run Chart is a line chart showing process performance over time and highlghts the dynamic aspects of process variation. It is easy to plot, interpret and usually no calculations are necessary.

    A run chart can focus attention on major changes in the process such as trends, shifts or cycles.

    Run Chart

  • To test if the variation in a single X (Cause) has a significant effect on a Y (Effect) we use a scatter diagram and correlation analysis. The variables are plotted on axes at right angles to each other and the scatter in the points gives a measure of confidence in any correlation shown.
    Degrees of CorrelationNo Correlation. There is no demonstrable connection between X and Y.Possible Correlation. There may be a connection between the variables.

    Positive Correlation. It is likely there is a connection between the variables.

    Perfect Correlation. One variable is exclusively and fully linked with the other.

In practice Minitab, Excel or other graphical software packages are more commonly used to perform the analysis and this is the recommended approach.

Downloads

Search – Measure

Graphical Analysis

Purpose

Graphical analysis is commonly used to visually display the relationship between two or more variables. There are many graphical techniques such as Bar Chart, Dot Plot, Pareto Chart, Control Chart, Histogram, Run Chart, Scatter Plot.

The shape of a graphical plot can reveal a great deal about a process, giving a picture of the
variation, highlighting unusual values and indicating the probability of particular values occurring.

The latter three techniques are further explained below.

Method

  • To look at a static display of process variation – use the Histogram to understand the shape, central tendency and variation of the data without any dynamic time element.

    It gives a picture of the extent of variation, highlights unusual areas and indicates the probability of particular values occurring and can also be used to compare a process to requirements, if the specification lines are drawn on the histogram

    Histogram

  • A Run Chart is a line chart showing process performance over time and highlghts the dynamic aspects of process variation. It is easy to plot, interpret and usually no calculations are necessary.

    A run chart can focus attention on major changes in the process such as trends, shifts or cycles.

    Run Chart

  • To test if the variation in a single X (Cause) has a significant effect on a Y (Effect) we use a scatter diagram and correlation analysis. The variables are plotted on axes at right angles to each other and the scatter in the points gives a measure of confidence in any correlation shown.
    Degrees of CorrelationNo Correlation. There is no demonstrable connection between X and Y.Possible Correlation. There may be a connection between the variables.

    Positive Correlation. It is likely there is a connection between the variables.

    Perfect Correlation. One variable is exclusively and fully linked with the other.

In practice Minitab, Excel or other graphical software packages are more commonly used to perform the analysis and this is the recommended approach.

Descriptive Statistics to Understand Process Effectiveness vs Efficiency

Purpose

Descriptive Statistic consists of basic statistics and graphical techniques used to summarize data

  • Measures of central location
  • Measures of spread (dispersion)
  • Evaluation of symmetry & skewness

Measures of central tendency include the Mean, Median and Mode while measures of spread include the Standard Deviation (or Variance) and the Range.

Typical graphical techniques

  • Histograms
  • Box plots
  • Dot plots
  • Normal probability plots

Continuous Data is frequently but does not always follow a Normally Distribution.
This Bell-shaped curve has important properties that allows a lot of analysis to take place. It is completely described by its mean and standard deviation.

Method

Normal Distribution

  • Accuracy is an indicator of process Effectiveness and is calculated by the Mean or Average of the data set.
    Precision is a useful indicator of process Efficiency and is calculated by applying various statistical principles to the concept of Sigma or Standard deviation.
  • The Mean of a sample of data points would be the Sum of all of the Data Points divided by the Number of Data Points. The MS Excel formula for this function is =AVERAGE(A1:Ax)
  • Given a set of numbers, the Range is equal to the distance between the extreme values of the data set. (Highest – Lowest). The MS Excel formula for this function is: =SUM(MAX(A1:Ax)-MIN(A1:Ax))
  • The Standard Deviation is a measurement of the average distribution of data points relative to the mean. The MS Excel formula for this function is: =STDEV(A1:Ax)

Baseline Process Capability

Purpose

Process Capability helps us to answer the question – “Is the process capable?”. It measures whether the process, given its natural variation, is capable of meeting established customer requirements or specifications. It can be useful for:

  • Measuring continual process improvement over time
  • Prioritising processes to improve
  • Determining if there has been a change in the process

Method

To measure the degree to which a process is or is not capable of meeting customer requirements, capability indices have been developed to measure the distribution of your process in relation to the specification limits.

The two most common capability indices are;

  • Process Capability Index, Cp
    • Confirmation that very few products/services are being produced outside tolerance limits when these are specified.
      Cp

      Cp = Tolerance band/Process variation

      = 2T / 6σ

      =(usl – lsl) / 6σ

      usl = upper specification limit

      lsl = lower specification limit

  • Process Capability Index, Cpk
    • Confirmation that the majority of products/services are being produced at or close to a nominal target value
      Cpk

      Cpl = x̄ – lsl / 3σ

      Cpu = usl – x̄ / 3σ

      Cpk = the lower of Cpu and Cpl

Targets for Cpk of 1, 1.33, 1.67 and even 2 have been set by many industries.

Cpk % Defects Conclusion
0.7 27% (274,000 DPMO) Not capable
1.00 6.7% (66,800 DPMO) Barely Capable: Part of process distribution outside of specification
1.33 0.6% (6,210 DPMO) Minimum Acceptable Process: Process distribution barely within specification
1.67 0.02% (233 DPMO) Acceptable Process: Process distribution within specification
2.00 0.0003%(3 DPMO) World Class Process

Downloads

Search – Measure

Control Charts

Purpose

Control Charts are used to monitor, control and improve process performance over time by studying variation and its source. This helps improve a process to perform consistently and predictably for higher quality, lower cost and higher effective capacity. It provides a common language for discussing process performance.

  • It helps you read process data so you can see and understand what is happening.
  • If you can know what is happening, then you can start to think about if any action is necessary and if so, what action to take.

A control chart is basically a run chart with control limits. The centre line represent the mean value of the data and the control limits are positioned +/- 3 standard deviations either side of the centre line.

Control Chart

  • Control charts do not tell you if the process is performing to customer specification; they tell you if the process is ‘in control’ (Voice of the Process).
  • You need data from a process operating in a stable condition to be able to set up control charts. This means you may have to eliminate special causes of variation before you can set up a control chart.

Method

There are many types of control charts and the one you decide to use will be determined by the type of data and the sample size you have.

  1. Select the process to be charted.
  2. Select the appropriate control chart using the Control Chart Decision Tree available as a download to determine which type will best fit your situation.
  3. Determine sampling method and plan
    • Determine the statistically significant sample size required taking into priorities. As a rule of thumb, collect at least 20 -25 groups of samples.
    • As much as possible, obtain the samples under the same technical conditions: the same machine, operator, shift and so on.
    • Frequency of sampling will depend on whether you are able to recognize patterns in the data. Consider hourly, daily, shifts, monthly, annually and so on. Once the process is “in control” you might want to consider the frequency with which you sample.
    • Consider using historical data to baseline.
  4. Initiate data collection
    • Run the process untouched and gather sample data.
  5. Software packages such as Minitab or Excel do all the calculations and this is the recommended approach to create the appropriate control chart.
  6. Interpret the Control Chart
    • Determine if the process mean (centre line) is where it should be relative to your customer specification or your internal business objective. If not, then it is an indication that your process is not currently capable of meeting the objective.
    • Analyse the data relative to the control limits, distinguishing between common causes and special causes. The fluctuation of the data points within the limits results from variation inherent in the process e.g design, choice of equipment, preventative maintenance and can only be affected by changing the process.
    • However, points outside the control limits or non random patterns within the limits come from special causes e.g., human errors, unplanned events, freak occurrences that is not part of the way the process normally, or is present because of an unlikely combination of process steps. Special causes must be eliminated before the Control Chart can be used as a monitoring tool. Once this is done the process will be in “statistical control” and samples can be taken at regular interval to ensure that the process hasn’t fundamentally changed.

Downloads

Search – Measure

ANALYSE

Purpose

  • Now we know what is happening, we need to ANALYSE and understand why it happens
  • This will help us distinguish the causes from the symptoms
  • Then we can identify which elements of the problem we should solve first

The analyse step builds on process understanding developed in the Measure step to help pinpoint the specific causes of the problem.

This is where you will organise your data to identify gaps between current and goal performance an look for process problems and improvement opportunities.

You should analyse process maps by identifying which steps directly add value to meeting your customer requirements. This is the time to look for common sources of process variation such as bottlenecks, rework, lack of compliance to standard operating procedures and the nine elements of waste.

A cause and effect diagram can help you break down and identify root causes of the problems in the process.

It is important to apply statistical rigour to test your hypothesis and verify the root causes and projected benefit in numerical terms.

Output

  • Detailed “As Is” process maps analysed (Value / Non-Value Added)
  • A prioritised list of potential sources of variation of data
  • Reduced list of potential key input variables that affect the output(s)
  • Graphical and Statistical analysis of “vital few” root causes
  • Revised project charter
  • Quantify the Cost of Poor Quality (COPQ)

Tools

  • Cause & Effect (fishbone)
    • Start to collect and organise thoughts around possible causes
  • Root Cause Analysis
    • Identify the vital few root causes to the problem and factual justification backed by data analysis
  • Value Analysis
  • Hypothesis Testing
  • Design of Experimentation (DoE)

Moments of Truth Analysis

Purpose

Moment of Truth is any point at which the customer draws a critical judgement (positive or negative) of the quality / value delivered. Simply put, any interaction with a client is “a moment of truth”. It tests the connection between the process team’s value proposition, strategy, implementation, staff, skills and systems and it generates the experience that lasts in the customer’s memory until they interact with the process team again.

A “moments of truth analysis” therefore starts by identifying the interactions a process team has with its customers, stakeholders and suppliers.

It helps to identify;

  • What are the critical steps that “have to go right” in order to consistently deliver value to the customer?
  • At what point(s) does the customer get an impression about the process?

Method

  1. When do customers interact with the team?
    1. Review the list of CTQ’s generated during the Define phase of your project.
    2. For each CTQ, identify the times and context in which the customer interacts with the team. These are the “moments of truth”.
    3. For each moment of truth, find out why the customer is interacting with the process team and what they want to achieve out of the interaction.
  2. What happens when customers interact with the team?
    The next step is to understand how the team interacts with the customer in the moment of truth.

    • Who initiates the contact?
    • Is it a one off thing or part of an ongoing process?
    • What happens?
  3. Mapping the interactions from the process team’s perspectiveThis is the point where you start to analyse anything that could go wrong or be improved.For each interaction you can ask questions such as:
    • How often does this occur?
    • How long does it typically take?
    • What would cause this?
    • What would happen if it did?
    • What would happen next?

Assess the supporting processes if relevant

Moments of Truth

Once you have mapped out the moments of truth you can also map the internal processes that the team performs in order to be ready for those moments of truth.

Hand-off & Bottleneck Analysis

Purpose

Handoffs: Where an item or service is passed across the boundary between two distinct organisational functions inside or outside the organisation

A Bottleneck is any process step with a capacity equal to or less than the demand placed upon it

The purpose of the Hand-off, Bottleneck Analysis is to help explore potential causes of variation and address non value add steps and process waste.

Method

Use process maps and data analysis to guide you to pinpoint potential causes (sources of process variation).

Handoffs

  • Many problems originate at Handoffs – general rule: fewer handoffs is better.
  • Handoffs are points in the process where things typically start to break down and Bottlenecks are created.
  • Handoffs can be used to define internal customer boundaries and remove Bottlenecks from the process.

How well does the process add value? If you identify:

  • Gaps
  • Disconnects
  • Overlaps
  • Hand-offs
  • Duplication of Effort
  • Constraints
  • Unnecessary process steps

Then you may be able to implement some very valuable actions quickly.

Firstly, avoid trying to analyse or improve the actual value-adding activities themselves – people have focused on these for years so your efforts will likely result in small, incremental improvements. But if you focus on the gaps, you may be able to make order-of-magnitude changes to drive out sources of process variation.

Day In Life Of (DILO) Studies

Purpose

Day in the Life of Studies (DILOs) gives us a powerful insight into what’s really happening “on the shop floor”…

What is a DILO ?

A DILO is a short interval review of each and every activity performed by an individual as well as observations, quotes, comments or any other data collected by the person performing the study.

Why use a DILO ?

  • To understand what really happens day-to-day in the operation
  • To identify the types, quantity and causes of non-value added activities
  • To assess behaviors and effectiveness
  • To identify operational issues, their root causes and impactability
  • To gain an understanding of the environment and culture

Method

Day In Life Of Studies (DILOs)

The objective of the study is not a personal evaluation, but to understand what issues and problems arise on a day to day basis and how an individual reacts to the problems. This in turn will highlight many issues and root causes.

Emphasize that our interest is in understanding the problems they typically face, the systems and processes they use – it is not an individual evaluation or a time & motion exercise.

  • The study should be conducted over a full shift or working day
  • Know beforehand what the individual’s working hours (including time and duration of breaks) should be – so you can assess late starts, break over-runs, early finishes
  • Focus on identifying and quantifying value-added (VA) vs non-value-added (NVA) time
  • Color coded narrative : VA is green, NVA is red
  • Do ask, as well as objectively evaluate, whether it is a “typical” day
  • Detail, detail, detail, detail. Collect as much detail during the study as is possible
  • Take detailed notes – don’t rely on your memory – and get copies of critical documents (e.g. the daily plan, the work order, SOP)
  • Don’t transition from one color code to the next without an explanation – must be clear why each block of time is coded the way it is
  • Don’t comment on or criticize what you see while it’s happening; be sensitive to the perspective of the individual you’re studying
  • The text of the DILO should refer to the person you are following as ‘subject’—no names.

Downloads

Search – Analyse

The nine elements of Waste

Purpose

Eliminating waste is one of the most central themes in the Lean philosophy. Everything that cannot be characterized as directly value-adding for the end-user customer is to be regarded as waste.

Toyota clustered the unnecessary activities – waste – into seven different categories.

In addition to the original seven, it is nowadays also customary to add “misuse of resources and ignoring creativity” and “searching” as the eighth and ninth. Let us look closer at these.

Method

Using process maps generated earlier during your DMAIC project, review process steps and tasks to decide if these

Transportation

If one neglects to plan the production well enough in advance, situations will likely arise later where half-finished items for example, are needlessly transported between various sites in the plant.

The time used for transport is not directly creating value for the customers, and should therefore be kept to an absolute minimum.

Inventory

Unfinished products are often found sitting around, waiting to be completed, or finished products are waiting to be shipped out to the customer.

This type of storage can occur in an attempt to reduce unit costs. The reasoning is, “By producing many units of one type, the unit cost will be low.”

But what can happen is that the next stage in production line is not ready to receive and process the units produced.

For that reason, they must be placed in interim storage, where they tie up capital and incur costs. If, in addition, one does not have a clear over-view of demand, it is also easy to produce too many of the finished goods, leading to the goods having to be stored, again, leading to higher costs.

Interim storage or inventory of work in progress (buffers) makes the organization feel comfortable about having enough products in stock for value adding activities. Unfortunately stocks have a tendency to hide the critical bottlenecks in the process flow.

We always have to strive find the right level for all kinds of inventory, either it is raw materials, unfinished products or finished products.

Motion

It is natural for people working with a product to move around a certain amount in order to carry out their job. However, one often observes that employees move around unnecessarily and thus waste time on movements that do not create any direct value.

For example, this may be because the worker must search for parts at another section of the building or that office staff must go from one end of a department to the other in order to look for, or deliver something. Unnecessary motion does not add any value and thereby increases the company’s costs.

Waiting

It is a waste that persons have to wait unnecessarily before beginning the next step of their work process,
There are surely bottlenecks at various stages of a production chain, and if it is congested at one site and the people at the next stage in the process are working faster or have a greater capacity than previous processes, we can easily end up with quite a bit of waiting. This waiting is not value creating and therefore constitutes a waste of resources and an extra cost.

Overprocessing

This is a type of waste that is somewhat less specific than the others and more difficult to address. Some examples of this type of waste, are when e-mails are printed in order to be stored in hard copy, when extra time is used to grind off the rough edges after a piece of material has been cut, or when surfaces of components are being made neat looking only to appear in background structures never visible for the end user. If the law does not require e-mails to be stored in hard copy form, then this is completely unnecessary, and it should be possible to cut material properly on the first try.

Furthermore, it is a waste when machinery is not good enough and is more expensive to operate than it ought to be.

Overproduction

As mentioned earlier, it can be difficult to produce precisely the amount of product that is needed relative to the expected demand.

What often happens is that companies tend to over-produce instead, and use an unnecessary amount of money and resources to produce goods that no one will buy. Or at least not purchase until much later.

Defects and Corrections

If there is a defect in a finished goods or service, this also constitutes a form of waste.

If the defect is discovered before the product is sold, then this can always be fixed, but it costs money.

If the product reaches the market, it will likely be sent back or corrected at a later time. This causes both additional cost and the risk of lost re-purchase from the customer. The latter situation represents perhaps one of the most serious consequences of defects and poor production quality. Customer value and satisfaction is the focal point for a Lean company. Everything that can have a negative impact on customer value and customer satisfaction must be moved to the top of the priority list as quickly as possible.

Underutilized human capacity

Many have not identified their employees’ skills and are unaware of the competencies and creativity residing in their employees. It is common that individuals who are either over- or under-qualified occupy the wrong positions.
Promotions are traditionally used as rewards. This can have unfortunate consequences and there are lots of examples of promotions leading to the company losing a good, perhaps the best, specialist and getting a mediocre middle manager instead.

Searching

Of all the ways in which waste can occur, perhaps searching is the easiest to recognize. In this, most of us already have lots of experience; searching around is universal, and it occurs both in offices, production facilities and on computers.

We search for tools, raw materials, information and consumer goods. And it is only a small comfort that “everything pops up again so long as it has not been thrown away or stolen”. Time is lost, hindering full utilization. Time spent searching could have been used for value creation.

5 Why Root Cause Analysis

Purpose

The primary goal of the 5 Why technique is to determine the root cause of a defect or problem. (The “5” in the name derives from an empirical observation on the number of iterations typically required to resolve the problem).

5 Why Example

  1. The patient was late in the operating theatre, it caused a delay. Why?
  2. There was a long wait for a trolley. Why?
  3. A replacement trolley had to be found. Why?
  4. The original trolley’s safety rail was worn and had eventually broken. Why?
  5. It had not been regularly checked for wear and tear. Why?

The root cause – there is no equipment maintenance schedule. Setting up a proper maintenance schedule helps ensure that patients should never again be late due to faulty equipment. This reduces delays and improves flow. If you simply repair the trolley or do a one-off safety rail check, the problem may happen again sometime in the future.

Method

The basic premise of the technique is that by repeating why (something happens) five times, the nature of the problem as well as its solution becomes clear. The questioning could be taken further to a sixth, seventh, or higher level until you have identified a true actionable cause to a problem – a Root Cause.

  1. Write down the specific problem. Writing it down helps you formalise the problem and describe it accurately. It also helps a team focus on the same problem
  2. Use brainstorming to ask why the problem occurs then, write the answer down below
  3. If this answer doesn’t identify the source of the problem, ask ‘why?’ again and write that answer down
  4. Loop back to step three until the team agrees that they have identified the problem’s root cause. Again, this may take fewer or more than five ‘whys?’

General Guidelines

  • A cross-functional team should be used to problem solve
  • Don’t jump to conclusions or assume the answer is obvious
  • Be absolutely objective

A key phrase to keep in mind in any 5 Why exercise is “people do not fail, processes do”.

Downloads

Search – Analyse

Cause & Effect (Fishbone) Diagram

Purpose

The Cause-effect diagram is a simple tool to connect causes and effects.
It is also called the ‘Fishbone’ diagram, because of its shape, or ‘Ishikawa’ diagram, after its originator, Kaoru Ishikawa.

This tool allows a team to identify, explore and graphically display, in increasing detail, all of the possible causes related to a problem to discover its root cause(s).

Method

It is usually used as a brainstorming framework to find causes, as follows:

  1. Write down the ‘effect’ in a box at the right of the page.
  2. Draw a long ‘backbone’ arrow to the right of the effect box, pointing into the box.
  3. Add smaller boxes containing major cause areas, with angled arrows pointing into the backbone.
  4. If you are not sure about the major areas, use the ‘6 Ms’, as in the diagram.
  5. Use brainstorming to collect ideas for possible causes, writing them on sub-arrows that point into the cause area arrows.
  6. Note that are not actual causes yet – these are selected later.
  7. You can write these possible causes on Post-its, so you can reorganise them later.
  8. Use voting to select the probable key causes.
  9. Look for the causes that if you fixed would result in the biggest improvement.
  10. Perform experiments or otherwise use data to prove that these probable causes are actual key causes, as you had believed.

Downloads

Search – Analyse

Identify Common vs Special Causes of Variation

Purpose

Sources of Variation comes in 2 types – and we treat the 2 types differently.

Special Cause Variation (Assignable)

  • Accounts for approx. 15% of all Variation
  • Unpredictable
  • Easily detected and corrected
  • May come and go sporadically

A process with special cause variation is called unstable

Common Cause Variation (Environmental)

  • Accounts for approx.. 85% of all Variation
  • Predictable
  • Constant and difficult to eliminate
  • Common to all occasions and places
  • Degree of presence varies
  • Each cause contributes a small effect to the variation in results

A process with only common cause variation is called stable

Method

Dealing with Special Cause variation

Goal is to eliminate the specific special causes
to make an unstable process stable

  • Get timely data so special causes are signaled quickly
  • Identify what is ‘special’
  • Immediately search for a cause. Find out what was different on that occasion.
  • Isolate the deepest cause you can affect.
  • Develop a longer-term remedy that will prevent that special cause from recurring
  • Or, if the special cause produced a better result, implement the special cause such that it becomes a positive common cause

Special Cause variation requires immediate diagnosis and action

Dealing with Common Cause variation

  • Disciplined data analysis to determine root causes
  • Focused experimentation to test potential solutions
  • On-going control to monitor effectiveness and potential for further improvement

Common cause variation requires more careful, longer-term treatment

Understand different data distributions

Purpose

What Good are Mathematical Distributions?

  • Distributions are the probability curve of a random effect occurring
  • We use critical values, based on our level of risk, to determine significance as compared to our observed values
  • Percent probability represents risk
  • Significant events happen are least likely to be due to random chance
  • Lookup tables and statistical software is used to calculate and correlate risk and test values

Method

  • Mathematical Distributions are the basis for inferential statistics
    • We go from what we know to what we don’t know
    • We know we have sample data
    • We deduce that our data fits an appropriate mathematical distribution
    • We apply statistical principles and make assumptions based on each distribution
    • We draw conclusions based on the behavior of these mathematical distributions
    • Construct a confidence interval
    • Test for differences in samples
    • Examine associations between samples
  • Some Different Types of Data Distributions
    • Normal Distribution (The “Bell” shape curve)
    • Poisson Distribution
    • T Distribution
    • Binomial Distribution
    • Chi Square Distribution
  • There are many other types of Distributions

Scatter Plot

Purpose

Scatter Plots (also called scatter diagrams) are used to investigate the possible relationship (or correlation) between a y or Dependent variable and an x or Independent variable that both relate to the same “event.”

Method

Scatterplots are used to analyze patterns between a y and an x variable. These patterns are described in terms of linearity, slope, and strength.

y is called the dependent variable or response variable

x is called the independent variable, predictor variable or control variable

Scatter Plot

Things to look for:

  • Imagine drawing a straight line or curve through the data so that it “fits” as well as possible. The more the points cluster closely around the imaginary line of best fit, the stronger the relationship that exists between the two variables.
  • If it is hard to see where you would draw a line, and if the points show no significant clustering, there is probably no correlation.
  • Linearity refers to whether a data pattern is linear (straight) or nonlinear (curved).
  • Slope refers to the direction of change in variable Y when variable X gets bigger. If variable Y also gets bigger, the slope is positive; but if variable Y gets smaller, the slope is negative.
  • Strength refers to the degree of “scatter” in the plot. If the dots are widely spread, the relationship between variables is weak. If the dots are concentrated around a line, the relationship is strong.

Additionally, scatterplots can reveal unusual features in data sets, such as clusters, gaps, and outliers.

Caution! – Correlation does not imply causation! A third variable may be ‘lurking’ that causes both x and y to vary

Downloads

Search – Analyse

Correlation & Regression

Purpose

Correlation is a measure of the strength of linear association between two quantitative variables with the aid of data (Eg. Pressure and Yield)

  • Measures the degree of linearity between two variables assumed to be completely independent of each other
  • Pearson Correlation coefficient is a way of measuring the strength of the correlation
  • Describes the nature of relationship with the help of an equation which can be used for predicting and forecasting
  • A significance test is used to draw a conclusion about a population, using data from a sample

Regression provides an equation describing the nature of relationship useful for predicting and forecasting.

Method

  • Correlation Coefficients are numbers that indicate the strength of the relationship between two factors (linear only)r – strength and the direction of the relationshipr2 – percentage of variation in Y attributable to the independent or X variable.
  • Add precision to visual judgment of correlation
  • Test the power of your hypothesisHow much influence does this factor have?Are there other, more important “vital few” causes?
  • NOTE: A drawback of the correlation coefficient is that it does NOT provide a meaningful number for curvilinear relationships. For those, you’ll need to plot the data.
  • Correlation Coefficients are numbers that indicate the strength of the relationship between two factors (linear only)r – strength and the direction of the relationship.r2 – percentage of variation in Y attributable to the independent or X variable.
  • Adds precision to visual judgment of correlation
  • Tests the power of your hypothesis to identify important “vital few” causes?

Interpreting Pearson Correlation Coefficient, r

  • Falls on or between -1 and 1Calculated in Minitab or MS Excel
  • Figures below -0.65 and above 0.65 can be considered meaningful correlation*1 = “Perfect” positive correlation-1 = “Perfect” negative correlation
  • The MS Excel formula for this function is: =PEARSON(A1:Ax,B1:Bx)



Regression Equation

Y = b0 +b1X1

b0 = Predicted Value Of Y When X1 = 0

b1 = Slope Of Line. Change In Y Per Unit Change In X1

Hypothesis Testing

Purpose

In hypothesis testing, relatively small samples are used to answer questions about population parameters (inferential statistics) to identify sources of variability using historical or current data:

There is always a chance that the selected sample is not representative of the population; therefore, there is always a chance that the conclusion obtained is wrong.

The general goal of a hypothesis test is to rule out chance (sampling error) as a plausible explanation for the results from a research study.

Method

Step 1

  • State the hypotheses and select an α level.A hypothesis is a statement that might or might not be true.In a significance test we use sample data to decide whether or not the null hypothesis is true.

    The null hypothesis refers to the population and can be written before we even take a sample.

    It is often written as an equation.

    e.g. Population mean = 0.57 kg

    e.g. Population correlation = 0.0

    • The null hypothesis, H0, always states that there is no effect (no change, no difference). Assume the status quo.
    • The α level establishes a criterion, or “cut-off”, for making a decision about the null hypothesis. The alpha level is typically assumed to be 5% but it can be set to other values.

Simplified Perspective

Null Hypothesis (Hspan class=”sub”>0) e.g. defendant assumed innocent

Prosecuting attorney must provide evidence beyond reasonable doubt that assumption is not true. Reasonable doubt = α

Step 2

  • With some assumptions, inferential statistics allows the estimation of the probability of getting an “odd” sample and quantifies the probability (p-value) of a wrong conclusion.
  • The p-value is calculated from the sample data.
  • The p-value is the probability of getting such extreme data if the null hypothesis were true.
  • Whenever we obtain a small p-value, we know that, either:The null hypothesis is not true, orThe null hypothesis is true, but we have drawn a sample which is not representative of the population, or

    One of the assumptions underlying the significance tests is not satisfied

Step 3

  • Interpret the p-value
    • What is a “small value”? Typically, p<0.05
    • If the p-value is less than 0.05, we reject the null hypothesis.
    • If the p-value is greater than 0.05, we cannot reject the null hypothesis.

IMPROVE

Purpose

  • We now need to generate some ideas to remedy the symptoms and fix the root cause.
  • It may be sensible to tackle the root causes one at a time, or at least, not all at once.
  • The solutions will need to be selected, planned and then tested prior to implementation.

In the Improve step solutions are generated and the ones best addressing the root cause(s) are selected and implemented to prevent the problem from reoccurring.

To generate solutions you first need to develop boundaries of the kinds of alternatives you will consider and then be creative in considering all the different alternatives.

Revisit your process map and develop a new “future state” process map describing how the process should work. This is a good time to apply Lean thinking as a set of guiding principles to make improvements.

Check that your solution has addressed the key process problems identified earlier and consider running a pilot first to validate the potential impact of the solution.

Once you have a validated solution, it is time for implementation planning.

Successful implementation will depend on both a plan for implementing the technical aspects of the change and a strategy for dealing with the change resistance to get ‘but-in’ from key stakeholders.

Output

  1. Solutions generated to address root causes
  2. “Should-Be” process map agreed
    • A new (improved) process performance baseline and capability
  3. Cost Benefit Analysis for Proposed Solutions
  4. Validated ‘significance’ of improvements
    • Solutions tested on small scale
    • Solution selected and implemented
  5. Confirmation of planned results

Tools

  • FMEA
    • Preventative thinking – what might go wrong
    • Proactive management of failure modes
  • Poka Yoke (Mistake Proofing)
    • How can we prevent defects from occurring in our Future State process

Apply Critical Thinking to identify potential solutions

Purpose

Much of what we call “intelligence” is our ability to recognise patterns. The human mind is great at recognising patterns. We recognize sequences, cycles, shapes, processes, similarities, probabilities, etc. This is our comfort zone.

Patterns give us the power to understand the world and, as a consequence, they rule our thinking. They become the “rules” and “mental models” we live by. We use them to determine what is “typical”, “likely”, “estimated”. We use the rules and models to infer meaning. From meanings we make assumptions. We act based on the conclusions we reach from rules & assumptions.

The problem is – How do we make creative connections if our minds are full of existing patterns.

Our ‘pattern recognition machines’ (minds) mean that new combinations (thinking laterally and being creative) when identifying solutions is often difficult.

There is a difference between ‘logical’ and ‘lateral’ thinking. Neither is good or bad; better or worse. One is not necessarily more successful than the other. But they are different and you need different tools/techniques during a Lean Six Sigma project. Most of us have a preference for the logical. But that preference doesn’t mean we can’t be creative thinking laterally. It’s a skill which can be learned.

Method

Apply

Brainstorming is a group or individual creativity technique by which efforts are made to find a conclusion for a specific problem by gathering a list of ideas spontaneously contributed by its member(s).

General Brainstorming Guidelines

  1. Define your problem or issue. This is extremely important. Problem Statements should start with: “In what ways might we…?” or “How could we…?” Be concise and to the point.
  2. Give yourselves a time limit.
  3. Once the brainstorming starts, participants shout out solutions to the problem while the facilitator writes them down – usually on a white board or flip-chart for all to see. There must be absolutely no criticizing of ideas. No matter how silly, it must be written down. Laughing is to be encouraged. Criticism is not.
  4. Once your time is up, select the ideas which you like best. Make sure everyone involved in the brainstorming session is in agreement.
  5. Write down about five criteria for judging which ideas best solve your problem. Criteria should start with the word “should”, for example, “it should be cost effective”, “it should be legal”, “it should be possible to finish before 2015”, etc.
  6. Give each idea a score of 0 to 5 points depending on how well it meets each criterion. Once all of the ideas have been scored for each criterion, add up the scores.
  7. The idea with the highest score will best solve your problem. But you should keep a record of all of your best ideas and their scores in case your best idea turns out not to be workable.

Mistake Proofing (Poka Yoke)

Purpose

Mistake proofing, or its Japanese equivalent poka-yoke, is a technique for eliminating errors such that it is impossible for an error to occur or makes the error immediately obvious once it has occurred.

Originated by Shigeo Shingo in Japan (pronounced PO-ka yo-KAY)

“poka” means an inadvertent mistake

“yoke” means to prevent

Our minds are not made for repeating the same task or for understanding every detail of what we do. We also have deep drivers that can steer us off course, pushing us to think or do one thing when we should be paying more attention elsewhere.

We need to look at these issues when we think of mistake proofing.

Benefits

  • Enforces operational procedures or sequences
  • Ensures quality at the source instead of quality after the fact
  • Eliminates choices leading to incorrect actions
  • And reduces waste (cost & time)
  • Provides a better working environment
  • Relieves need for constant attention to detail
  • Removes barriers caused by inspection procedures
  • Enables focus on skills rather than problems

Method

Error proofing is a very simple technique. You should keep it in mind at all times, but particularly when you are designing the solution or the improvement.

  1. Obtain or create a process map and/or process risk assessment (FMEA). Review each process step, thinking about where and when human errors are likely to occur.
  2. For each potential error, work back through the process to find the root cause.

Use Shingo’s four error proofing methods. Ideally you should redesign the system to Eliminate the chance of the error taking place. Start with the first, if that can’t produce potential ways to make it impossible to make mistakes. If you can’t do this, you should then try to;

  • Flag (identify quickly, every time the problem occurs),
  • Facilitate (make it difficult to create the problem)

Eliminating Waste

Downloads

Search – Improve

Basic Lean Techniques – 5S, A3, 5 Why and Morning Meetings

Downloads

Search – Improve

Decision Matrix

  • Purpose—what we intend to achieve in this situation
  • Perspective—the context that sets the stage for a decision
  • Scope—the boundary that we use to distinguish what is included and excluded in considering the situation

Decision Quality can be Evaluated by Six Factors

100% is the point at which additional improvement efforts would not be worth their cost.

Downloads

Search – Improve

“Future State” Process Map

Purpose

The “Current State” process chart or map shows the “real” process as it presently exists. As all solutions are being determined and improvements are identified, these are illustrated with a “Future State” map.

A future state map bridges the gulf between the current and ideal states. It incorporates the realities of technical limits, budgets and time.

Method

The primary tool is Group brainstorming which uses the experience and knowledge of all project members. Start with current process map and use the process mapping techniques explained in the Measure stage.

  • Review solution for effect on process steps
  • Eliminate, combine and/or re-sequence
  • Define process boundaries (starting and ending points)
  • Describe where to collect data.
  • It’s an iterative exercise, redraw the process map several times
  • Walk-through the process as it will happen
  • Verify your ‘finished’ map with the people who are involved in the operation of the process

Evaluate Benefits

Purpose

Cost / Benefit Studies are performed once the Future State Process Map has been agreed, to validate if benefits outweigh costs.

  • Consider all costs of implementation
    • Present and future
  • Determine ALL benefits
    • Expense
    • Time
    • Customer/employee satisfaction
    • Safety
    • Others

Method

There are at least three points at which financial calculations should be completed

  • A high level preliminary estimate during the Define phase
  • A detailed financial cost/benefit analysis when improvement actions are identified which is updated during later phases as more information is obtained
  • A final estimate before the project is closed

The cost section of a Cost/Benefit Analysis includes the costs incurred to research and implement the proposed solution and Future State process map. Only incremental costs should be considered in this analysis. Incremental costs are costs which are only being incurred because of your decision to pursue implementation. For example, if you recommended that a new computer be purchased in order to improve customer service, the purchase cost of the computer would be an incremental cost.

Additional examples of project implementation costs would be team members’ time on the project, time to implement improvements and material costs

You may also want to look at Opportunity Costs. These represent the potential benefit that is lost or sacrificed when the choice of one course of action requires the giving up of an alternative course of action

Some Examples Of Direct Costs:

  • Material Scrap Costs
  • Direct Labour Costs
  • Cost Of Capital

Some Examples Of Indirect Costs:

  • Supervisor/Management Review Time
  • Fire-Fighting Time To Resolve Problems
  • Correcting Other Departments Work Before Processing

Consider a trial implementation of a proposed improvement, conducted on a small scale under close observation and scrutiny to verify the financial benefits.

Design of Experimentation (DOE)

Purpose

DOE is an experimental strategy or structured manipulation of controllable factors (independent variables, X’s) at different levels to see their effects on a response (dependant variable, y).

Y = f(x1, x2…..xn)

A DOE IS USEFUL TO

  • Identify important factors. Of all the variables that were identified during the Analyse stages of the project, which drive the largest change in the outcome of the process?
  • Establish process stability. It gives us the ability to set our process at its optimum point and consistently obtain predictable results which is useful during the Improve stage of the project.
  • Find best operating conditions. For the given conditions, we can have the ability to predictably obtain the most out of our process.

Method

  • DOEs are performed in the “Analyze” & “Improve” phases of the “DMAIC” Breakthrough Strategy.
  • DOEs determine the impact that variables have on a product / process characteristic.
  • Process Maps & FMEAs identify the inputs (or x’s) to be investigated and / or validated.
  • Process Maps & FMEAs identify what outputs (or y’s) need to be measured.
  • Measurement System Analysis (MSA) is performed on all important measurement systems before the DOE.

There are several methods of experimentation.

Trial and Error

The objective is to provide a quick fix to a specific problem. The quick fix occurs by randomly and non-randomly making changes to process parameters, often changing 2 or more parameters at the same time. The result often is a “band-aid” fix as the symptoms of the problem are removed, but the cause of the problem goes undetected. In trial and error experimentation knowledge is not expanded but hindered.

One Factor At a Time (OFAT)

The old theory in running an experiment is to hold everything constant and vary only one factor at a time.
With this method, one can be sure that any changes in the response would be due to the manipulated factor. But would they?

Is it reasonable to assume that one can hold all variables constant while manipulating one?

Imagine a large number of possible factors that affect the response variable

How long would OFAT take to identify critical factors and optimize these factors?

Is it possible that the knowledge gained is suspect?

Don’t know the effects of changing one factor while other factors are changing (reality)

Unnecessary experiments may be run

Time to find the factors that have an effect on the response is significant

Full Factorial

An experimental strategy that allows us to answer questions most completely. Full factorials examine every possible combination of factors at the levels tested.

Fractional Factorial

A fractional factorial, as it sounds, looks at only a fraction of all the possible combinations contained in a full factorial. Therefore, if there are many factors being investigated, information can be obtained with less investment.

DOE Terminology

  • RESPONSE
    • A process or product characteristic that is measurable.
    • A response is generally a “Y”.
  • FACTOR
    • A process variables being investigated.
    • A factor is generally an “X”.
  • LEVEL
    • The values or settings at which a factor is evaluated.
    • Example: Oven Temperature (i.e.,100°F & 220°F).
  • INTERACTION
    • The failure of one factor to produce the same effect on the response at different levels of another factor.
    • Example: Time & Temperature.

Downloads

Search – Improve

Piloting

Purpose

When to pilot

  • The scope of the change is large
  • The change could cause far-reaching, unintended consequences
  • Implementing the change is a costly process
  • The change would be difficult to reverse

Method

Plan

  • Identify future state process characteristics to validate:
  • Functional
  • Performance
  • Reliability
  • Identify pilot team
  • Prepare test cases
  • Plan for collecting data

Do

  • Perform the pilot tests
  • Collect data

Check

  • Review data
  • Establish status

Act

  • Act on the results of the pilot. Either:
    • Stop – the solution doesn’t work
    • Modify – change the solution, possibly re-pilot
    • Roll-out – no changes necessary

Even if a pilot shows excellent results, beware of ‘scaling-up’ issues that will make the full implementation different or more complex than a pilot

Change Management

Purpose

Change Management is defined as being the processes and actions to take into account all human and social aspects of the change from the launch of the project to its end…

It ensures the adequate involvement of the people concerned in the definition, communication and implementation of the new solution to secure “Buy-in”.

It covers a range of activities tightly embedded with Project Management activities.

Method

The majority of people who resist change fall in the “Not Knowing&quot; category. They resist because they do not receive adequate information about the change, its benefits or the actions they should take.

Those in the “Not Able to&quot; category resist because they feel unable to make the required changes or work effectively in the new environment with additional support (such as training).

The last category, “Not Willing to”, includes
people who resist because they have
personal concerns which go unanswered
(such as “Will I survive this change?”)

Change management is about taking
all categories of people in the triangle
to engage them in the new model!

Implementation Planning

Purpose

During Implementation the scope and nature of your activity changes and there is a shift from analysis and decision making to actions. But the implementation is still a time for learning and improvement.

There is an increase in task complexity and a decrease in your control over the tasks. There are greater numbers of people, all those directly involved in the solution and all those indirectly involved in its impact.

There may be a requirement to review the team make-up and you will definitely need to re-plan using Project Management principles.

Resources such as time, money, equipment and commitment become even more critical. All of this is about transferring responsibility from the improvement team to the process team (the people who run the process).

This transfer must be explicitly clear.

Method

Key Features of Implementing an Improvement Plan

  • Project Management
    • Define your objectives
    • Clarify the project scope
    • Understand your constraints
    • Identify key milestones and deliverables
    • Plan activities and tasks to be performed
  • Commence implementation
    • Assess Change Readiness and phase the implementation accordingly
    • Agree start
    • Review and adapt plan to suit calendar
    • Kick off meeting
    • Communication with Stakeholders and Project Team
  • Monitor progress
    • Resource draw down
    • Review team behaviour charter
    • Review team/project KPIs
    • Review/refine/develop implementation charter
  • Communicate progress
    • Issue Reports
    • Control Budgets
    • Manage/Mitigate Risks
    • Handle Roadblocks
    • Verification of Actions
  • Ensure successful completion
    • Deliverables meet plan?
    • Data/results collection
    • Quick wins and unexpected gains/losses
    • Sign off
    • Modification to plan
    • People issues

Downloads

Search – Improve

CONTROL

Purpose

CONTROL ensures that you hold the results and establish Standard Operating Procedures to reinforce new pratices that are followed and at this stage all of the key elements of process managementshould be coming together.

This is also the time to take stock of the journey and transition back to the operation for on-going monitoring.

The purpose of the Control step in the DMAIC roadmap is to ensure that the process improvements, once implemented, will “hold the gains”, so that the problems dealt with earlier, stay solved.

This is done through documentation, on-going monitoring and ensuring all key stakeholders are committed to the new ways of doing things.

The new process must be documented to reinforce compliance to new practices, procedures, policies and communicate what is required in a user friendly and accessible way.

On-going monitoring of the process is important to ensure that the process is stable, reporting predictable results and to look for further opportunities for improvements.

Monitoring requires that you establish standards for key input, process and output measures and control charts are a good tool for monitoring the stability of your process. Specific guidance is also provided to the process team regarding corrective actions to take if the control chart shows that your measurement is out of control.

The Lean Six Sigma mind-set and attitude reinforces the need for the process team to look for improvements on a continuous basis through change initiatives such as Kaizen events.

Output

  • Assessed Impact of the Solution on the Key Measures
  • Documentation required to maintain improvements
  • Monitoring Plan
  • Statistically validated evidence to prove Process is Standardised and meeting targets
  • FMEA is being used on a regular basis
  • Align organisational systems and structures to reinforce and anable the new process

Tools

Control Plan

  • Overall document explaining roles, responsibilities, monitoring measures, managing exceptions, response mechanism of process goes out of control
  • SPC
  • Mistake Proofing
  • FMEA

Develop Standard Operating Procedures (SOP’s)

Purpose

A Standard Operating Procedure (SOP) is the documented sequence of steps and other instructions necessary to carry out the required activities within a process.

It’s primary audience is all those directly involved in operating the process.

Method

The following six-step procedure will be useful and effective in most situations.

  1. Name the SOP using descriptive action words. Examples: Recording, Identifying and Handling.
  2. Write a scope for the SOP. To do this, answer these questions: Which specific
    operations or tasks will be covered? Which are not covered? Who is the SOP written for?
  3. Develop an overall task description including sub-tasks. Consider using a series of pictures or a flow chart or audio instructions in addition to written information.
  4. Describe each task in detail. In this section include the following:
    • Specific order in which activities are done
    • Timing sequences and times allowed
    • Materials, IT systems or tools used and how they are used
    • References to other associated SOPs
  5. Get everyone on board. Inform everyone about the SOP and where to find it. Talk with all employees to gain agreement that procedures and expectations are understood and achievable.
  6. Set up a system to monitor the SOP regularly.

General Pitfalls in writing SOP’s

  • Not involving the affected persons in the creation of the procedures
  • Not readily available
  • Obsolete procedures not destroyed
  • Procedure documentation not readily available
  • Not testing the procedures
  • Omitting information
  • Results to be obtained
  • How to do a step
  • Lowering the importance of procedures
  • No method to update procedures
  • Ignored by management

RACI Chart

Purpose

RACI Charting is used to ensure clarity on ‘who will do what’. It is a key tool for clarifying roles and responsibilities within the “Future State” process but can also be used as part of the Analyse stage.

The scope includes critical activities / steps / decisions / Information and communication flows

The focus should be to ensures that Accountability for an activity / decision is clearly defined and understood within the business

The RACI Matrix (below) is built on the assumption that only 1 role/person is Accountable for an action or decision, although many others many be involved in the ‘doing’ of the task or activity

  • Highlights the primary co-ordination mechanisms and hand-off points required to ensure that the flow of the process is sustained
  • Helps to optimise information and communication flows

Method

RESPONSIBLE: Has responsibility for completion / execution of the task. May actually do the task him / her self

ACCOUNTABLE: Has overall ownership of the decision / task & the results or outcome of actions taken

CONSULT: Gives advice or information to support the decision / task. There is a two-way communication (getting their inputs in the action / decision too)

INFORM: Is given information after the decision or action. There is just a one-way communication (just letting them know).

The difference between Consulted and Informed categories is that in the former case and in the latter case

  1. Define action/decision
  2. Define primary roles
  3. Prepare RACI Matrix
  4. Facilitate RACI session involving all key process stakeholders
  5. Ensure agreement on outputs and decisions
  6. Document the outputs and align with procedures and job descriptions

Skills Matrix

Purpose

Once we know what role is performing what task with the process, we can assess existing resource capabilities. The Skills Matrix is a simple visual tool to aid in the management, control & monitoring of skill levels and to aid resource planning.

It is best used to quickly identify current available skills and future requirements.

  • To examine where our strengths & weaknesses are.
  • As a day to day planning tool to use skills where they are most needed.
  • As a planning tool to organise adequate cover for holiday & sickness.
  • If done fairly, to keep employees motivated & reduce boredom.

Method

Plan

  • Establish all tasks required in the team or area.
  • Establish all the team members.

Do

  • Develop all procedures for all tasks.
  • Define skills required for each task.
  • Define level of ability/competency.
  • Define how we measure performance.

Check

  • Measure team members against procedures & performance measures.

Act

  • Make visual on the skills matrix the ability or competency for each team member against each task.

Skills match the detailed procedures and where possible contain information on:

Tasks and activities

Standard Operations

Timings and frequencies

Standard cycle times

Service specifications

Design Specifications

Product and Service quality characteristics

Quality Specifications

Statistical Process Control

Purpose

SPC is used to monitor the consistency of process performance against statistically predicted patterns to monitor the Voice of the Process. It is used to identify the source and nature of potential process improvement opportunities – Common and Special Cause variation. It helps you read process data so you can see and understand what is happening.

If you can know what is happening, then you can start to think about if any action is necessary and if so, what action to take.

Using SPC provides the ability to make appropriate responses to variations which ‘disrupt’ process effectiveness and efficiency. When a process is performing within the predicted pattern (limits), we say that it is ‘in control’.

However, SPC does not necessarily include measurement of performance against Customer Value requirements. A process may be ‘in control’ but not producing to customer specification.

Method

Statistical Process Control uses Control Charts to monitor and analyse process variation and determine if a process is operating “in control”. A control chart is a run chart with control limits.

Control Chart

  • There are different types of Control Charts.
  • Software such as Minitab packages do all the calculations and create the control charts.
  • All types of Control Chart, are built from the same principles – it is only the statistics in calculating control lines that differ.
  • For continuous data, if the sample size is:
    • 1, then use Individual and Moving Range chart (I-MR)
    • 2-9, them use X-bar and Range charts (Xbar-R)
    • Greater than 9, then use X-bar and S-charts
  • For discrete data, determine if measuring defectives or defects
    • Defectives, equal sample size: use np-charts
    • Defectives, unequal sample size: use p-charts
    • Defects, equal sample size: use c-charts
    • Defects, unequal sample size: use u-charts

Statistical Process Control Flowchart

  1. Look for obvious trends, shifts, cycles, etc. Go through this loop one time only
  2. Collect a minimum of 30 data points on the run chart before calculating control limits
  3. If possible, identify and eliminate special causes for points out of limits

After three times through this loop, use limits and watch control chart to see if it settles down, i.e., no more special causes

After about 30 samples, recalculate limits

Response Plan

Purpose

A response plan is a documented method for how a process owner/team should respond to any out-of-control conditions that may occur in a process. A suite of Key Performance Indicator’s (KPI’s) to prioritise what needs to be looked at.

A good response plan will help ensure a timely and appropriate response to processing problems as they occur, thereby decreasing the risk of defects getting to the customer. It contains of 6-8 key measures to help us drive performance to the CTQ.

All Xs and Ys that are being monitored as part of the new process must be included in the response plan and reviewed by all process stakeholders.

Method

  • Provide a response plan for each monitored X and Y, with:
    • Specific action to be taken.
    • Timing of action.
    • Owner of action.
  • Tools for developing the Response Plan.
    • Common sense and experience/knowledge of those who work in the process.
    • Failure Mode and Effects Analysis FMEA.
    • Statistical Process Control.
  • And consider whether Error Proofing might also help.

A response plan could simply be an Action Log which includes measures of key drivers of process perfomance.

For each measurement tracked, document the specific action to be taken when an out-of-control event occurs. The action that is taken will depend on the severity of the situation. If the situation is minor, specific corrective actions can be executed as planned. In the event of a serious problem, the process owner may need to become more involved.

The plan should specify a time frame in which the team needs to respond for each out-of-control situation.

Finally, the person responsible for each X and Y must be named in the response plan.

Be specific about action to be taken. Remember to operationally define any terms that may be ambiguous so that no confusion exists about the specific response needed.

Key Questions to Ask.

  • Where data is taken on the process
  • Who takes the data
  • How (by what methods) measurements are taken and recorded
  • When (how often) data is collected
  • Includes Measurement Systems Analysis (MSA) to ensure integrity of the data
  • Who takes action based on the data
  • What action to take
  • Where to find trouble-shooting procedures

Downloads

Search – Control

Failure Mode Effects Analysis (FMEA) to assess Process Risk

Purpose

Failure Mode and Effects Analysis (FMEA) was devised by NASA early in the US Apollo space program in the 1960s.

NASA created the tool to alleviate the stress between two conflicting objectives:

  • “To successfully complete the mission and return the crew”
  • “To predict the unavoidable failure of at least some components, to prevent them when possible, to plan for them, and to build in the ability to overcome them”

In the late 1970s, automotive companies, driven by liability cost issues, began to incorporate FMEA into the management of their processes.

Many service organisations, automotive, aeronautics and electronics companies use FMEA principles to manage design, process and system risk.

Method

FMEA is a technique for driving improvement and it is a ‘live’ action plan for managing risk.

FMEA Chart

The 4 key terms on the FMEA are

  • Severity – Sev. Does it matter if this happens or are the effects of failure so small that we would hardly notice them?
  • Occurrence – Occ. Should we worry about this happening or is it so unlikely that we could use our efforts on other things?
  • Detectability – Det. If this happens, can we identify it or is it completely hidden until much later?
  • Risk Priority Number – RPN. SxOxD which gives us a ranking of the risks of the Failure Modes or ‘Criticality’ , also known as the ‘Risk Priority Number’ or RPN
  1. Identify product or process name and collect all relevant background information.
  2. List “failure modes” by Brainstorming with key process stakeholders who understand most about what is happening with the process or product.
  3. Identify “effects” for each failure mode.
  4. List causes for each one at a time.
  5. Use FMEA Rating Scales to score on a scale of 1 to 10 the Sev, Occ and Det.
  6. Calculate RPN. Sev x Occ x Det.
  7. Identify corrective actions.
  8. Implement corrective actions.
  9. Score and re-calculate RPN to see if it has made a difference.
  10. Update FMEA and use as a live action log for managing risk.

Downloads

Search – Control

Align Organisational Systems & Structures

Purpose

The process interfaces, systems and structures in the organisation must support and reinforce the new behaviours and actions that have been implemented as part of the new process implementation. The following organisational interfaces need to be considered as part of the overall solution.

  1. Staffing: Acquiring / Placing Talent
  2. Development: Building Competence / Capability
  3. Measures and Rewards
  4. Effective Communication
  5. Designing Organisations (Structure)
  6. Information systems (technology, MIS)
  7. Resource allocation systems (e.g. budget, finance, strategy)

Method

Systems and Structures Assessment

Systems and Structures Assessment

REWARDING PERFORMANCE

Using the following response choices, please indicate what would be most
likely to happen if you were to do each of the actions described below:

  • 1 = the action would usually bring reward or approval
  • 2 = the action would probably bring neither approval nor disapproval
  • 3 = the action would probably bring punishment or disapproval
  • 4 = response to this action is unpredictable, ranging from approval to disapproval

Rewarded or Disciplined?

Downloads

Search – Control

Visual Management

Purpose

Visible status and information boards allow timely and consistent information to be available to all. This improves communication through better accuracy (quality) and less searching around (time)

The purpose of putting information on a Visual Management Panel (VMP) is to communicate

And the VMP is used to communicate to suppliers who are visiting, customers who are visiting, people from other departments passing through, senior executives who are visiting in fact anyone with any interest. But it’s main target audience are those who work in the process (or process area)

And it is best used as a meeting point where the team meets (once a day/week/ month) and discusses what is happening in their process

And this facilitates the most effective Control – SELF CONTROL which is the basis for CONTINUOUS IMPROVEMENT

Method

A Typical Daily Management Board
Call Centre Performance Indicators
Visual Management Diagram

Lessons Learned and Best Practice Transfer

Purpose

The purpose of lessons learned is to bring together any insights gained during a project that can be usefully applied on future projects.

Lessons learned are useful; time spent in doing the work better is time well spent. Getting it right the first time is cheaper and easier than doing it now and fixing it later.

Transferring best practices requires the support of expert facilitators and change agents to identify and learn from best practices and applying them in a new configuration or a new location. The key factor in the transfer of best practices is to make the ‘recipients of best practices’ understand the need for the implementation of best practices.

Method

Typical barriers to overcome include:

  1. Senior management’s failure to focus on the importance of transfer of best practices.
  2. Improper sharing of concepts of best practices.
  3. ‘Silo thinking’ in an organisation. It means departments work independently to meet their own goals instead of working together.
  4. Poor communication. Employees being unaware of best practices.
  5. Lack of time.

Besides removing obstacles, organisations have to create an environment, which facilitates the transfer of best practices. Successful organisations have adopted many features to facilitate the transfer of best practices. The most common among them are:

  • Benchmarking teams: They are formed to take care of internal and external benchmarking projects.
  • Best practice teams: Benchmarking teams generally have a clear life-span, where as best practice teams are a permanent part of the networking infrastructure of an organisation.
  • Knowledge and practice networks.

The following are critical factors for the effective transfer of best practices:

  • Having a common and a consistent goal that links best practices to business objectives.
  • Communicating to employees of the organisation, the business needs and importance of best practices.
  • Creating an environment, which encourages sharing of best practices.
  • Providing an updated IT infrastructure to support the sharing of best practices.
  • Resolving the problems of empowerment and team leadership.
  • Reinforcing the required behavioural competencies.
  • Demonstrating practices that are successful in other organisations.

The ‘transfer of best practices’ has the potential to achieve massive benefits and is too important to be ignored.

Downloads

Search – Control

Hints and Tips

Left Arrow
Right Arrow

Define

It is important to define a project that is both important to the business and do-able.

It is difficult to stay focused on the problem without jumping to solutions. Keep focused on the problem at this stage.

Ensure that the resources required are allocated to the project.

Getting sponsor ‘buy-in’ and visible support for the project is a critical success factor.

Change Management starts when your project starts. Focus on involving as many key stakeholders as is appropriate.

Defining the Problem Statement

When bringing together a team to achieve a particular purpose, provide them with a problem statement.

Customer Types

If we do not fully understand our customers’ needs, we cannot even hope to satisfy them.

There is more than one customer for most processes. SIPOC helps to identify them.

You may need to segment your customers into types or categories to make better sense of their differing needs.

It is critical that you think from your customers perspective.

Be certain to focus on what the customer said. Not the interpretations that have been made of their words.

You might need to make sure that the customer understands that simply asking for information is not an intention to act.

There may be a lot of Voice of Customer (VOC) data which already exists – but beware, get the source data, not others’ interpretations and conclusions.

If we do not fully understand out customers’ needs, we cannot even hope to satisfy them.

Translate Customer Needs into Critical to Quality (CTQ) requirements

If we do not fully understand our customers’ needs, we cannot even hope to satisfy them.

There is more than one customer for most processes. SIPOC helps to identify them.

You may need to segment your customers into types or categories to make better sense of their differing needs.

It is critical that you think from your customers perspective.

Be certain to focus on what the customer said. Not the interpretations that have been made of their words.

You might need to make sure that the customer understands that simply asking for information is not an intention to act.

There may be a lot of Voice of Customer (VOC) data which already exists – but beware, get the source data, not others’ interpretations and conclusions.

If we do not fully understand out customers’ needs, we cannot even hope to satisfy them.

Project Charter to start the team on the right footing

You should record any assumptions you make in your charter, so you can refer to them in future revisions.

When bringing together an improvement team to achieve a particular purpose, focus them by providing a draft Project Charter for them to rework.

It is sometimes easy, other times difficult, to quantify the cost of poor quality that is being produced by your process. If scrap or rework is being produced – record and quantify it.

SIPOC Process Mapping

SIPOC is a good high-level ‘presentation’ tools that can help to convey the scope and boundaries of the process to stakeholders

Mapping makes processes visible and tangible so you can see and understand them; then you can improve them

The people who carry out the process must be the people who provide information about the process

Before you start mapping a process, check what work has already been done (and how it was done) and re-use when appropriate

Balance having enough detail to see where problems occur without having too much detail to be able to see problems clearly

Don’t try to construct in one pass, process mapping is an iterative process

Stakeholder Mapping to understand the resistance to change

Analyse stakeholder needs throughout DMAIC but the preparatory work must be done in Define, Measure, Analyse phases.

The project sponsor should play a key role in managing stakeholder expectations.

Ensure there is a shared need for the project amongst the stakeholders.

Be specific about the desired outcome of the project to enable all stakeholders to see the Vision.

Ensure focused communication to all stakeholders, addressing their true concerns and requirements.

The outcome of a stakeholder analysis is some form of communication plan. Communication is two-way – part of the initial communication plan will be to understand why some people hold the views they do.

Don’t give all the communication actions from your stakeholder analysis to one person – share them around the project team. The Sponsor plays a key role in communications.

Don’t try to guess, ignore or analyse the reasons behind someone’s opinion, instead, measure the facts.

Many positions adopted by the different parties may seem to be irrational but there is always a rationale in the individual interests of different stakeholders. Try to understand their point of view.

Build your stakeholder perceptions, through cross information, exchange point of views with team members.

Don’t make the mistake to think people will adopt and stick with an opinion ! Things, actions and positions will move… all the time ! Because that’s what Change management is all about.

The Business Case – Cost of Poor Quality

You should record any assumptions you make in your estimate, so you can refer to them in future revisions

Contact your company Finance Department to help you with financial benefit reporting

Kano Model to prioritise needs

Must Haves are assumed – these are usually unspoken so may not appear in the Voice of Customer analysis

Need to haves are usually expressed explicitly by the customer

Delighters are something that maybe the customers weren’t expecting so they could be unspoken

The Kano model provides a framework for identifying the type of need a customer is expressing

Linkage of CTQ’s to Key Output Measures, Y’s

Use the Needs Statements and Output Characteristics you identified to generate several Critical to Quality (CTQ) statements

Identify those aspects which you feel are most important for ensuring value delivery to the customer

Select the KOMY that you feel most clearly reflects how the customer would evaluate your performance

Generate specific Needs Statements for those groups that you think are most important

Remember: You are focusing on customer needs, not solutions

Validate the need (check) with the customer

Ensure you “hear” what the customer is really saying

You might need to make sure that the customer understands that simply asking for information is not an intention to act

No Guessing! No Interpreting! Get genuine VoC

Measure

Apply any quick wins identified but be sure this is not a solution but an obvious issue which has a straightforward and obvious fix.

Don’t try to measure everything.

Involve the people in the process who know what actually happens.

Be wary of using summarised, aggregated reports and data.

Verify processes and data with the individual people who know most about it – those who do the work.

Avoid getting bogged down in the detail of the process during this stage. We first need an understanding of the whole end-to-end process before we tunnel down into the details.

If you have rework in the process Flowcharting is the right techniques to use to map the process as it will capture the alternative decision path.

Pareto Analysis Focus on vital few

The figures 80 and 20 are illustrative – the Pareto Principle illustrates the lack of symmetry that often appears between work put in and results achieved.

Excel can easily create Pareto Charts.

If necessary collect process data for a period of time to collect real data.

Make sure you have a reasonable sample size. This will make your analysis more meaningful.

Try not to have a category heading of ‘Other’ because it is not definitive.

Calculate Process Yield Metrics

Final Yield ignores the “Hidden Factory” or rework.

Time and resources expended through rework is referred to as the Hidden Factory, because the rework loop diminishes overall process capacity.

Is a complicated process better: NO. Most process steps create more opportunities for errors. We need to reduce the number of opportunities.

Each process step induces another opportunity to make a mistake – create a defect condition.

Data Collection Plan

Accurate data collection is essential to maintaining the integrity of the analysis and minimize the impact of measurement variation.

Distorted findings result in wasted resources and can mislead to pursue fruitless avenues of investigation.

“Count what is countable, measure what is measurable and what is not measurable – make measurable”– Galileo, 17th Century

“When you can measure what you are speaking about and express it in numbers, you know something about it; but when you cannot measure it, when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind”– Kelvin 1883

Data Types, Continuous vs Discrete

Discrete (Attribute or Countable) data has whole numbers (integers)

Ask yourself “Is it possible to have ½ of one of these?” If the answer is “No”, you probably have discrete data

Beware: Percentages seem like continuous data but are often based on discrete data e.g. An exam pass rate (represented as a percentage) is based on discrete data; ‘how many passed?’

Determine Significant Sample Size

Extrapolation from sample to population requires a representative sample.

Data is obtained using samples because we seldom know the entire population.

Avoid Convenience Sampling; Data collected from the easiest source, Sampling only when the problem is seen, Not following a clear Data Collection Plan.

Other types of sampling bias; Environmental bias, caused by changes in external conditions during sampling.

Other types of sampling bias; Response bias, caused by the act of sampling changing the behaviour of the population.

Other types of sampling bias; Measurement bias, Caused by inconsistent operational definitions or data collection procedures.

Flowcharting

Use of ‘brown paper’ approach allows you a larger canvas for your work.

Capture the flow of work as it really occurs.

Do not try to make the flow ‘too neat’.

Be certain to capture the potential ‘loops’ through which activity flows.

Use process mapping to begin to gather data on the frequency of issues or problems within the process.

Remember: You are using the flowchart to describe the process, not the people who are involved.

Ensure that you show alternate courses of action when a ‘gate’ decision occurs.

Avoid two or more successive decision boxes. If you find this is happening, question whether you are really documenting a process or whether you are creating a decision tree?

If the flowchart cannot be drawn on one page then group boxes together and replace them with a single box. The replaced boxes then appear at a lower flowchart level.

Gauge R & R

Don’t plug your data into software and blindly accept the %GRR values.

Statistics is a means to an end, not the end in itself. There has to be a purpose and pay-off for conducting the analysis.

Suggested Actions to improve Repeatability; Repair, replace, adjust equipment, implement SOP.

Suggested Actions to improve Repeatability; Identify the problem gauge and take actions to fix.

Suggested Actions to improve Reproducibility; Training, compliance to SOP.

Suggested Actions to improve Reproducibility; Error Proof (Poka Yoke).

Measurement System Analysis (MSA)

Suggested actions to improve Resolution; Use a measurement device or system that can measure smaller units

Suggested actions to improve Resolution; Measure to as many decimal places as possible

Suggested actions to improve Resolution; Larger sample size may overcome the problem in the long term

Suggested actions to improve Accuracy; Calibrate the measurement device or system when needed/scheduled

Suggested actions to improve Accuracy; Use standard operating procedures to collect data

Suggested actions to improve Accuracy; Create Operational Definitions

Suggested actions to improve Accuracy; Use only in restricted range where linearity exists

Suggested actions to improve Stability; Ensure compliance to standard operating procedures (SOP)

Suggested actions to improve Stability; Change/adjust data to ensure like for like comparison

Suggested actions to improve Stability; Use control charts to better understand stability

Suggested actions to improve Stability; Use/update current SOP

Suggested actions to improve Repeatability; Repair, replace, adjust equipment

Suggested actions to improve Repeatability; Identify the problem gauge and take actions to fix

Suggested actions to improve Reproducibility; Training, Compliance to SOP

Suggested actions to improve Reproducibility; Error Proof (Poka Yoke)

Operational Definitions

Confirm understanding of all definitions with stakeholders.

You may need to implement measures and operational definitions to collect the data if it doesn’t already exist.

Verify operational definitions and data with the individual people who know most about it – those who do the work.

Trust nobody – check all operational definitions.

Don’t try to measure everything.

Value Stream Mapping

Use of ‘brown paper’ approach allows you a larger canvas for your work.

Don’t ever use computer to build the map, the process is too fluid and dynamic – maybe to communicate at the very end but better never.

If you can’t understand it in 20 mins it’s too big and complicated.

The one sheet can be right around the room. The key point is that it should only be on one page.

Calculate Process Yield Metrics

Calculate Defect Metrics Process Sigma and Z Metrics

Histogram is convenient for large amounts of data, particularly when the range is wide.

Generally, at least 20 data points are required for a Run Chart or Histogram.

A Scatter Plot can be used to make predictions of values lying outside the measured range.

Generally at least 30 data points are needed for a Scatter Plot.

Use graphical analysis to identify what causes defects/errors, plotting them on Bar charts and cascading Paretos down to the level where you can work on the causes.

Graphical Analysis

Histogram is convenient for large amounts of data, particularly when the range is wide.

Generally, at least 20 data points are required for a Run Chart or Histogram.

A Scatter Plot can be used to make predictions of values lying outside the measured range.

Generally at least 30 data points are needed for a Scatter Plot.

Use graphical analysis to identify what causes defects/errors, plotting them on Bar charts and cascading Paretos down to the level where you can work on the causes.

Descriptive Statistics to Understand Process Effectiveness vs Efficiency

Range is considered to be a approximation of spread where standard deviation is a much more accurate measure of spread.

When you have small samples n << 30 Range is nearly as good as Standard Deviation; with n > 30 only Standard Deviation should be used.

Be sure to carefully analyse your data collection plan to obtain representative samples.

Be conscious of the validity and integrity of the data, how the data was collected and how the data was interpreted.

Always generate a run chart before doing more sophisticated statistical analysis.

Must always first test sample data prior to analysis to see if it follows a normal distribution.

Baseline Process Capability

While Cp relates to the spread of the process relative to the specification width (tolerance band), it DOES NOT look at how well the process average is centered to the target vale.

Performs tests to check if your continuous data is normal.

If the process is near normal and in statistical control, Cpk can be used to estimate the expected per cent of defective products or services.

Process Capability represents a “snapshot” of the process over a specific time duration.

Control Charts

Control Chart also serves as useful tool for on-going control of a process.

Make sure the samples collected are random. To establish the inherent variation of a process, allow the process to run untouched i.e according to standard operating procedures.

Your process is in “statistical control” if the process is not being affected by special causes. All the points must be randomly dispersed about the centre (mean) line for an in-control process.

Any points outside the control limits, once identified with a cause(s) should be removed and the calculations and control chart redone.

Points within the control limits but showing non random patterns such as trends, shifts, cycles or other instability are also special causes.

When a Control Chart has been initiated and all special causes removed, continue to plot new data on a new chart, but DO NOT recalculate the control limits. As long as the process does not change, the control limits should not be recalculated.

Analyse

Projects typically under commit and over deliver: So this is a good time to use the analysis revisit your projected benefits in your project charter:

Use the ‘5-Whys’ tool to make sure you have identified a true, actionable cause

Understand that you need to verify that your root causes are the actual root causes

It is often difficult to stay focused on causes, without leaping forward into solutions. But you must stay focused on causes up to this point. Problems will arise later if you don’t.

Moments of Truth Analysis

A Moment of Truth is Not restricted to only those points at which we ‘touch’ the customer.

Not every customer contact will be a Moment of Truth.

What measures or service level agreements does the team have in place?. These could be potential Moments of Truth.

Hand-off & Bottleneck Analysis

Where does the process touch the customer (or allow them to judge us)? These are priority places to ensure the process is satisfying customers.

Unless you first understand the value delivered to the external customer you will be unable to distinguish between Value-adding, Value-enabling and Waste activities and Identify the most important gaps and delays.

If you don’t impact the Constraint, you have no Effect.

Lack of end-to-end view based on the external customer’s Value Requirements will generally result in Duplication of Unnecessary actions, reviews or controls, process activities performed in the Wrong sequence, Over-processing.

Start with Customer view of ‘value’, not your internal perception of what you think they require.

Handoffs: Good points for data collection.

Day In Life Of (DILO) Studies

First question to ask is, “What is your job”—what do they do during the day, what are the responsibilities, who do they report to and then observe if this is what they actually did during the day.

DO NOT Comment, criticize or make jokes about the speed or quality of the work; don’t share your judgments on what is VA/NVA during the study.

Understand use of and compliance to systems and Standard Operating Procedures.

Don’t let the subject spend too much time explaining things, wait for intervals in the work which are obvious rather than letting the subject draw you into extended discussion which itself disrupts their day.

Take notes as it happens – You will miss detail if you don’t.

Collect all volume / waste info etc. as it happens – it often cannot be collected afterwards.

Ensure you’ve measured the day of study vs historical performance to see if the day of study is typical.

DO NOT: Use a “stop watch”.

DO NOT: Confront or try and problem solve for them – leave that for the project!

Don’t write vague comments like: “Entered something into computer” YOU MUST know what they are entering, into what system. You can’t always sit quietly. Sometimes you must ask intrusive questions.

DO NOT Forget your common sense – you don’t have to follow people to the bathroom!

The nine elements of Waste

A good definition of value adding is what a customer is willing to pay for. Would you be willing to pay for waiting, counting, administration, moving around, defects,scrap etc.

The potential for lead time improvement is far higher by focusing efforts on waste elimination.

5 Why Root Cause Analysis

Avoid the tendency to stop at symptoms rather than going on to lower-level root causes.

Ensure you have the right level of support to help ask the right “why” questions.

Results are not repeatable – different people using 5 Whys come up with different causes for the same problem.

Avoid the tendency to isolate a single root cause, whereas each question could elicit many different root causes.

5 Why, when combined with other problem solving methods, is a very effective tool.

The real key is to avoid assumptions and logic traps and encourage the team to keep drilling down to the real root cause.

Cause & Effect (Fishbone) Diagram

If ideas are slow in coming, use the major cause categories as triggers to help guide you, e.g, “What in Methods is causing …”.

Key causes are those which, if addressed, would give significant gains – These are usually root causes.

Use the cause and Effect diagram to begin to search for root causes.

Avoid the tendency to stop at symptoms rather than going on to lower-level root causes.

For each cause, continue to push for a deeper understanding, but know when to stop.

Ensure everyone agrees on the problem statement. Include as much information as possible on “what”, “where”, “when” and “how much” of the problem, and try to use data to specify the problem.

Identify Common vs Special Causes of Variation

Common Cause variation is; Always present, Expected and Predictable.

Special Cause variation is; Not always present, Unexpected, Un-predictable.

Dealing with Special Causes – Look for differences between data points and Take action based on differences.

Dealing with Common Causes – Study all the data points and Make basic changes to the process.

Understand different data distributions

Be sure to carefully analyse your data collection plan to obtain representative samples.

Continuous Data is frequently but not always Normally Distributed.

Common Assumptions in Statistical Tests is that the data follows a Normal Distribution. It typically allows us to perform robust statistical analysis to better understand process performance.

Scatter Plot

Correlation tells how much linear association exists between two variables.

Regression provides an equation describing the nature of relationship.

In general, we need a minimum of 30 data points to determine correlation.

Correlation & Regression

A drawback of the correlation coefficient is that it does NOT provide a meaningful number for curvilinear relationships. For those, you’ll need to plot the data.

In general, we need a minimum of 30 data points to determine correlation.

Hypothesis Testing

Hypotheses are statements about population parameters, not sample statistics

Improve

During Improve there is a Shift from analysis and decision making to action, but the implementation is still a time for learning and improvement

Review of team make-up may be useful during Improve and you will definitely need to re-plan (so Project Management becomes important) and the Resources needed (time, money, equipment, commitment) even more critical

At the end of Improve, transfer responsibility from the improvement team to the process team (the people who run the process). This transfer must be explicitly clear.

Apply Creative Thinking to identify potential solutions

Creativity exercises, relaxation exercises or other fun activities before the session can help participants relax their minds so that they will be more creative during the brainstorming session.

Focus on quantity: the greater the number of ideas generated, the greater the chance of producing a radical and effective solution.

Welcome unusual ideas: They can be generated by looking from new perspectives and suspending assumptions. These new ways of thinking may provide better solutions.

Combine and improve ideas: Good ideas may be combined to form a single better good idea.

Mistake Proofing (Poka Yoke)

Errors can appear in all parts of the company – which means error-proofing is everyone’s business.

Remember, it’s not just products that are designed, processes are designed!!!

The more potential errors that are identified and dealt with at the design stage, the fewer that are likely to emerge later.

Error Proofing is Inexpensive, Very effective, Based on Simplicity and Ingenuity.

Decision Matrix

Inappropriate Frames Can Result in Poor Decision-Making

“Future State” Process Map

The people who carry out the process must be the people who are involved in developing the Future State process map.

Describing work as a process applies to all work, whether repetitive in nature or ‘one-of-a-kind.’

Evaluate Benefits

Concentrate on direct costs and direct benefits

Use indirect costs that are generally acceptable to all stakeholders

Use the process map and personnel in associated departments to identify cost and benefit information

Keep the analysis simple, focus on cost of implementation and a few key benefits that clearly exceed the cost

Use standard methods and rates in your calculations

List all activities that contribute to either cost or benefit and identify as much as possible how these activities will be measured

Keep the presentation simple and easy to understand

Design of Experimentation (DOE)

It takes money, time and effort to perform DOE.

A range methods that are available to statistically validate the results of a DOE. ANOVA is one such method.

Questions play an important role in ensuring DOE success

Ingredients for a Good Experimental Strategy; Clearly Defined Objectives and Knowledge of the Process.

Plan your DOE before starting your experiment.

Piloting

Whenever you develop a solution, consider piloting it on a small scale first.

A pilot should clearly validate that the proposed solution will actually improve performance, or not.

User testing is always important.

Change Management and Communication Planning

THE MOST IMPORTANT ELEMENT OF DRIVING ACCEPTANCE IS TO PROVIDE EMPLOYEES WITH CONTEXT

Implementation Project Plan

As we improve, new problems will arrive that will require solving

Ensure that the organisation knows what is happening, when and how

Be clear whether it is a Big Bang, in Phases or ‘drip feeds’

If you are doing a phased roll-out, must manage parallel operations

You must ensure that the customer will not feel the impact of your internal transitions

Control

The use of Visual Management Panels facilitates the most effective Control – SELF CONTROL which is the basis for CONTINUOUS IMPROVEMENT

Documentation – it’s not just paperwork; it should start you thinking about training.

Monitoring – what do we need to watch and how will we do the watching

Response Plan – who will do what when things go wrong

Develop Standard Operating Procedures (SOP’s)

People need consistency to achieve top performance. Doing jobs the same way every time rather than wondering, “How does the boss want it done today?” improves productivity.

Well-written SOPs facilitate training.

People tend to be supportive of the things they help create. Involving employees in developing SOPs.

People can’t remember more than 10 or 12 steps, so they tend to have difficulty with long SOPs.

RACI Charts

Can be used to Analyse phase to understand key interfaces and hand- off points in the process and resolve overlaps and gaps.

Rarely does everyone on a team have the same understanding of who is responsible for what.

Skills Matrix

Consider, What if someone does not do a job for a while, do they keep the same skill level?

It is a simple tool to aid resource planning.

Statistical Process Control

Control Charts are not a solution tool – they are aids to help you identify problem areas that need solutions.

Not everyone will understand or trust the concept so they continue to make uninformed interventions.

Do not ignore Special Cause variation or make fundamental process changes in response to Special Causes of variation.

Update data on control charts regularly.

Corrective Action Plan

Do not focus only track back to the process output, Y, this will only provide indication of a performance shortfall after the event.

A response plan is not only senior management’s responsibility

It is a highly visible, living document that is used by all process stakeholders to understand performance on an on-going basis.

Failure Mode Effects Analysis (FMEA) to assess Process Risk

FMEA is different to traditional Project Risk Matrices which uses just Occurrence x Severity to assess specific project risks.

This is a simple and effective tool complicated by jargon.

Always work in a group on FMEA and use people who know the process.

Start with a process map, then brainstorm then prioritise.

Avoid time-consuming arguments about the scores. The trick is to use FMEA rating scales to remove subjectivity.

Do not forget the ACTIONS at the end. It is a pointless, academic exercise unless you make a difference to the risk profile – Error Proofing as a minimum.

Align Organisational

Consider, How well do existing rewards test against the characteristics of a good reward system?

A shared recognition of a need and logic for change is important for business results and critical to the organisation over time. The need for change has to be greater than the resistances to change.

Visual Management

Visual management panels are not JUST to look at!

Lessons Learned and Best Practice Transfer

By incrementally capturing lessons learned and turning that hindsight into best practices you will achieve far greater long-term project success.

Spend some time validating the lesson learned with the same individuals that were responsible for the improvement initiative in the first place.

Some of the most important lessons we learn come from failures.