Six Sigma Part 1

Hill Farmhouse 20 Chishill Road Heydon Royston Herts SG8 8PW Tel: 01763 837111 Fax: 01763 837223 www.wdpconsulting.com

wdp consulting

six sigma

Contents – Part 1
Introduction Six Sigma Model Project Management Process Mapping Measurement Statistical Thinking Basic Statistics Lean Operation Leading Teams

wdp consulting

six sigma

Introduction to Six Sigma
So What Exactly Is… Six Sigma? In 2000, Six Sigma was a phrase that was relatively new to the majority of businesses in Britain. Only 16%, according to an exit poll at the Best Factories conference in Birmingham, thought that it was a current manufacturing initiative (reported in Works Management July 2000). However, in the USA it was heralded as the next step on the road to quality improvement and business excellence. Since then Six Sigma has received a higher profile in the UK with many large multi-national companies adopting the methodology. However, Six Sigma is still seen as a methodology for large companies and has struggled to make headway into small medium sized companies. What’s it about? Six Sigma is about understanding the variation within products or processes. Reducing this variation produces a more consistent functionality in a product, or more consistent output from a process – leading to better processes, more reliable products, lower costs and ultimately happier customers. Where did it come from? The origin of Six Sigma goes back to Motorola’s quality initiatives of the eighties. In 1985, Bill Smith of Motorola presented a paper showing that if a product were assembled error free, the product rarely failed during early use by the customer. One of the members of Motorola at that time was Mikel Harry. He set up his own consultancy (Six Sigma Academy) and modified the principles. Harry says, ‘I was just the godfather. I put the math to it and dressed it up for school’. In the mid-nineties, Jack Welch proclaimed that great savings were being made at GE due to Six Sigma. From there, other major American companies joined the Six Sigma devotees. The ideas have since crossed the Atlantic. What is this Sigma and why Six? The term ‘sigma’ is used to describe the spread of the output from a process. As sigma increases the process spread reduces, product reliabilty improves, the need for testing and inspection diminishes, work in progress declines, costs go down, cycle time goes down and customer satisfaction goes up. The benefits can be understood if we consider a car which just fits into a garage. One small shift to either side and the car will be damaged. This depicts the situation where a processes output can just meet the required specification. (This would be ‘Three’ sigmas) However, if we halved the width of the car, getting it into the same
wdp consulting six sigma

in process terms.once the new process is up and running. standardise and monitor to ensuring that the gains are realised. wdp consulting six sigma .determine how to significantly change the process and reduce the defect levels. Define the goals of the existing system: . reduce the number of defects to as low as 3. Improve the system: . The aim of Six Sigma is to reduce the width of the car such that two car widths would fit into the garage. More advanced statistical tools such as Experimental Design may be used at this stage.garage becomes far easier. Analyse the system: . Control the new system: . This may involve the use of Statistical Process Control (SPC). map and measure the process. This would represent ‘Six’ sigmas.4 defects per million opportunities (DPMO). The simple quality tools and basic statistics are used at this stage.select characteristics critical to quality. Achieving this would. Understanding variation Variation around target Six Sigma Three Sigma Target Lower Spec Measured Characteristic Upper Spec 6σ Six Sigma Model The generic Six Sigma process follows the accronym DMAIC.choose a process or product that is critical to business success Measure the current system: .use benchmarking to identify potential areas for improvement.

one week training sessions. Motorola. One company has combined Six Sigma and Lean Manufacturing to create Lean Sigma! The American Society for Quality (ASQ) has produced its ‘Body of Knowledge for Black Belts’ although this is by no means accepted as standard. Nothing New under the Sun? Those who have lived through quality initiatives during the last twenty years will recognise many of the concepts. there is focus on projects with measured improvement. there is little difference except many of the successful quality techniques have been brought together. This is complimented by a project applying the techniques learnt.Six Sigma Training. These are taught to proposed project leaders (Black Belts) usually in four one week sessions with one session taking place each month. Some companies say a Black Belt qualifies having successfully completed two projects. In addition. wdp consulting six sigma . Typical Six Sigma training programmes rely strongly on statistical methods. something which has for many years been advocated by Juran. giving accredition to those who have simply completed their training course. has additional Yellow belts and one consultancy even has the position of Brown belt. Green Belts (project team members) would receive two. Six Sigma tools Subjects covered FMEA Design of Experiments Statistical Process Control QFD Measurement Systems Analysis Basic Statistics Reliability Engineering Regression Analysis Seven Quality Tools Mistake Proofing Taguchi Methods Benchmarking Hypothesis Testing T-Tests ANOVA Response Surfaces 6σ Six Sigma Standards Unfortunately there is no standard describing Six Sigma and each provider offers a different view and curriculum. itself. others are happy to take your money. So why is this different? In simple terms. some say three.

For many of the large companies small savings can become huge when applied worldwide. it is interesting to note that GE along with many others had past initiatives using tools similar to those of Six Sigma that gradually faded out. Team skills? Knowledge of teamworking and facilitation is essential. By using experimental design the factors affecting bag length were understood and the production was able to reduce the average bag length without creating risk for the customer. reduce errors. As with any successful initiative. However. A positive aspect is that. In fact. On a smaller scale these improvements may not reach billions of dollars but can still be significant. The savings on material amounted to about £0. 40% of executives’ incentives are tied to key Six Sigma achievements. The benefits of Six Sigma are often talked of in terms of savings to the company. the process was more stable and meant that the customer was even less likely to get an undersized bag. wdp consulting six sigma . In Motorola.005 per bag. Benefits of Six Sigma As with any improvement methodology for which the appropriate resources and commitment are applied the benefits can be enormous. a manufacturer of plastic packaging had tended to make the bags overlarge to avoid the customer receiving a bag that was too short for the application. improve products and processes and as a result save money. provision of appropriate resources and appropriate training but certainly not four weeks of statistical methodology.000 per year.So what is the secret? Advocates of Six Sigma would probably say that it is the correct application of the steps and tools. Not a huge saving. it should always be remembered that the original aim of Six Sigma was to improve customer satisfaction. Black Belts are released from their day-today work to be full time on projects. The key to Six Sigma is that the leadership at the highest level of an organisation drives it. Six Sigma requires top management commitment. except that the company produced 40 million bags per year giving a saving of £200. For example. In GE. Black Belts receive bonuses for project success. Although individuals are trained. However. in most cases. and basic project management skills will be needed for any projects that are developed. it is highly unlikely that they will be able to achieve their project aims without the help of others. More advanced tools can be picked up as and when they are needed.

objectives and plans. Six Sigma Success Advert for Satellite Broadcasts Oct 2001 ‘Dr Harry will show why successful installation of Six Sigma has little to do with statistics but.’ 6σ wdp consulting six sigma .Benefits of Six Sigma Produces significant bottom line results Requires and gets top management support Provides better products and services for customers Develops and empowers employees Works in all parts of the business Sustains gains over time Uses previous tools 6σ Six Sigma Pitfalls Six Sigma cannot be introduced as something that only impacts in one area eg. The key to achieving this is to ensure that Six Sigma is aligned and woven into the organisational strategy. establishing valid deployment plans and defining execution tactics…. manufacturing. Although the tools dominate the training it is not simply enough to have the tools without management commitment to changing the way the organisation operates. Six Sigma is in essence a platform for change. If it is to be successful it has to be applied throughout an organisation.. rather. with leadership setting performance goals.

Measure. and what their expectations are Define project boundaries the stop and start of the process • Define the process to be improved by mapping the process flow Measure the performance of the Core Business Process involved. • Define who customers are. Analyse. their Critical to Quality (CTQ) issues. • • Create innovate solutions using technology and discipline Develop and deploy implementation plan wdp consulting six sigma . Define Control Measure Improve Analyse Define.e. (Transfer) Define the Customer.Six Sigma Model The DMAIC model is intended to be cyclical i. and the Core Business Process involved. Control. Improve. what their requirements are for products and services. the improvement cycle never ends. • • • Identify gaps between current performance and goal performance Prioritise opportunities to improve Identify sources of variation Improve the target process by designing creative solutions. • • • Develop a data collection plan for the process Collect data from many sources to determine types of defects and metrics Compare to customer survey results to determine shortfall Analyse the data collected and process map to determine root causes of defects and opportunities for improvement.

Control the improvements to keep the process on the new course. transfer the learning to other relevant areas of the company. Some companies add Transfer after Control. That is. wdp consulting six sigma . • • Prevent reverting back to the "old way" Require the development. documentation and implementation of an ongoing monitoring plan Institutionalise the improvements through the modification of systems and structures.

S. Prepared a goal statement defining the results expected from the project. Developed and tested data collection forms to ensure the collection of consistent and complete data? wdp consulting six sigma . (note: results not solution)? 5. Reviewed the charter with the Champion / Sponsor and confirmed their support? 7. Been given a brief project rationale explaining the potential impact of the project on customers. Identified the types of measures to be collected and have a balance between effectiveness / efficiency and input / process / output? 3. team membership. with a measurable target. Confirmed that the project is a worthwhile improvement priority for the organisation and is supported by the business leaders? 2. Determined what needs to be understood about the process and where to go to get the answer? 2. 1. reasonable choice between gathering new data or taking advantage of existing data? 6. Developed clear. Tested the operational definitions with others to ensure their clarity and consistent interpretation? 5. Identified the primary customer and key requirements of the process being improved and created a SIPOC diagram of the areas of concern? 8. Composed and agreed a two to three sentence description of the situation as currently seen? 4. unambiguous operational definitions of the things to be measured? 4. Completed a Team charter including a list of constraints. 1.Six Sigma Checklist (adapted from P. Clarified the factors needed to facilitate the analysis of the data? 7. profits and its relationship to the company’s business strategies? 3. Prepared a detailed process map of the areas of the process where the initial measurement will be focussed? Measure Has the project team…. Pande “The Six Sigma Way”) Define Has the project team …. Made a clear. a preliminary plan and schedule and process scope? 6.

identified reasons for variation in the process and identified possible areas for improvement? 4. Conducted a value and cycle time analysis. Prepared and tested the measurement system. Ensured that the key areas of the process are understood so that a new process can be created to meet the needs of the customer efficiently and effectively? For Process Improvement 6. locating areas where time and resources are devoted to tasks that are not critical to the customer? 3. including proportion defective and yield? Analyse Has the project team … 1. Examined the process and identified potential bottlenecks. Investigated and understood whether they are dealing with Special Cause or Common Cause variation? 7. Analysed data about the process and its performance. If Common Cause. Created a list of innovative ideas for potential solutions? 2. including training of collectors and assessment of data collection stability/ 10. identified the significant factors influencing the variation in the process? Improve Has the Project team … 1. Identified clear criteria for a successful solution? 3. disconnects and redundancies that could contribute to poor performance? 2. Identified an appropriate sample size and sampling frequency to ensure a valid representation of the process under study? 9. Used data to prepare baseline process performance measures. as opposed to process improvement and confirmed the decision with the Champion / Sponsor? For Process Design / Redesign 5.8. identified and verified the root cause? 8. Made a final choice based on the above criteria? wdp consulting six sigma . Evaluated whether the project should focus on process design or redesign. If Special Cause.

schedule etc. Determined key charts / graphs for a process ‘Scorecard’ on this process? 4. 7. Verified that the solution will give the required improvements? 5. Forwarded other issues / opportunities. Selected on-going measures to monitor performance of the process and continued effectiveness of the solution? 3. Developed a plan for piloting and implementing the solution including a pilot strategy. Identified an ‘owner’ of the process who will take over responsibility for the improvement and for managing continuing operations? 6. Evaluated the results and confirmed that the results in the Goal statement can be achieved? 8. Prepared all essential documentation of the revised process. Agreed with the Champion / Sponsor and received the go-ahead? 6. to senior management? 8. Celebrated the work and efforts of the team? wdp consulting six sigma . Considered potential problems and unintended consequences of the solution and developed preventive and contingent actions to address them? Control Has the Project team … 1. Identified and implemented refinements to the solution based on lessons learnt from the Pilot? 9. that were identified during the project. Compiled results data confirming that our improvement has achieved the Goal defined in the Team charter? 2. Documented the team’s work and data collected during the project? 7. including key procedures and process maps? 5.4. Created and put in place a plan to completely implement the solution? 10. results assessment. action plan.

and things to think about. As a project manager. albeit rarely with balls!! What you should not do is expect the impossible. A good juggler will have an awareness of their own ability. and will be catching and throwing them up in the air again so they do not fall in the ground. What it does do is give some key pointers. The balls they have in the air relate to different things they are trying to keep in balance at any one time. that is down to an individual’s style and aptitude. the mood they are in when juggling.Project Management Introduction These notes have been compiled to give the reader an introduction into project management skills. how many jugglers have you seen spinning 20 plates at the same time as juggling 5 chainsaws? wdp consulting six sigma . It should be used as a reference point and as a guide. the people who may be juggling with them and those who may just be watching. you will have to do very similar interventions. The picture below shows a juggler. It cannot tell you exactly how to manage projects. Project management is often like juggling. At all times the juggler must be aware of where the balls are.

usually for being grossly overspent!! Projects can be both large and small. see diagram below: - Type 2 Project No Product developmentt Type 4 Project Research & Organisational change Greater chance of failure Methods Well Defined Type 1 Project Engineering Yes Yes Type 3 Project Applications software development Greater chance of success No Goals Well Defined Type one projects: goals and methods are well-defined . is limited in time and scope. Project management has been described as. and the methods of achieving them. wdp consulting six sigma . are well understood throughout the project. and managing people as a temporary team. This concept leads to the definition of four types of project. There are a variety of definitions of a project.typically large engineering projects. involves a variety of resources.g. which include: “A human activity that achieves a clear objective against a timescale. controlling. and change” “A human endeavour that creates change. and whether the methods of achieving them are well defined”.” Projects can be judged against two parameters: “whether the goals are well defined.Definition of Project Management Projects come in different forms. a fixed timescale. Different Types of Project “Many traditional definitions of projects assume that the objectives of a project. Type two projects: goals are well defined but the methods of achieving them are not e. no practice or rehearsal. and is unique” All projects have to be managed. “the process of planning. product development projects. Many people think of a project as being the Channel Tunnel or British Library or other large engineering projects that hit the news headlines. This is no different in Six Sigma. a team of people.one clear objective. Projects nearly always have the following characteristics . has mixed goals and objectives.

of projects. There are four key stages in the life of a project. Above all else customer demands may create the need for projects. Some of these demands will be focused in an organisation’s annual strategy. The conventional view of projects as complex. Type four projects: neither the goals nor methods are well-defined . This will inevitably create a project to implement the technology. For example.g. External influences can also create the need for projects. typically organisational development projects. and many large projects will often fall into this category. They may be part of the logical strategy of an organisation and be created on an annual basis. thereby fixing the goals and methods.Type three projects: goals are not well defined but the methods are . particularly who the stakeholders are. but well defined. For example. Some people argue that the best way of achieving these projects is to move the project into the type three or two category (preferably type two).g. There are also other sources that may drive projects. but others will be much more immediate. if not the majority. and create projects to deal with the external changes. then customers will place demands upon the organisation to make changes in the way the service is delivered. PROJECT LIFE SPAN BIRTH Proposal and initiation MATURITY Implementation and control W DP GROWTH Design and approval DEATH Finalisation & control & evaluation wdp consulting six sigma . and then into the type one category as quickly as possible. in order to build appropriate relationships and satisfy competing demands. if service provision falls down to an extent that a great deal of bad publicity is created. one of the first things the project manager must do is to identify the external environment. Whatever the type of project. The Development of Projects Projects can come from a variety of sources. software development projects where the user requirements are hard to specify. sequences of activities to deliver clearly defined goals and objectives may be invalid for a large number. an operation manager may see the scope to introduce technology to make the operational process much more efficient. If there are plans to change legislation then organisations often have to react to this. programmed into the ongoing efficient running of an organisation.e.e. Milestones become key decision points and configuration management is suggested as a way of managing this.

This may be the end of the project as it quickly becomes apparent that it is a not feasible. One of the worst crimes of the project manager is to ‘press on regardless’. Even at this stage of the project. monitored and controlled. only to suffer drastic consequences. Project Leader: who is leading the efforts? Champion / Sponsor: who is responsible for “creating the environment” for the team leader and team to undertake the project. Areas to be considered: Project Title: this should give others a quick view of the project and use key words or phrases. Charter / Scope Writing a scope document that incorporates the necessary information will mean that there will be a formal agreement between Project Manager and the Project Champion as to what should be delivered.The first is the Birth stage. hopefully completed. Defining The Project As already described. It should be a document that is constantly referred to and ensures that all work is focused and that only the right amount of work is undertaken. thereby ensuring an effectively managed project. However.Scope. progress may and should be aborted if things change so radically as to make the project no longer feasible. This will be where the seeds for the project are sown. Stakeholders. if the project passes successfully through the first stage then it moves into the Growth stage. The Death stage of the project marks the end of the project. and the outline proposal documentation is drawn up. This document will become ever increasingly useful to the Project Manager as the project progresses. During this phase the project design is made clear. and will be the time that project evaluation and learning takes place. There are four key elements to this: . Within Six Sigma this known as the Project Charter. where despite iceberg warning the ship continued on at high speed. This is followed by the Maturity stage of the project wherein the plans will be implemented. Project Start date: the date the project formally started. Analogies can be drawn with the Titanic. the definition of the project will be paramount to a successful completion. the scope of the project agreed and people understand the likely impact on resources. Self and Team. wdp consulting six sigma .

the process of scope management begins and this is covered later. Stakeholders can include a variety of people.Project End date: the Champion will probably set the anticipated project end date. Process: which process is to be improved? Why does it need improving? How will it affect the customer? I.e. This will give an idea of the potential savings and size of the project. The team needs clear boundaries to help it focus on the extent of the project. Stakeholders The management of a project’s “stakeholders” means that the project is explicitly described in terms of the individuals and institutions that share a stake or an interest in the project. Project objectives: what is the project expected to achieve? Are the objectives SMART? OBJECTIVES Specific Measurable Achievable Relevant Time based W DP Process Measurements: what are the measures that will determine the success of the project? Team members: who should be part of the team? Subject matter experts? Project Milestones: what are the major milestones within the projects? Once the scope is defined. what is the need for this improvement? Project boundaries: the team cannot solve everything. often outside the wdp consulting six sigma . Cost of Poor Quality: quantifying the total cost of quality can be difficult but there may be information on such things as scrap produced. The time allowed for the project should give enough time to complete the project given its complexity and current business conditions. excess hours spent etc.

and what does exist is often based on simple forms identifying the stakeholders and their level of power and interest. Only in this way will all external influences be considered. it is necessary to look at the project life cycle. Self Only by knowing him or herself / herself will the Project Manager ensure that they make an effective contribution to the project. It is only by identifying what is important to each of the stakeholders in terms of both hard and soft expectations. and dovetail the two accordingly since stakeholders may appear during the life of the project. By being realistic about this. The following questions will be useful in assessing this: What experiences have you had? What are you good at? wdp consulting six sigma . that the project manager can develop options for managing any interference or conflicting expectations and for managing the project to ensure the stakeholders are satisfied at its conclusion. There are few prescribed tools and techniques for the management of stakeholders. the Project Manager will be able to see their own strengths and weaknesses.control of the project manager and therefore presenting serious management problems. This information can be used to map how this might affect the project and then the necessary action can be taken to ensure the project does not falter due to a weakness in a key skill area such as planning. as demonstrated in the model below: Level of interest Low High Low Minimal effort Keep satisfied Keep informed Key players Power High It is critical that all stakeholders are identified and none overlooked. While deciding the approach to take regarding stakeholder management. and possibly using some assessment tools to aid the process.

Group pressures can have a major influence on the behaviour of individual members within that team and their performance to the team’s overall aims. and so the project manager should seek to overcome a weakness by using the skills of others where necessary. The development of a weakness may well take some time. the people within it influence each other in many different ways. Once in a team. Both may be different to that of the organisation from which it originates. Strengths are often easier to build on and identify whereas weaknesses are often hard to admit to. Admitting to the weakness does not have to be a part of this.What are you not so good at? What do you want to develop? How will you measure this? The strengths and weaknesses identified should then be matched against expected role and performance. although doing so may well allow the designated person to act as a mentor for the development of the skill in question. The team may even develop its own hierarchy and its own leaders. Teams Teams are an essential feature of the work pattern of almost any organisation. The activities of the team can also be closely associated with the process and style of leadership adopted by the leader / facilitator of the team. wdp consulting six sigma . accept and then either develop or resolve.

the easier it is to promote cohesiveness.There are two different types of group: formal and informal. However. The performance of a team is essentially its effectiveness. This is influenced by: (1) (2) the givens . wdp consulting six sigma . problems arise with communication and co-ordination. not necessarily related to the task being undertaken. with the occasionally ‘whole project’ team briefings to manage communication. Formal groups are created to achieve specific organisational objectives and concerned with the co-ordination of work activities. The optimum performance of team size is probably five to seven people. the task. they can also cause difficulties and conflict may arise when members are in competition with each other. For larger projects it therefore makes sense to have teams and subteams for different work packages. If a team becomes too large it may split in two sub-teams with friction developing between them. As team size increases. where people are so in tune with what the others are thinking they do not always see the broader picture and may overlook something incredibly obvious. It is very unusual for an informal group to be exactly the same as a formal group. processes and procedures leading to (3) the outcomes .leadership / facilitation style. Compatibility of members of a team is also important. People are brought together on the basis of defined roles and the structure of the organisation. Informal groups are based more on personal relationships and agreement of group members than on defined role relationships. the environment the intervening factors . they can suffer the phenomenon of ‘group think’. The nature of the task to be undertaken is an important feature of the formal group with goals being identified by management and certain ground rules.the team. The more homogenous the team in terms of background. Understanding the informal group network can be critical in determining how to facilitate formal teams. The Givens The size of the team is an important factor in its performance. and can under perform. Teams with more than seven or eight permanent members start to split into sub-groups. attitudes and values. The membership and roles of informal groups can cut across formal structure. motivation. If teams are too cohesive. These groups tend to satisfy psychological and social needs. interests. Differences in personality or skills in the team may actually complement each other and help the overall performance. relationships and perhaps norms of behaviour established.productivity and member satisfaction. These groups tend to devise ways of attempting to satisfy member’s affiliation and other social motivations that are perhaps lacking in the work situation.

team performance is likely to be hindered. However. wdp consulting six sigma . if the task requires ambiguity. either by talking to individuals separately. It is important for the team leader / facilitator to reveal hidden agendas. It is important that the task is clear and unambiguous. hidden agendas can be overcome by careful allocation of task The type of task will naturally affect the kind of team that is formed. Once recognised. the more important it is for the team to work well together and the need for consensus on a focal person or leader. This enables greater demands to be placed upon the team. Belbin has done some study into this and defines a variety of roles within a team. the more important the task is to the individual within the team. but not too long as procrastination may occur. therefore having implications for the leadership of that team. For example. but not so good at communication. they will be more suited to the co-ordination of the project plan. or preferably encouraging members to admit their own issues to the team. It has been found that the more complex the task is. For as long as hidden agendas exist. This can result in work of a lower quality than if the timescales were broadened.The individual roles in a group can therefore be important. the Plant will be introverted but have many ideas. making a particular alliance. However. When setting up a team it is useful to examine individual strengths and weaknesses and recognise the most beneficial role each team member can play. The benefits of task allocation are that it allows a structured approach. and will prevent trust from building. Hidden agendas can give rise to conflict. Hidden agendas may be protecting the interest of one sub-group. the Shaper will be the task leader and is needed to spur action. If the timescales are very tight. if someone is particularly good at planning. This will assist the team through the stages of team development. Good performance therefore requires that a leader / facilitator should be picked because of ability to direct and influence rather than status. The team that is formed to formulate ideas about the detail of the task needs a more supportive management style and may not initially have a structured approach to its work. the more committed they are to achieving it. impressing the boss. then the team may not develop to deal with individual needs. performance in terms of quantity may be a lot higher with team members motivating each other. or even covering up past errors. Another important factor affecting team performance is the ‘salience of the task’. rather than primary liaison with stakeholders. The timescale of the task also reflects how the team will work together. For example. the Finisher will check details and deadlines. then this can lead to the formation of different roles within the team and also increase stress. The timescale set should therefore be realistic so that good performance can be achieved.

Performing the group structures itself or accepts a structure that fits most appropriately a common task. Storming conflict begins to emerge between individuals or sub-groups. A model that is frequently used splits the development into four stages:Forming this is the initial stage of development. and should not be overlooked even if the team already know each other. have a growth cycle. and what behaviour is or is not appropriate. testing to discover the nature of the situation.Team Development When examining the components of team performance. This results in progress and achievement. like individuals. cohesion and unity begin to be experienced by participants. and performance expectations must be altered accordingly. norms emerge for handling conflict. Teambuilding exercises may help with this. Discussion causes differences of opinion to emerge. The group begins to apply more of its energy to being effective in the area of its common task. Roles are established and accepted in relation to the task. Communication of views and feelings begins to develop. indeed its purpose for existing is challenged. Participants try to discover the nature of the task. Performance will improve if teams can ‘get through’ the first stages as quickly as possible. Participants can be anxious. and team-building exercises should not just be saved wdp consulting six sigma . and the ground rules. Teams. mutual support develops. the authority of the leader. It is also important to recognise that teams need to be maintained. The team will function differently at various stages of development. trying to discover what help is available from the leader or facilitator. BUILDING THE TEAM 4 stages of team development forming storming norming performing motivation is key Task needs W DP Balancing needs Group maintenance needs Individual needs Norming the group begins to harmonise. it is also important to consider team development. the make up of the group. Plans are made and work standards laid down.

they should focus on being heard and having control. but it can be used to help someone recognise his or her behaviour and its effect on the team request that the person change his or her behaviour to be more consistent with the ground rules. Here are some things you can do to encourage team responsibility from the start. Example: Joe. and explain their use. or suggest the desired behaviour. Team members take turns answering. Each member is asked to respond to a few questions – such as. When Sarah offered her idea. your focus should be on group dynamics and positive confrontation to help team members find workable ways to address their differences. That not only establishes immediate ownership for the success of the team. and correlate it to the ground rules. wdp consulting six sigma .’ invite comment. As facilitator. Sarah. Check progress Team progress or process checks are an easy way to encourage team ownership of results and how the team works together to achieve them. Progress checks are led by the facilitator and are usually the last item on a meeting agenda. This is the stage at which many teams (and facilitators) either make it or break it. They look to the facilitator for guidance and assistance. Would you be willing to hear how Sarah’s idea might work? Clarify roles Working with the team. That can be the beginning of a long-term dependency. team member. The Forming Stage During the initial stage of development. but also for enforcing them. This step isn’t always necessary.for the beginning of a team’s task. so they don’t express their negative feelings – they don’t want to rock the boat. but undertaken at various intervals. Provide a mini lesson on how to intervene regarding breaking the rules: • • • make a statement or ask a question that describes the inappropriate behaviour. and team leader. Help the team develop ground rules. The Storming Stage Once team members are oriented. Example: What did you intend by that? Or. you said ‘You need to get out of the office more. conflict is often evident. I felt that comment was a cheap shot. but also individual responsibility for fulfilling roles and expectations. one of our ground rules if no cheap shots. throughout the life of the project. At this stage. Encourage team members to assume responsibility not only for adhering to the ground rules. What did we do well? What did we not do well? What do we need to do differently the next time? The facilitator asks the questions one at a time. the primary concern of new team members is to be included. as appropriate. define and spell out in specific terms (preferably written) the roles of facilitator. They typically wonder: ‘How do I fit here?’ ‘what’s expected of me?’ They may think that if they speak out they’re misbehaving. and then the facilitator offers his or her view.

What’s going on? Janet. in the Norming stage. you just said to Eileen ‘There’s no way I’ll go for that’ and suggested another alternative. The Norming Stage By now. There are two reasons why some don’t let go and the dependency deepens. The Performing Stage This stage leads to the final handoff. Your key focus is to encourage the team to become self-facilitating. that’s the third time you rolled your eyes as John has responded to that? (Patti’s response: ‘Yes. Example: Paul. and taking steps to balance everyone’s participation. Ron. Your primary focus as facilitator is to uncover unspoken issues and promote the full exploration of ideas. would you be willing to paraphrase what Eileen proposed and state your concerns? Use graduated interventions. You should make such statements as: • • • Why do you think you were so effective today? Or. a handoff means transferring most or all of a facilitator’s typical duties to the team. But in general. So when an issue crops up that the team could handle on its own. Now. Example: Patti. Make observations aloud to get team members to see the effect of their actions or behaviours on others personally and on the team and its tasks. you wouldn't be so blind. A key role of the facilitator during this stage is to play back the specifics of what’s happening. wdp consulting six sigma . you drop your eyes and start to doodle. depends on the type of team. Before you talk about how your idea will work. he said.Get the team to own responsibility for recognising and addressing difficulties. Were you aware you were doing that? What are your concerns? This stage is also a great time to provide idea-generating tools that stretch people’s thinking or encourage differences of opinion. One it feels good to be the expert or advisor. they should focus their energies on being open and encouraging others. Devil’s Advocate is an exercise in which one team member is appointed the devil’s advocate and has to argue why certain ideas or proposals won’t work. They have become comfortable giving and receiving feedback. taking risks. however. Why do you suppose you had such difficulty? I sense that you’re not comfortable with what has been proposed. each time the group mentions the benchmarking visit.’ How do you think these exchanges between the two of you are affecting the team?). The process of letting go. Two. It’s here that the ties that bind are cut. although simple. the more second nature his or her responses are. The definition of final handoff. is difficult for many facilitators. ‘If you cared about anything other than your promotion. team members have worked their way through the struggles in the storming stage. You haven’t said anything in the past 10 minutes. the more skilled a facilitator is. the facilitator steps in automatically.

and we should probably see how our customers would feel about it. Example: How will that help you overcome the difficulties you were having? (Team response: ‘It will force us to look at one alternative at a time and thoroughly explore it. a superlative one inspires and motivates. ask questions that encourage them to test the idea. What process? What do you think you need to do? 3. Ironically.The process for empowering a team to become self-facilitating is Socratic. if needed. 4. Knowing that a team depends on you can be satisfying.’) 5. A mediocre facilitator tells a team what to do. This means paraphrasing what was said.’) 2. Wait until this step to offer your suggestions. Ask outcome questions. This step involves helping the team see the link between the proposed approach and the desired results. a good one demonstrates and encourages. Is there any value in checking those alternatives with some others outside the group? 6. or expertise.’) Will that provide you with all of the information you need to make an informed decision? (Team response: ‘Probably not. Check Members’ understanding. Rather than correct team members or elaborate on their ideas. Rather than give the answer. 1. Example: You said you could probably look at the pros and cons of each idea. Ask ‘What will you do’. Team independence is the best measure of a facilitator’s success. But the absolute best facilitators are able to step aside and allow a team to take charge. Example: I think those are both good approaches. Example: What’s happening right now? (Team response: ‘We’re foundering. ask a question that will help you determine whether team has the skills and knowledge to address what’s happening. It’s here that you need to draw on your questioning abilities. ask leading questions that will determine its direction. If those are OK the team then puts its plan to work. Offer relevant observations and experience. This final step gives you an opportunity to check the end result and action plan. the skills for putting yourself in the background are greater than for staying at the forefront. wdp consulting six sigma . If the team can’t resolve barriers to success using steps 1 through 4. one at a time. assistance. Check your understanding. These six steps will help you hone them. Ask a question that causes the team to recognise what’s happening. We need some data from X. By asking the right questions. Example: You say some kind of process will help. We probably need to use some kind of process to hash this out. you can get a team to recognise what’s happening (observe its own process) and determine what to do.

there is a core set of techniques that are commonly used. which will often mean that different levels of a plan will exist. Before looking at the various ways in which plans can be developed it is worth understanding the sequence of planning. It is also worth noting that plans should be developed with the project team. evolved and established to co-ordinate complex activities . turn it into a series of molehills. and other relevant stakeholders. It simply follows the principle that when faced with a mountain. For example. partly as a result of nations’ competition for supremacy in weapons and defence systems. The plan should be targeted at its audience. it allows work to be delegated in coherent and often independent packages. For example. Milestone plan this is a tool that allows the project manager to show how the intermediate deliverables build towards the final objectives of the project. and it allows risk to be contained more easily. the Egyptian pyramids. senior people in the organisation will not be interested in the detail of the tasks to be undertaken. Although variations exist. It is something that provides better control of work definition. However it has not been until the later part of this century that many of the techniques have been developed. but they will be interested in the overall picture and some of the key milestones to be achieved. It wdp consulting six sigma . it allows work to be defined at an appropriate level for estimating and control. The methods and tools for planning projects can be as follows: Work breakdown structure (WBS) this involves the division of an activity into sub activities and sub-sub activities in order to control the project effectively.Project Planning & Control SEQUENCE OF PLANNING Consider the ‘big picture’ Break the project into sub projects Determine the relevant milestones Break each sub project down Clarify resources W DP Project planning is a form of management.some man-made ones still survive today.

the reason for this is that you could arrive at work by bus should your car breakdown. if not. Responsibility charts can be described as a matrix with tasks shown as rows and organisational resources as columns. as is often the case in open projects. Symbols are placed in the body of the matrix to represent the involvement of each resource type in each work element.therefore allows different members of a project team to see the interfaces between their activity and others. This method can be invaluable in managing and allocating known tasks. but the key issue is ‘what will have been achieved’. the milestone plan is likely to be much more broadly focused. but not ‘when I have driven to work’. but what is to be achieved. Once milestones have been defined a result path can be drawn to show the interdependencies between tasks. An example of a milestone would be ‘arrival at work’. Where these are not always clear. MILESTONE PLAN WBS1 WBS2 WBS3 tasks activities (sub WBS3) W DP interdependencies milestone result path Milestones are useful to show the logical sequence of events. Defining milestones can be a tricky task. as we see illustrated here. but still providing a useful tool for the project manager. They do not show how something is to be done. If tasks are predictable they can be developed in advance. highlighting dependencies. RESPONSIBILITY CHART Companies / Departments / Functions Resources / People \\\\ W DP Milestones Activity / Task Roles / Responsibilities wdp consulting six sigma . they can be used over shorter time frames. This can get quite complicated.

Estimating costs at the start of a project clarifies the total budget required and predicts cash flow requirements. but something!!! The use of time charts as named above allows people to see what is going to happen and when. Managing to achieve deadlines is most critical. is unknown then it can be difficult to draw. the late start and finish and the float of elements in a project. The critical path shows the overall duration of the job. '98 Oct 25. ' MTWT F S SMTWT F S SMTWT F S SMTWT Tue 10/13/98 Wed 10/14/98 2 days Thu 10/15/98 Thu 10/15/98 Thu 10/15/98 Fri 10/16/98 Fri 10/16/98 Fri 10/16/98 3 days Thu 10/15/98 0 days Thu 10/15/98 0 days Tue 10/20/98 8 days Fri 10/16/98 4 days 10/15 Task E Task E1 Task E2 Task E3 Task F Task G Task H Mon 10/19/98 Mon 10/19/98 1 day Tue 10/20/98 Tue 10/20/98 3 days Mon 10/26/98 Mon 10/26/98 1 day Tue 10/27/98 Tue 10/27/98 2 days Wed 10/28/98 Wed 10/28/98 1 day If plans also calculate the early start and finish. or bar charts also show time.Labour. Management. See example below: - GANTT CHART ID 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Task Name Task A Task B Task C Task C1 Task C2 Task C3 Task C4 Task D Half Way Party !! Duration 1 day 2 days 8 days 2 days 2 days 2 days 2 days 1 day 0 days 3 days 1 day 1 day 1 day 1 day 1 day 1 day Start Wed 10/7/98 Thu 10/8/98 Wed 10/7/98 Wed 10/7/98 Fri 10/9/98 Finish Work Wed 10/7/98 1 day Fri 10/9/98 1 day Fri 10/16/98 9 days Thu 10/8/98 3 days Mon 10/12/98 1 day Oct 4. and a classic tool for managing time is the critical path method (also known as programme evaluation and review technique). scope and organisation. Subcontractors. They depict the projects schedule by showing the duration of an activity. Gantt charts. start dates. It is the longest path of the project and therefore the shortest time in which the job can be done. Some projects will be very capital wdp consulting six sigma . but needs to be linked to the WBS. even for open projects where the stakeholders will often want to see something by a certain date . Cost can be measured through a cost breakdown structure or project financial plan. Estimating project costs can be achieved through careful planning. its planned start and finish dates and the logic of events. Overheads and Administration.they may not be sure what or how. Materials. ‘what-if’ analysis can be more readily conducted. The WBS provides the means to so this. Plant and Equipment. assessing historic projects and group brainstorming. Controlling and assessing variance enables plans to be made to manage some of the inevitable overruns!! It is a simple technique really. managing cost. Clearly if activities. or preferably a combination of all three. The costs that occur can be split into a variety of categories . Another traditional technique is the Cost specification methodology that focuses on three objectives.Time is often seen as a critical element of projects. Fees and Taxation etc. '98 Oct 11. '98 Oct 18. Any delay in the critical path activities will cause the project to overrun. logic etc. and to control costs you must compare the actual expenditure not to the schedule of expenditure. but to some measure of the value of the value of work done. Some projects are time critical others are not.

Techniques are employed to produce the best product and service possible . people’s time (it is often just fitted in!!).For example. tested. and any variances are eliminated. recorded and analysed. OD projects by their very nature are less capital intensive. but a management standard system. As the facility is erected and commissioned it is checked against the specification to ensure it is of the required standards. but it is a means to an end and should not be seen as anything more. periodic reviews and reasonable change control. hidden .intensive whilst others will not be. but we always find time to do it twice!’ W DP Quality can be managed through a variety of means. Quality control is the process of diagnosis and cure.through constant innovation. At the start of the project the manager should draw up a quality plan to define how quality will be achieved. The costs will frequently arise in the expertise bought in to assist with certain interventions. such as ISO9000 are not a product standard. either as part of the organisations approach to projects generally or specifically produced for major projects. appropriately qualified people. Three tools that can be adapted from regular initiatives include the Cause and Effect diagram. It provides a framework of procedures with which to achieve quality. Quality assurance systems. The activities by which this is done must be planned. It is geared to the achievement of quality of production. In order to assure the quality of the project there needs to be a clear specification with defined standards. not the production of quality. and are often. how the company’s procedure will work on this project. Pareto's 80/20 rule and ‘Just-in-Time’ methodologies. wdp consulting six sigma . and how the manager intends to assure and control quality. and the costs can often be. A project handbook may well be used. Complete cost predictability is therefore impossible until the tasks are known! QUALITY Give the customers what they want! Achieve by: Carefully developed & managed processes An inbuilt attitude of mind ‘We never have time to do anything. relating to how quality is provided.

The systems are used for planning and controlling projects. internal .technical. Microsoft Project is an example of a computer software package that can help in the management of a project.pass the buck contingency high risk W DP Simple low risk Some innovation New high Risk management – There are probably five sources of risk.identify and replan deflection . It should not however be assumed that by purchasing such a system. and group brainstorming. Project Management Information Systems a PMIS is used for collecting data from across various functions. assumption analysis. external-predictable. providing timely control information. decision drivers’ analysis. A typical system has two key purposes. For example. The systems are also used for recording historical data for estimating future projects. external-unpredictable. These can then be reduced by avoidance. internal .skills and teamwork uncertainty high Overcome by: avoidance . and analysing and presenting those data in a form suitable for all the parties involved in a project. throughout the project management lifecycle. and legal. a project will automatically be well managed and planned. deflection or contingency planning.RISK Outcome Process Simple Medium High complexity Affected by: technology . THE PLANNING SYSTEM data • • • • • scope organisation time cost quality structured information THE CONTROL SYSTEM • purpose • resources • management management decision and direction wdp consulting six sigma . The techniques that can be used to identify these risks include .non-technical.expert judgement. Whilst risks exist in every project. to act as a planning system and also as a control system . Risk can be identified by using the simple table shown.new and old time people . the above categories may not be appropriate to OD projects. internal-technical risks may not be relevant to a quality improvement programme.as illustrated below. plan decomposition.

Summary A key part of Six Sigma is the running of projects to achieve the improvements within a process or product. it is necessary to use the knowledge developed from running projects to help ensure that the Six Sigma project is successful. and relies upon much of the data set-up in the ‘scoping tools and techniques’ (as illustrated in the previous diagram). Thus. and therefore the same arguments apply when considering their usefulness with open projects. The key parts of the project are : Defining the Charter Selecting the most appropriate team Planning and monitoring the project steps using tools described Communication with all stakeholders Process Mapping wdp consulting six sigma .This sort of system builds.

You will discover that individuals will have a different understanding of what the map looks like and how the process works. Customers and Suppliers may be either internal or external to the organisation. CONTROLS SUPPLIERS INPUTS PROCESS OUTPUT CUSTOMERS RESOURCES Definitions: wdp consulting six sigma . In its simplest format it can be seen as: INPUT PROCESS OUTPUT The process transforms the inputs into the outputs.A Process Map is a graphical representation of a process. Mapping can be applied to anything from the movement of an invoice to the flow of materials in manufacturing or the steps involved in making a sale. showing the sequence of tasks or events using a number of symbols. SUPPLIERS INPUTS PROCESS OUTPUT CUSTOMERS This diagram is sometimes referred to as SIPOC. a group of logically related tasks. What is a Process? A process is a set of steps/tasks/activities that have a beginning and an end. decisions or activities which together produce a result or output. This simple diagram can be expanded to include Suppliers and Customers. The objective of creating a process map is to help understand the process. Hence maps should be created by a team rather than by an individual. highlight where problems exist and ultimately improve the process. A further development of the diagram brings in reference to Controls and Resources. (see appendix ). Another definition could be. A good map is the foundation for continuous improvement.

E.P. equipment or material not transformed by the process Supplier – individual.I.C. department or organisation that supplies the inputs Customer – individual. have they considered the requirements of all the customers listed? Suppliers Inputs Process See Below Outputs Customers Requirements Step 1: Step 2: Step 3: Step 4: Step 5: wdp consulting six sigma .g.Process – any activity that transforms inputs into outputs Inputs – material and/or data transformed by the process to create the output Outputs – the results of the transformation of the inputs Controls – regulators or influences on the process Resources – people. department or organisation that receives the outputs In looking at a process for the first time it can be useful to list the elements under each of the S. This helps remind the team of aspects of the process that need to be considered.O.

High Level Process (Core Process) E.g.g.SIPOC Example Suppliers Manufacturer Inputs Copier Process Outputs Copies Customer s You O ffice Supply Company Paper File Toner Making a photocopy O thers Yourself Original Pow er Company Electricity Process Steps Remove Remove originals originals and copies and copies Put original Put original on glass on glass Close Lid Close Lid Adjust Adjust Settings Settings Press Press START START Levels of a Process Your position within an organisation will affect how you see a process. Launch Product. wdp consulting six sigma . Key Process (Major Process) E.g. introduce new products to the market Identify Market Need Develop Product Design Develop Production Capability Launch New Product Within this there are Key Processes e. A process can be shown at a number of levels. Launch Product Prepare Sales Literature Train Sales Team Pilot Sales of New Product Sell New Product Each of theses processes can be further broken down into Activities or Tasks.

g. It is important to agree from the start at which level the process is being considered. However. that it was acknowledged. Example A chemical process was showing high and at times irregular variability. Design brochure Write Text Agree Layout Prepare Artwork Produce Draft Brochure If needed even each of these could be broken down further. Whatever level is being considered. the Supervisors denied they did this and it was only under pressure from Operators saying they did. For most improvement situations. It was only during conversations with operators that it became known that at the start of a shift. For example it may be decided that a radical overhaul of the method of identifying a market need is required. Honesty is a key requirement before any real improvement can be made. wdp consulting six sigma . Prepare Sales literature Design Brochure Agree Design Print Brochure Distribute Brochure Activity / Task E. Once the Supervisors stopped changing the process that they were able to move forward and improve the process which ultimately led them to running a line for several days without any breaks or changes. the new Supervisor got out his little black book and changed the process settings. this may not always be the case. the more detailed activity / task level will be used. When the Supervisors were consulted over this they had no answer even though some of the changes appeared to be shift related. it is important that an accurate map of what currently happens is created.g.Sub-Process E. Initially. It is very tempting for teams to produce a map of what should happen according to procedures or what they would like to happen. a situation unheard of before. once the previous Supervisor had left.

e. then there needs to be an understanding of which processes are linked to which key business objectives. suppliers and other stakeholders. Amongst the key business objectives should be those that directly relate to serving the external customer. Process Mapping Steps Step 1: Decide on the process to be mapped Step 2 Decide where the process begins and ends (boundaries) Step 3 List the outputs and identify the customers for these outputs Step 4 Determine the requirements for these customers Step 5 List the inputs and their suppliers Step 6 List the main activities and decision points in the process Step 7 Arrange these activities in their proper order.Link to Key Business Objectives There are many processes within an organisation. using arrows to show the direction of flow Step 8 Agree this represents the current ‘as is’ process Customers and Suppliers (Steps 3 – 5) In drawing the Process map it is important to recognise those who may have some interest in this process. Make sure that the objectives are not purely internal. customers. i.e. the process having the greatest impact on the business objectives. Think of a process that you are involved in: Who are your main customers? wdp consulting six sigma . thus if significant improvements are to be made. A decision making process (see Tools) can then be used to identify the most critical process i.

Map the Process (Steps 6 – 7) In order to display the process than different methods may be used. Systems Flowcharting Operations Flowchart (See appendix for description) 10 20 30 40 wdp consulting 50 six sigma .Do you know what their requirements are? Are your customers aware of what you can or cannot do for them? Who are your main suppliers? Do you know what their capabilities are? Are any of your customers also your suppliers? Do any other groups have an interest in your process? Too often organisations produce Process Maps without any reference to customer needs and expectations.

Example: Washing dishes after a meal Clear dishes from table yes Food on dishes? No yes Dispo se of food? No Store food Put food in bin Stack dishes in dishwasher Add detergent Set dishwasher controls and turn on wdp consulting six sigma .

However. A step that contributes to customer satisfaction adds value. Each step will take time. material and add cost to the process. it is important that each step adds value to the process. Measurement The measures chosen should relate to customer requirements or necessary controls required to ensure that the process works correctly. If the answer is no. cost or labour Supplier’s effectiveness: how good is the input material or information? Process controls: how stable is the process? In deciding on the measures to be used. the key measure was on-time delivery. It was not until part way through the project that it was discovered that half of the team thought that on-time meant as measured against the factory scheduled delivery date whilst the remainder thought that it was measured against the customer’s required delivery date. If this step were eliminated then the customer would notice. This is the total amount of time to complete the process from boundary to boundary. then improvements must be made.Improving the Process Map Simply getting a team to draw a Process Map can lead to the identification of areas for improvement. Typical areas of measurement might be: Effectiveness: output measures of your customer’s requirements Efficiency: measures of the amount of time. take care to ensure that their definitions are agreed. sometimes there were weeks between them. Thus. Value Analysis There may be many steps within a process. For example in a delivery performance project. For example discovering output that goes nowhere or realising some things could be done earlier in the process. In many cases it may not be possible to answer this question immediately and it is necessary to collect data from the process in order to identify areas for improvement. The definition of whether a step adds value or not needs to meet three criteria (Eckes): wdp consulting six sigma . on studying the final Process Map it is necessary to ask whether or not this represents the ideal process flow. A typical measure of a process is that of Cycle Time.

If the step does not satisfy a customers’ needs then look to eliminate it. 5.1. reject. 6. The step must physically change or transform the product or service 3. 2. reclaim. The customer must be willing to pay for that step in the process 2. try to brainstorm ways of reducing the time taken without losing the value. wdp consulting six sigma . 3. At each step consider key customers’ needs. For each step ask: Does this step add value in our customers’ eyes? What would happen to the customer if this step were eliminated? Are others in the organisation affected by this step? If a value-adding step adds considerably to the cycle time.g. rewrite. Overproducing Waiting Transporting Inappropriate processing Unnecessary inventory Unnecessary motion Defects Having identified non-value adding steps it is useful to calculate the impact they are having on the total cycle time of the process. 7. review. retest. The step must be done right the first time It is often easy to identify non-value added work as the process step begins with the prefix “re” e. Given that most processes have been in operation for many years then it is quite possible for steps and procedures to have been added for someone or purpose many years ago but are no longer of value. Eckes identifies six major categories of non-value adding work: Internal failures – steps related to correcting errors/ rejects in the process External failures – steps relating to failures at the customers Control / Inspection – does this add value? Why can’t the process get it right first time? Delays – often seen as waiting for the next stage of the process Preparation/set up – the time this activity takes up has led to the demand for long runs Movement – excess movement is often created due to poor factory layout Compare this list to the Seven Wastes (see Lean Operations) identified by Taiichi Ohno: 1. It will also enable priority to be given to the step that has the largest detriment on the process. 4. recall.: rework.

such points should be kept to a minimum. wdp consulting six sigma . However.Inspection / Control Points A point of contention is the identification of inspection or control points. such that there is often little knowledge of exactly what the customer requires and from the customer’s point of view exactly what the Supplier can provide. Inspection does not add value and appears when a step within a process cannot produce its output right-first-time. After several rejected batches and varied excuses from the Supplier. Suppliers The quality of the inputs will influence the quality of the process. Where an inspection point is deemed necessary then it is important that the standards to which the inspection is made are clear. Needless to say. they eventually admitted that they could not maintain the hole-tolerance of +/-1mm. A lot of unnecessary hassle and frustration could easily have been avoided. An example in the electronics industry found a Supplier receiving returns from a customer for oversized mounting holes in a circuit board. whether internal or external. In the interests of process improvement it would be better to not have such inspection. Too often Suppliers take on contracts knowing that they will find it difficult to meet them. Further investigation at the customer’s revealed that the tolerance was a default tolerance and the process could easily cope with a tolerance of +/-2mm. a figure the Supplier could achieve. Often there is little communication between Customers and Suppliers. within critical processes some inspection may be necessary in order to monitor the process and to check for any unforeseen changes to the process. Inspection/ Control points should be positioned close to where an error may be first detected.

Three criteria for measurement (Eckes) are: 1. 3.” Measurement is an important part of process / product improvement. By measuring we can: Ensure customer requirements have been met Monitor progress against objectives Create standards for comparisons – Benchmarking Provide visibility for everyone to understand their performance level Estimate the costs of poor quality Identify areas for improvement However. There needs to be a clear understanding of why something is being measured and whether it is the most appropriate measure. how do you know whether things have improved? Within an organisation there are many things that could be measured but for historical reasons the things that are currently measured may have little to do with the areas that need to be improved. do not complain about illogical behaviour. Measures of effectiveness in which you have no history of customer dissatisfaction should be avoided. The only problem that Management had to confront was to account for the thousands of pounds of missing raw material at the annual stock check. If you don’t measure. Goldratt: The Haystack Syndrome) For example: in a chemical company each shift was measured by its yield. wdp consulting six sigma . Measure only what is important to the customer.” (E. The choice of measurement is important. A poor yield resulted in aggressive messages on the notice board and strong words from the departmental manager. In most cases the operators were powerless to affect the yield yet they still were blamed. If you measure me in an illogical way. Only measure those outputs of a process that you can improve. 2. So to avoid the hassle. the Supervisors underreported the quantities of raw material used. beware what gets measured affects behaviour. This led to an improvement in the yields and a quiet Management. “Tell me how you will measure me and I will tell you how I will behave.Measurement “What gets measured gets done.

It is important that several aspects of the measurement system are known. opinions etc and if possible quantitative data should be collected. The individual is then asked to mark on the line the appropriate level of pain.e. Qualitative data can be made to look more like quantitative data. “moderate pain”. A word of warning. “severe pain” wdp consulting six sigma . yes/no. scales etc. “a little pain”. Thus. incorrectly. For measuring dimensions this will include gauges.In Six Sigma a term that is commonly used is that of Critical to Quality or CTQ. (This can also be referred to at Critical to Customer or CTC). This is better than a qualitative five-point scale: “no pain”. Measurement Systems Measurement Systems are the means by which data is obtained. For measuring equipment then a Gauge R & R study should be carried out before data is collected. Concentrate on the critical few. that the tools used to create the data are suitable for the job. It is often assumed. Accuracy – the measured value reflects the true value Repeatability – the same person repeating the measurement gets the same result Reproducibility – a different person repeating the measure gets the same result Stability – the same measurement repeated over time does not vary Adequate Resolution – the measure has enough different values to pick up differences. if they are critical to the customer then they should be measured. “a lot of pain”. For process related measures it may be the point at which the data is collected or the data itself. For example in the measuring of “pain” in drug trials a 20cm line is used marked “no pain” at one end and “severe pain” at the other.This refers to any features of a product or aspects of a process that a customer would view as being important in the performance of the product or process. Just because a piece of equipment has been calibrated does not guarantee its suitability. there are many things that can be measured. This point is then measured in either centimetres or millimetres. This process is more difficult for qualitative data i.

Too much variation from the measurement system and .Measurement Variability When collecting data it is recognised that there will be variation from the process. Accuracy is were the data on average reflects the true value. How big is the measurement system variability? What are the sources of the measurement variability? Is the measurement system stable over time? Is the measurement system capable of making the required measurements of the process? What needs to be done to improve the measurement process? The purpose of measurement systems analysis is to quantify the variation due to the measurement system itself.the measurement system may give rise to a false belief that a change in the process has taken place.changes in the process may be hidden . but there will also be variation introduced by the measurement system. Questions to be answered. Ideally a measurement system should be both accurate and precise. Sometimes accuracy is called bias – the difference between the observed average and a reference value. Precision is where the data shows very little variation. Measurement System requirements the measurement system must be in a state of statistical control (stability) the variability of the measurement system must be small compared with the manufacturing process variability and any specification limits the increments of measure must be small relative to either the process variability or the specification limits the measurement system should give the true or reference value. This should be close to zero. wdp consulting six sigma . within acceptable limits (accuracy and precision) Accuracy and Precision Confusion often exists between these two terms. It is important to know how much variation is being introduced by the measurement system hence the need for an assessment of the system.

This is a set of trials conducted to assess the Repeatability and Reproducibility of the measurement system.Precision / Accuracy Accurate not precise Precise not accurate Neither precise nor accurate Precise and accurate Repeatability – this is the variation observed when a gauge is used by one person repeatedly to measure the same part. however. As a rule of thumb the tool should be able to detect a change of a tenth of the total process spread. Repeatability is usually much smaller than the precision of the system. For example: 3 operators measure ten parts three times. the smaller the changes that can be detected. Theoretically the observed value should always be the same. Measurement Discrimination – this sis the capability of the measurement tool to detect small changes in a measured characteristic. in practice this is not the case. wdp consulting six sigma . Variation of Measurement System (precision) = Repeatability plus Reproducibility = σ2 rpt + σ2 rpd σ2 ms Stability – this is the variation present when the same part is measured by the same person over time. The higher the discrimination. Gauge R&R This stands for Gauge Repeatability and Reproducibility. In simple terms – several operators measure several items a number of times. This can be assessed using a Control Chart and must be stable. Reproducibility – This is the variation present when the same part is measured by different people.

25) / (0.1 mm? Sample size = ((2 x 0.25. This is simply a metric that reflects the ability of the process to meet the specifications of the process.The calculations associated with this study will be covered in the training. Formula: Sample size > (2 x process sd / difference to be detected)^2 (95% confidence level) Example: Process has an sd of 0.000 wdp consulting six sigma . Sigma is derived from DPMO (Defects per Million Opportunities). What sample size is required to detect a difference of 0. is knowing how much to collect. in collecting data.1))^2 = 25 units Basic principles to follow are: Collect enough data to represent the total variation of the usual process. Try to collect quantitative data Ensure that the data represents “common cause” variation and not “special cause” variation Sigma Calculation Six Sigma has introduced the measure known as Process Sigma or the Sigma metric.000. Sample Size One of the difficulties. It is important to ensure that the specifications used represent the customer requirements and do not as often happens represent the organisations view. There are statistical methods of calculating the minimum sample size but the result of the calculation is often a size that is beyond the practicalities of the study. DPU = number of defects / number of units DPMO = (DPU / Opportunities for error)*1.

4 Process Sigma 2 3 4 5 6 Where Does Sigma come from? The Sigma value is based on mathematical calculations of the area under the Normal distribution curve. Notice that this is not the same value as in the tables.45 = 4. Normal curve Under the standardised Normal curve it is known that 68. Thus if we consider what lies outside of two standard deviations from the mean we can calculate 100-95.5 sigma).55.537 66.210 233 3.5 sigma. The Sigma metric has both benefits and problems Benefits Allows different processes to be compared. If we convert to a proportion (divide by 100) and multiply this by 1.45% of the data lie within two standard deviations of the mean.000.27% of the data lie within one standard deviation of the mean. Accounts for the fact that one process may have many steps whilst another only has a few.000 we have the number of defects we would expect in every million items. This is because the table values are adjusted to allow for a long-term shift of the process (estimated at 1.807 6. So at the level of two sigma we would expect 45.000 defects. (45. Thus two sigma is really 0.000 defects). 95.DPMO is then converted to Sigma from tables (see Appendix) DPMO 308. Moves the mindset away from percentages (many managers would stop improvement if they had a process that was producing 95% good product) Provides a baseline to compare improvement Disadvantages wdp consulting six sigma .

From tables Sigma = 3. DPMO = 50.Is based on the theoretical Normal distribution Is used as an exact measure when at the edges of any distribution the ability to predict accurately declines The 1.05. wdp consulting six sigma .5 sigma allowance may be inappropriate Sigma can be inflated by increasing the number of opportunities for error Some organisations have avoided the problem of defining an “opportunity for error” by simply using defects per unit.1. Thus if a process has a failure rate of 5% then DPU = 0.000.

process improvement cannot be achieved. Write your signature several times – they are different. a screw has to be put into a block.Statistical Thinking Understanding Variation Variation appears in all processes. Without understanding variation. In essence statistics is the study of variation. Or a bearing is produced slightly large creating increased friction causing a poor performance of the product. This has been the basic premise of Quality improvement. You can predict what is going to happen and therefore be prepared. Measure the output from a machine – they will never be exactly the same. median. This is especially true for the output from a manufacturing process or service provision. most things do vary and so we have to allow for the variation. wdp consulting six sigma . mode. no one is exactly the same – even identical twins have differences. A service call takes longer than expected thus impacting on planned calls for the rest of the day. standard deviation and range. If you knew that your journey to work would take exactly twenty minutes every time then you know what you have to do to arrive at work on time. However. For instance. In order to describe the amount of variation then a number of measures are used: mean. Together these will describe the variation. Variation presents problems in both manufacturing and services. Understanding the variation within a process will help determine what needs to be done to reduce it. Variation can mean that the screw and block do not match. These values will be discussed in basic statistics. Why reduce Variation? Life becomes so much easier if every time you do something the outcome is exactly the same. Understanding the variation will allow us to reduce the main sources and thus incur less problems and failures. Look around you.

a rigorous and logical problem solving methodology. such as RISE. To identify and remove Special Causes for which the root cause is unknown. Use process improvement.there all the time . is required. others may be difficult. Six Sigma looks at reducing Common cause variation and in doing so may change the process such that Special Causes are eliminated. In terms of responsibilities: Common causes – management must change the system.Special Cause and Common Cause Variation Variation can be of two types: Common Cause Special Cause .part of the natural variation inherent in the process . Special Causes must be removed to allow a clearer understanding of the Common Cause variation. Deming estimated that 94% of processes exhibit Common Cause variation and 6% exhibit Special Cause variation Sources of Variation Variation can be found in a number of sources.unpredictable It is important in starting process improvement to be able to distinguish between Common Cause and Special Cause variation.exceptional . For a process the sources of variation may be understood by use of MEPEM or the M’s. if Special Causes are present in the process then they need to be identified and removed. However. some of which may be easy to affect. They must also be removed because they may fail to appear during the data collection and give a false sense of security to the improved process only to appear later and produce unexpected rejects. Six Sigma methodologies Special cause – operators/administrators to identify cause and eliminate. wdp consulting six sigma .

5. ink viscosity Wear – nib wear. hand size Environment – temperature System to system – type of paper In looking for improvement then there needs to be an understanding of what factors are influencing these sources in order to make changes to reduce the effect. This focuses on the specification limits. Taguchi points out that the philosophy should be to focus on the centre point (mean) and minimise variation around that point irrespective of the specification limits. However. 4. pressure applied. 3. Consider a fountain pen. Taguchi’s Loss Function Six Sigma tends to have the philosophy that the aim is to reduce variation such that it is well within the customer specification. angle of writing. size of ink cartridge/ holder. wdp consulting six sigma . size of casing. Piece to piece variation Changes over time (wear) Customer usage External environment Interaction with neighbouring sub-system This is known as a Noise Tree.MEPEM Materials Equipment People Environment Methods M’s Materials Machines Men Mother Nature Methods Measurement Maintenance For a product then the above categories can be used or the following: 1. amount of ink Customer Usage – user style. storage. 2. Why might a pen not function as intended? Piece to piece – variation due to size of nib.

Understanding Variation Variation around target (Loss function) Loss Loss Function Target Value 6σ Although both will give potentially similar improvement. The Normal Distribution The Normal distribution is a theoretical distribution created by Gauss. the failure rate goes down and the Sigma value goes up. the philosophies are a little different. By using specification limits then the danger is the process can appear to have been improved by simply widening the limits. Equation: Its importance is for several reasons: It is a good approximation to the data from many processes It is symmetrical It can be used to estimate the probability of events occurring outside of defined parameters wdp consulting six sigma . Taguchi’s Loss Function can be a little too hypothetical for many people who need to have cut-off points to determine whether a product or service is acceptable or not.

73% of the data lies within three sd of the mean Calculating areas under the Normal Curve. of the mean 95.45% 99.Six Sigma Metrics Area under Normal Curve µ−4σ µ−3σ µ−2σ µ−1σ µ µ+1σ µ+2σ µ+3σ 68.45% of the data lies within two sd of the mean 99. Calculate the probability that a value lies beyond 14 cm Z = (14 – 10) / 2 =2. Z = (x – mean) / sd Look up the z value in tables.023 that a value will lie outside of 14cm Or we would expect 2. Example A process has a mean of 10cm with an sd of 2.73% 99.02275 i.e.27% of the data lies within one sd.99% µ+4σ 6σ Areas under the Normal distribution 68.0 gives 0.0 From tables a Z= 2. there is a probability of 0.3% of the data to lie outside of 14 cm wdp consulting six sigma .27% 95.

These boundaries should be carefully defined to make it clear in which class each result should fall.Decide on the data to be collected 2. Class Width – the width of the interval into which data is grouped.Create a Check Sheet on which to record the occurrence of each event or data value.Collect data wdp consulting six sigma .Displaying Data A graphical representation of data can have far more impact than simply a set of numbers.10 –5. For numerical data the groups may be in the form of class widths.1 – 5. Check sheet (Tally Chart) This is an easy to understand method of grouping data. For count data the groups will be simple categories. It can also help in the understanding of the variation.19cm is clearer than 5. Class Boundaries – the extremes of the class widths. Frequency distribution – is a display of the pattern created by plotting data frequencies versus quality characteristics.Decide on the period of data collection and from what source 3. 4. Here we will look at: Check Sheet Histograms Dot Plots Pareto Chart Run Charts Scatter Diagram Definitions: Frequency – is the number of times a value or values within a defined category occurs.2 cm. 1. For example: if data is to two decimal places then a boundary of 5.

Decide on the process measure – data measured on a continuous scale is best 2.58 I II IIII IIIII IIII IIIII IIIII II IIIII II IIIII III I Total Tally Total 1 2 4 9 12 7 5 3 1 44 6σ Histogram This graphical method displays the number of data points in each class width as a bar. Creating a Histogram 1.: Histogram Check Sheets Diameter 5.54 5.51 5.57 5.55 5. Thus it is possible to have bars of different widths. high pattern. However. Guidance: Number of data points <50 50 .50 5. spread out pattern. In its purest state the area of the bars of the histogram are proportional to the frequency.20 wdp consulting six sigma .This type of chart is often the forerunner of a graphical method e.g. too many class widths will produce a flat.52 5.56 5.Decide on the class width – be practical Note: too few class widths will produce a tight. The height of which represents the number of data points in that class width.53 5.Gather the data – collect enough data to give a good representation of the process (some recommend 50. for convenience the bars are kept the same width so that the height represents the frequency of occurrence.100 100 -250 >250 Number of class widths 5–7 6 – 10 7 –12 10 .100 points) 3.

52 5. 4.Interpret the Histogram Interpretation Centre – compare with customer requirements Spread – is variation well within the customer requirements? Shape – Normal distribution? Bi-modal distribution? Skewed? Any strange shapes? If the number of measurements taken were very large and if the class widths were very small then the Histogram would look more like a curve than a series of steps.51 5.54 5.Use a check sheet or tally chart to sort the data into the appropriate class widths.Calculate the class boundaries 5. wdp consulting six sigma .5 5.56 5.55 5. 6. where each data value is represented as a dot.53 5. The Dot Plot is created in the same way as the Histogram except that the frequency of each class width is represented by a number of dots rather than a bar.Draw the Histogram 7. similar to the Histogram. Histogram Histogram 14 12 10 Frequency 8 6 4 2 0 5. This curve is known as the Frequency Curve and will be used to represent the shape of the distributions from where the data comes.As a rule of thumb the number of classes equals the square root of the number of data points.57 5.58 Tube diameter (cms) 6σ Dot plot A graphical method. Each data point within the class width is then represented by a dot.

56 5. In considering the 80:20 rule.50 5.Draw a bar chart to represent the frequency of occurrence of each cause 6.51 5.58 Diameter of tube (cms) 6σ Pareto Chart This was first created by Vifredo Pareto in the 19 Century.Order the causes according to their magnitude 5.57 5.55 5. That is 80% of the problems are caused by 20% of the sources.Interpret the results The largest occurring cause should be the one first dealt with as this contributes most to the problem.52 5.Quantify the number of times each cause creates the defined problem 4. He was looking at wealth distribution.54 5.Decide on the problem area to be considered 2.Dot Plot Frequency 5. Juran talks about the vital few and the useful many.Identify the causes of the problem 3.53 5. This led to the 80:20 rule. Creating a Pareto Chart 1. Pareto Charts Pareto Chart invoice incorrect Reason for Complaint wrong quantity wrong product discoloured material thin material poor seals too late 0 5 10 15 20 25 30 Number of Complaints 6σ wdp consulting six sigma . He discovered that 20% of the people held 80% of wealth.

Plot the data 6. This is represented by the degree of Correlation. A correlation coefficient of around zero suggests no association.Create a graph with a vertical line (y axis) and a horizontal line (x axis) 4.Decide on the variables to be studied 2.Interpret the chart (This is the basis for creating an SPC chart) Run Charts Run Chart 2 1.Plot the paired data values on the graph 5.Decide on the process measure 2.Determine the scale to be used on the y-axis Use a scale such that the data collected uses approximately 2/3rds of the y-axis scale 5.Collect paired samples relating to the chosen variables 3.Run Chart This is a graphical representation of data over time.Gather data over time 3. It is important to note that a correlation does not prove that variation in one variable causes the variation in the other.Calculate the mean of the data 7.5 % Rejects 1 0. wdp consulting six sigma .Create a graph with one variable on the x-axis and the other as the y-axis 4.Draw the mean as a horizontal line on the graph 8. Creating a Scatter diagram 1. A correlation coefficient of +1 or –1 shows a high degree of association.Interpret the data Look for either positive / negative relationships or simply no relationship. Creating a Run Chart 1.5 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Week Number 6σ Scatter Diagram This is a graphical method of observing the relationship between two variables. It is used to assess whether the variation in one variable could be related to the variation in another.

Note: beware of false relationships.: number of storks and the number of babies born in an area (which proves that storks really do deliver babies). There is often an unseen variable to which both are related.g.A lot can be gained from the plot rather than simply calculating the correlation coefficient. Scatter Diagrams Scatter Diagram 800 Number of babies born 700 600 500 per year 400 300 200 100 0 0 50 100 150 200 250 300 Number of storks in area 6σ wdp consulting six sigma . – the degree of urbanisation in the case of the storks – the storks nest in the chimneys. E. It is often seen that apparent relationships exist between two variables.

6.4.5. mode.xn ) / n 3. Measures of Location : mean. Derek Rowntree 0-14-013632-0 There are a number of parameters that typical in any statistical calculation. Measures of Location Three main measures : Arithmetic mean (Mean) Median Mode Mean (arithmetic mean) X X = ( x1 + x2 + x3 + ……….10 3.9.4.9. 4.9.11 : median = mean of 5 + 6 = 5. 5.10. 6. Ex: 3. 5. median Measures of Spread : range.4. 9.5. Ex: 3.5.5 wdp consulting six sigma .10. standard deviation. 10 therefore =5 If there are an even number of data items then the Median is the mean of the two middle values when the data is placed in order. inter-quartile range If we know these parameters then we have a good idea of what the distribution of the data looks like.5. Statistics without Tears.6.10 = 42 / 7 = 6 where n is the number of data points Example: Median (middle value) The middle value when placed in numerical order.5.6.6.4.5.11 3.Basic Statistics Recommended book. These parameters tell use where the centre of the data is (location) and how far the data spreads.5.9.5.

6. There may be one or more points with this value.5.10 Five occurs twice and is therefore the most frequently occurring value.9. Mode wdp consulting six sigma .4. =5 Relationship between Mean. Ex: 3. multi-modal sets of data.Mode This is the most frequently occurring value. Median. Hence there are bi-modal.5. Mode and Median Symmetrical data Mean = Mode = Median Measured Characteristic Mean.

Positive skew data Mode<Median<Mean Mode Median Mean Negative skew data Mean<Median<Mode Mean Median Mode Generally: Mode = Mean –3 (Mean – Median) wdp consulting six sigma .

Measures of Spread
Look at the data points for A, B and C. They have the same mean but there spread is very different. Hence knowledge of the spread of data is also important.

A
12 11
10 12 14

0

1

2

3

4
0

5
2

6
4

7
6

8
8

9

13 14 15 16 17 18
16 18 20

B

0

2

4

6

8

10

12

14

16

18

20

C

0

2

4

6

8

10

12

14

16

18

20

wdp consulting

six sigma

Range
Range = maximum value – minimum value A: 0 B: 8 C: 20

The Range is easy to calculate but is influenced by extreme values

Standard Deviation (sd)
Sometimes called root mean square deviation. Two measures; Population and Sample Population sd (only used when you have all the data possible)

σ=√

∑ (X − X )

2

n

Sample sd (most commonly used when calculating based on a sample of the data)

s=√
From the dot plots above: A: 0

∑ (X − X )
n−1

2

B: 2.83 C: 7.48

It is rare to have all the data so it is usually the sample standard deviation that is calculated. Why two different calculations? It can be shown that when taking samples of data then the second equation on average gives a better estimate of the true population sd than the first. In statistical jargon the second equation is said to be BLUE (Best Linear Unbiased Estimate).

wdp consulting

six sigma

The standard deviation is not so easy to calculate by hand compared to the Range but this is not a problem with calculators and computers. Remember to use the “n-1” calculation. The standard deviation is less influenced by extreme values compared to the Range.

Interquartile Range
This shows the range of the middle 50% of the data. Interquartile range = Q3 – Q1 Where Q1 = 25% value and Q3 = 75% value (can be found from a cumulative frequency) A: 0 B: 12 – 8 = 4 C: 16 – 4 = 12 Calculators do not calculate this value and hence is not used very often. Its advantage is that it is uninfluenced by extreme values.

wdp consulting

six sigma

Statistical Distributions
As seen in Histograms, if we take enough data from a population and break it into small enough class widths then the resulting shape will approximate to a curve.

Bi-modal

The description of the data is known as the frequency distribution and the curve can be represented by a mathematical function.
wdp consulting six sigma

This distribution is called the Probability Density Function (p.d. 68. X is the value being considered. x-bar is the mean value of the data. sigma is the standard deviation of the population and may be estimated by the sample standard deviation. wdp consulting six sigma . it is a good representation of much naturally occurring data it is symmetrical other distributions look like a Normal distribution in certain circumstances Because of its properties there is almost an obsession with it.In order to use the frequency distribution for calculations. It is now possible to calculate areas under the curve. the data is standardised such that the area under the curve equals one.) Specific Distributions Normal Distribution This is the most commonly used p. Z = (X – X ) / σ Where Z is in fact the number of standard deviations that the value is away from the mean.f. This is used in conjunction with the standard tables.f.27% µ−4σ µ−3σ µ−2σ µ−1σ µ µ±1σ µ+2σ µ+3σ µ+4σ The area under the Normal curve can be calculated from standardised Normal distribution tables.d.

2420 Thus we expect that 24.6 – 50. What percentage of the pipes will have: a.7 gives a value of 0. Calculate the % of balls that will have a diameter greater than 57mm. a diameter greater than 50.5mm.2% of the balls will have a diameter of > 57mm.2 wdp consulting six sigma . = 21.2119. diameters between 49.7 From tables Z = 0.2% > 50. Z = (X – X ) / σ Z = (57 – 50) / 10 = 0. The area being tabulated is usually shown along with the title. Z = (49. Z = (50.5 = +0.2) / 0.6mm and 50.5 = -1.6 – 50. µ−4σ µ−3σ µ−2σ µ−1σ µ µ+1σ µ+2σ µ+3σ µ+4σ Example: The mean diameter of a pipe is 50.d.6 b.Example: The mean diameter of a steel ball is 50mm with a s.6mm c.6mm? a.2 mm with a sd of 0.8 From tables 0. of 10mm. Standardised Normal Tables Take care if using these tables as the areas are sometimes tabulated differently. a diameter less than 49.6mm? b.2) / 0.

5)) = 67.3% Normal Distribution and Link to Six Sigma Graphical representation Six Sigma Three Sigma Target Lower Spec Measured Characteristic Upper Spec At three sigma the data from the distribution just fits within the specification limits. Present / Absent This distribution is often used when looking at numbers of defectives. Yes / No. wdp consulting six sigma .2%>50. Binomial Distribution The Binomial distribution occurs where there are only two possible outcomes: Pass / Fail.6 and 11.5% < 49.6 c. Good / No Good.2 + 11.From tables 0. if 21.5% < 49. = 11. At six sigma the variation of the data has been reduced such that the distribution of data would fit exactly twice into the specification limits.1151. That is the part or service is either defective or it is not.6 then the remainder must be between these values ie: (100 – (21.

5)0 = (0. Example: Tossing a coin Probability of a head = 0.p.n Cx px q n-x where nCx = (n!) / ((x!)(n-x)!) and n! = 1 x 2 x 3 x 4 x 5 x …………n p = probability of success.5)3 = 0.p The sd of a Binomial distribution is √n.5)3(0. n= number of data points The mean of a Binomial distribution is n .5)1 = (3) (0.5)3 = 0.5 Probability get three heads in three tosses Pr (3 Heads) = ((3!) / (3!)(0)!) (p3)(q0) = ((3 x 2 x 1) / (3 x 2 x 1))(0.5. q = probability of failure.125 Probability of getting two heads in three throws Pr (2 H) = ((3!) / (2!)(1!)) (0.5)1 = ((3 x 2 x 1) / (2 x 1)(1)) (0.q When n is large and p and q are not small then the Binomial distribution can be represented by the Normal distribution.5)2(0. probability of a tail = 0.375 Note: Pr (3H) + Pr (2H) + Pr (1H) + Pr (0H) = 1 wdp consulting six sigma .5)2(0.

(n > 20. x is the number of events For a Poisson distribution : mean = µ. variance = µ For large n and a small probability of success (occurrence) then the Binomial distribution tends towards a Poisson distribution.0067 Pr (1 failure) = (e-5)(51) / 1! = 0. Pr ( X) = ((e-µ)( µx)) / x! Where e = 2.1775 Pr (0 failures) = (e-5)(50) / 0! = 0. n = 500 p = 0.0337 Pr (2 failures) = (e-5)(52) / 2! = 0. Time between events (e.0842 Pr (3 failures) = (e-5)(53) / 3! = 0. Example: Probability that a product fails in its first year is 0.1404 Etc wdp consulting six sigma . What is the probability of four failures from 500 products sold. µ= mean = n.1).Poisson Distribution Events whose probability of occurrence is low tends to give a Poisson distribution.01 = 5 Pr (4 events) = (e-5)(54) / 4! = 0.01 µ = n.p = 500 x 0.g.p. Time between failures) can also be modelled by a Poisson distribution.01. p< 0.7183.

roll 5 die 20 times record each individual value and the mean of the five dice. The Central Limit Theorem states that the distribution of means from samples of a population tends to be Normally distributed with a sd of the sample means equal to ( / n) where n is the number in each sample taken. The Central Limit Theorem can be demonstrated by a simple dice experiment. Calculate range of the means Compare the histograms and ranges. a. Calculate range of the means b. roll three dice 20 times record each individual value and the mean of the three dice. Plot a histogram of the mean values Calculate range of the individuals. Plot a histogram of the individual values. Plot a histogram of the individual values.Central Limit Theorem This is important in understanding the relationship between a sample taken at random and its underlying distribution. The value ( / n) is known as the Standard Error of the Mean. wdp consulting six sigma . Plot a histogram of the mean values Calculate range of the individuals. This will be used in Statistical Process Control.

For example.Hypothesis Testing Too often people collect data without thought about what they are testing. Single-sided Hypothesis test H0 : µ1 = µ2 ie. i.e. statistics only ever rejects the Null hypothesis. Collect more data and we might change our minds. wdp consulting six sigma . if we wish to compare a set of data with a previous known mean. That is if H0 is not accepted then what is the decision. Jargon H0 : Null hypothesis – what is to be tested Eg. has the mean of the process changed then we need to decide how large a difference are willing to accept before declaring the process has changed and what risk are willing to take on making the wrong decision. In defining a significance test then a decision must be made beforehand what the Alternative hypothesis will be. The decision of a single-sided test or a double-sided test will affect the probability of ‘accepting’ or rejecting the Null hypothesis. the decision about tests to be applied should be taken before the experiment has been run or the data collected. It is all too common for people to collect data organise it and then test the top result against the bottom result. In statistics. The means of two samples are equal H0 : µ1 = µ2 For any Null hypothesis there must always be an alternative (H1). we are only interested in a difference in one direction. It never ‘accepts’ the Null hypothesis but rather states that ‘there is no evidence to reject the Null hypothesis’. Eg: the means of the two samples are not equal H1 : µ1 ≠ µ2 or H1 : µ1 > µ2 Note: to be correct. H1 : µ1 > µ2 or H1 : µ1 < µ2 Double-sided (two-sided) Hypothesis test H0 : µ1 = µ2 H1 : µ1 ≠ µ2 ie. This is where statistical significance is used. often totally ignoring all the data in between. we are interested in whether or not there is a difference and not the direction. The use of a hypothesis test also leads to a decision of at what point is the Null hypothesis rejected.

Significance Testing
In comparing any sets of data then there are a number of options that may arise. This is best represented as a table. Actual Situation No Difference difference
No difference Difference √ Type I Type II √

Decision Made

If we make the right decision compared to the actual situation then all is well. However, we need to be aware of the risks that we make an incorrect decision. i.e. we conclude that there is a difference between the groups when they are from the same distribution. This is known as a Type I error (α). We could also conclude, in error, that there is no difference when the two groups come from different populations. This is a Type II error (β). In significance testing it is necessary to decide on the risks we are prepared to take in making an incorrect decision. In most cases it is the Type I error (α) that we set. That is; what is the probability we can accept that we will conclude that there is a difference when there is no difference? i.e. we will reject the Null hypothesis. The Type I error is often set at 5% although in risk averse situations value of 1% may be set. In theory any value could be used. By choosing the Type I error we automatically set the Type II error. If the Type I error is set at 5% then the Type II error will be around 10%. That is that there is approximately a 10% risk that we will say that there is no difference when the samples are in fact different. As the Type I error goes down the Type II error goes up and vice versa. (1 - β) is known as the Power of the test to discriminate between groups. As the number of data points in the sample increases then so does the power. (100 - α) is referred to as the confidence level. Knowing whether a test is one-sided or double-sided and the risk that we are prepared to take in making an incorrect conclusion we are in a position to use statistical methods to compare sets of data.

wdp consulting

six sigma

t- Distribution

A distribution can be standardised by the use of the equation: Z = (X - µ) / σ According to the Central Limit Theorem the distribution of sample means with have a standard deviation of (σ / √n). Replacing this in the Z calculation we have: Z = (X - µ) / (s / √n) To show that the estimate of the standard deviation has come from a sample the symbol s is used rather than σ. This new formula has a value that follows the Student’s t distribution. (So named because the creator of this distribution did not want to be known and called it “Student’s t” to hide his identity). Thus when comparing data from samples we use t tables rather than Normal tables. As the number of data points in a sample increases then the t distribution tends towards a Normal distribution. If we have collected a sample of data, with mean x-bar, from a process and wish to benchmark it against a previously recorded value( ) to determine if the process has changed and wish to take a risk of 5% that we will conclude that there is no difference when there is, then the test would look like:
H0 : x-bar = µ H1 : x-bar ≠ µ (ie. two-sided test) α = 0.05 (5%)

Pr ( -t.975 < (x-bar – µ) / ( s / √n) < +t.975)
If x-bar lies in either of the shaded areas then we would reject H0 and conclude that the process has changed.
Example:

An efficiency expert claims that a new machine must reduce the time by eight hours to be acceptable. Six experiments gave an average decrease of 8.4 hours with a sd of 0.32 hours. Should the new machine be bought? H0 : x-bar = µ H1 : x-bar > µ (one-tailed test) where µ = 8 hours

N = 6 , x-bar = 8.4, s = 0.32 one-sided test use α = 0.01 (1% level)

Pr ( (x-bar – µ) / ( s / √n) > t.99)
wdp consulting six sigma

From t tables α = 0.01, degrees of freedom (ν) = n-1 = 5

t value = 3.365

( (x-bar – µ) / ( s / √n) > t.99)
(8.4 – 8.0) /(0.32/ √6) >3.325)

3.062 > 3.325 since this is false, have no reason to reject the Null hypothesis. That is the new machine has not reduced the time by more than eight hours. However, had we used α = 0.05 (5% level) the answer would have been: From t tables α = 0.05, degrees of freedom (ν) = n-1 = 5 t value = 2.015

3.062 > 2.015 is true therefore we would reject the Null hypothesis with 95% confidence. This reinforces the need to decide on the parameters of the test before seeing the data.

Confidence Limits
When we take a sample, the mean of the sample is an estimate of the mean of the underlying population. However, if we were to take a different sample from the same population the estimate of the mean would be slightly different. By using confidence limits we can predict what the range of estimates of the population mean are likely to be.
Example;

What are the confidence limits for a sample of 12 measurements with an average breaking strength of 7.38g and an sd of 1.24g. n=12, x-bar = 7.38, s = 1.24 assume α = 0.05 t value = 2.201

from t tables degrees of freedom (ν) = n-1 = 11, α = 0.05

Pr ( -t < (x-bar – µ) / (s / √n) < +t ) = 0.95
Pr( -2.201 < ( 7.38 – µ) /( 1.24 / √12) < 2.201) = 0.95 Pr( 7.38 –2.201 (1.24 / √12) < µ < 7.38 +2.201 (1.24 / √12)) = 0.95 Pr (6.59 < µ < 8.17) = 0.95

wdp consulting

six sigma

79 wdp consulting six sigma . This may be written as 7.38 ± 0.56 to 8.Thus the 95% confidence limit for the estimate of the population mean is 6.20.

N = 120.18. df = 5.16 / √6) 20 ± 3.025) = 1.16 2 tailed test.5.16 t (0. 16.Comparisons Statistical tests are often used to compare data sets e.5 hours. t-test Can be used for : Comparison of the mean of a sample with a population mean Comparison between the means of two samples Comparison of a Sample Mean with a population mean Example A company claims that the average life of a battery is 21. 22.025) = 2.5 hours.68 < µ < 23.g. Note the effect of increasing the number in the sample. t (0. Tests of six batteries gave the following results: 19. (Most of the analyses will be shown using Excel). 25 hours Do these results confirm the company’s claims? H0 : x-bar = µ H1 : x-bar ≠ µ µ = 21.32 batteries could have come from a population with 21. (s / √n) 20 ± 2.571 (3. 20.32 16. a before and after improvement.98 ( 2 tail) wdp consulting six sigma .571 x-bar ± t. s = 3. x-bar = 20. x-bar = 20. sd = 3.

group A and group B.67 0.5 hours. This is usually held on the original Excel CD or may be held on the computer but not activated.25 1.03 1.63 Group B 1.62 1.78 1.57 In this case we would reject the Null hypothesis.38 1. the batteries did not come from a population with a mean of 21. In order to do this Excel needs an add-in for Excel called ‘data analysis’.84 1. Un-paired t-Test Group A 1. The mathematical calculations will be shown at the end of this chapter for those who are interested.84 0.22 In this example we wish to compare two sets of data. Otherwise the data will be analysed through Excel.e.43 < µ < 20. i.98 1.57 19.41 1.69 0.98 (3.04 1. Comparison of Two Groups There are two situations one where the data from the two groups is linked (paired) and the other where the data is not linked.20 0.07 1. Select Tools Select Data Analysis wdp consulting six sigma .22 1. Analysis using Excel Many of the simpler statistical tests can be run in Excel.75 1.16 / √120) 20 ± 0.77 1.20 ± 1.22 0.

(Output range). Ensure that the output range does not go over your data. Select your alpha level (0. two sample assuming unequal variance.05 in this case). (selecting this option rather than assuming equal variance avoids having to make the decision whether the variances of the two groups are similar). wdp consulting six sigma .Select t-test. Identify an area on the spreadsheet to display the results. Identify the data for variable 1 (group A) and variable 2 (group B).

Note: that the results for a one-tail test and a two-tailed test are calculated.If the calculated value (t-stat) is greater than ‘t critical’ then the difference is said to be significant. Calculated Value Exact probability Table Value If the Calculated value is less than the Table value (t critical) then the difference is not significant If the Calculated value is greater than the Table value (t critical) then the difference is statistically significant df = degrees of freedom wdp consulting six sigma . is this a one-tailed or two –tailed test. Hence it is important to have made the decision before hand as to the Hypothesis i. If the t-stat is less than the ‘t critical’ then the difference is said to be not significant. The results also show the exact probability that the difference could have occurred by chance.e.

Select tools Select data analysis Select t-test: paired two sample Identify data set 1.41 1.62 1.77 1. wdp consulting six sigma .03 1.22 0.84 0.75 1.22 Here the groups’ data values are paired in that each group produces output on each machine. As described in the unpaired t-test. data set 2 and the output range.84 1.63 Group B 1. Set the alpha value.69 0.38 1.98 1.Paired t-Test This is a special case as we have additional information in the pairing of the results.04 1.78 1.07 1.22 1.67 0.20 0. Machine 1 2 3 4 5 6 7 8 9 10 Group A 1.25 1.

Therefore there is a significant difference between the two groups.262.1389 which is greater than the t-critical two-tail of 2. (d.In this case the degrees of freedom are 9 since there are ten pairs. The t-statistic is 3. = n-1).f. wdp consulting six sigma .

75 1. Identify the data as one group.98 1.20 0.22 In this case we only have one factor i. The t-test is a special case of ANOVA for one or two groups. Ensure that the output range does not go over your data.67 0. Identify an area on the spreadsheet to display the results.77 1.e.41 1.22 0. ANOVA can handle more than two groups.69 0.07 1. wdp consulting six sigma . We could have several groups within this one factor. one type of product. Select your alpha level (0.38 1.84 1. machines.Analysis of Variance (ANOVA) This is a method of analysing the variation due to different groups or factors.84 0..63 Group B 1. One-way ANOVA Group A 1.03 1.22 1. Select Tools Select Data Analysis Select Anova: single factor Note: Excel can only handle two factors at a time.25 1.78 1. (Output range).05 in this case). sites.04 1. However it can handle several groups of data for any one factor.62 1.

The results also show the exact probability (P-value) that the difference could have occurred by chance. Critical value = 4.65 Since the calculated value is less than the critical value we conclude that there is no difference between the groups wdp consulting six sigma .If the calculated value (F) is greater than ‘critical value’ then the difference is said to be significant.41 Calculated value = 1. If the F-value is less than the ‘F critical’ then the difference is said to be not significant.

(Output range). Ensure that the output range does not go over your data. As for the one-factor ANOVA we can have several groups of data for each factor.05 in this case). Since there are two factors there are two comparisons shown as rows and columns. If the F-value is less than the ‘F critical’ then the difference is said to be not significant. The number of groups for each factor does not have to be identical. wdp consulting six sigma . If the calculated value (F) is greater than ‘critical value’ then the difference is said to be significant. Method From the Data Analysis select ANOVA two factor without replication. output from three machines on five different days. Otherwise select. Select your alpha level (0. with replication. In the example rows are machines and groups are columns. e. The factors here are machines and days. Calculated values Critical values Since both calculated values are greater than their corresponding critical values. Identify an area on the spreadsheet to display the results.Two-way ANOVA Here we are considering two factors. if there is only one data point per combination of factor settings. The results also show the exact probability (P-value) that the difference could have occurred by chance.g. Identify the data as one group. we conclude that there is a difference between machines and a difference between groups.

55 C 21. However.42 which is greater than the F-crit of 6.15 7. Excel does not wdp consulting six sigma .56 7.20 7. thus there is evidence of a difference between the drivers.944.53 21.19 7. However. We want to know if there is a difference between the cars and / or the drivers.18 21. thus there is no evidence of a difference between cars. For example comparing several cars driven by several drivers. the F-value for drivers is 17.51 64. Time Car 1 2 3 Σ Driver A 7.ANOVA becomes more useful in the case where is more than one category being assessed.17 7.1428 which smaller than the F crit 6.19 21.13 7.14 21.18 7.42 Σ B 7.944.53 Select Tools Select Data Analysis Select Anova :two factor (in this case without replication) Identify data Select alpha Identify output range In this case Rows = cars Columns = drivers From the Anova table the F-value for cars is 1.49 21.

wdp consulting six sigma . This can be identified using mathematics as shown in the mathematical analysis. other than it is between drivers.show where the difference lies.

1 the die is considered to be unbiased.3)2 / 8.3 χ2 = ((observed – expected)2 / (expected) df = 6 – 1 = 5 (ie number of possibilities) χ2 = (12 – 8. If the die is unbiased then we would expect a similar number of each face. theoretically from 50 throws.3)2 + (6-8.3)2 + ( 10 – 8.Chi – Squared ( χ2 ) used to calculate confidence limits of sd’s used to test goodness of fit of models to data used in contingency tables Goodness of Fit Example throw a die 50 times and record the results.1 Since 3.3 4 9 8.3)2 + (7 – 8.53 < 11. wdp consulting six sigma . We have no evidence to reject the Null hypothesis.3 3 6 8.3 6 10 8.3 2 6 8.3 5 7 8.3 times (50 /6) Face Actual : observed (O) Theoretical : expected (E) 1 12 8.3 = 3.34 / 8.53 from tables χ2 (0.3)2 + (6 – 8.3 = 29.05) on 5 df = 11. we would expect each face to appear 8.3)2 + (9 – 8.

Ensure that the output range does not go over your data. the data that will create the equation Y= mX + constant The screen shows a confidence limit of 95%. This is achieved by minimising the variation from the data points to the line. The mathematical equations needed to calculate the regression line can be quite involved and so it is suggested that any linear regression is carried out using Excel or statistical software. Identify the independent data (X) and the dependent data (Y). (Output range).Regression This is a methodology to analyse the relationship between two quantitative variables. Identify an area on the spreadsheet to display the results. Method From the Data Analysis select Regression. The equation of a straight line relationship between variables is y = a + bx where a and b are constants a: intercept – the point where the line crosses the y-axis b: slope – the gradient or slope of the fitted line. i. Excel only performs simple linear regression. For more complex regression statistical software should be used. wdp consulting six sigma . This is equivalent of alpha = 0.05. Simple Linear Regression Dis tance travelled over Tim e 70 60 50 Distance (miles) 40 30 20 10 0 0 5 10 15 20 25 30 35 40 45 50 55 60 Tim e (m inute s) 60 mph 30mph In its simplest case the method fits a straight line to the data in such a way to give the best fit.e.

the slope gradient and the constant.e. wdp consulting six sigma .The output for the regression shows: The correlation coefficient R squared The significance of the slope (from an ANOVA table) The coefficients for the equation i. There are other options available in Excel but at this stage it is probably best to ignore them.

20 0.62 1.101 x 0.e.38 1.05 two tailed test) df = na+nb-2 = 18 : t = 2.333 Combined sd: sdcom = √(((na – 1) s a 2 + (nb – 1)s b 2) / (na +nb –2)) = √ (((9)0. x-bar = 1. (difference) = sdcom √((1/na) + (1/nb)) = 0.03 1. sd = 0.98 1.136) ± 2.987 wdp consulting six sigma .69 0.147 Comparison of means Pr ( difference) ± t (se(difference)) from tables t (0.84 0.75 1.136.22 1.325 – 1.Mathematical Analysis Unpaired t-test In this example we wish to compare two sets of data.22 0. Group A 1.07 1.67 0.189 ± 0.78 1.329 √ (2/10) = 0.04 1.84 1.325.63 Group B 1. group A and group B. sd = 0.77 1.22 H0 : x-bar = µ H1 : x-bar ≠ µ Group A : n =10.325 Group B : n =10.101 Pr( 1.147 0.41 1.25 1.3252 + (9)0.329 Also s.3332) / 18 = 0. x-bar = 1.

19 0.05). Paired t-test This is a special case as we have additional information in the pairing of the results.77 1.2 -0.78 1.38 1.63 Group B 1.20 0.20 0.189.69 0. we reject the Null hypothesis that there is no difference between the groups.e.41 Mean difference = 0. there is no difference between the groups.53 -0.176 since this range includes zero we would not reject the Null hypothesis i.13 0.84 0.09 0.75 1.f. s = 0.75 1. t (9 d.84 1. wdp consulting six sigma .77 1.69 0. Calculate the difference between the pairs.41 1.03 0. from tables (α = 0.98 1. = s / √n = 0.38 1.25 1.41 1.22 Here the groups’ data values are paired in that each group produces output on each machine.189 / 0.63 Group B 1.f.08 0.262 since 3. Machine 1 2 3 4 5 6 7 8 9 10 Group A 1.22 0.67 0.04 1.22 difference 0. n=10 s.22 0.84 0.07 1.62 1.06 0.22 1.145 on 9 d.98 1.04 1. Machine 1 2 3 4 5 6 7 8 9 10 Group A 1.35 0.22 1.67 0.03 1.25 1.84 1.0601 t = 0.) = 2.e.78 1.62 1.03 1.07 1.0601 = 3.798 < difference < 1.190.190 / √10 = 0.145 > table value .-0.

412 + …………….78 1.67 0.20 0.07 1.69 0.36)2 / 10) - .9481 for the two sample case this is also known as the Residual Sum of Squares wdp consulting six sigma + (∑xb)2 / n .36 Total Sum of Squares = ∑(x – xbar)2 = ∑ x2 – (∑ x)2 / n = 1.62 1. The t-test is a special case of ANOVA for one or two groups.38 1.98 1.1786 Total Sum of Squares = Between Groups Sum of Squares + Within Groups Sum of Squares 2.41 + ……..63 ∑ 13.41 1.382 + 1.4093 – ((24.22 1.1786 + Within Group SS ∴Within Group SS = 1.1.04 1.25 Group B 1.1267 = 0.77 1.Analysis of Variance (ANOVA) This is a method of analysing the variation due to different groups or factors.22 11.38 + 1.84 1.222 – ((1.84 0.. Group A 1.1.61)2 / 20) = 0.25)2 / 10) + ((11.61)2 / 20) Total Sum of Squares (TSS) = 2. ANOVA can handle more than two groups.22)2 / 20) = 32.03 1.1267 Between Groups Sum of Squares = (∑xa)2 / n ((24.(∑x)a2/ n = ((13.75 1.25 1.22 0.

Note: The √ 0.15 7.65 Using F tables : F 1.1786 1.51)2 / 3 .55)2 / 3 .05) = 4.53)2 / 3 + (21.56 7. The number of groups for each factor does not have to be identical. output from three machines on five different days.53)2 / 9 = 0.0048 Between Drivers SS = (21.1786 0.49 21.17 7.1267 Mean Square 0.g.53 21.41 F (from calculation) < F from tables ∴no difference between the groups.ANOVA Table Source Between groups (na –1) Residual Total (n –1) 1 18 19 df Sum of Squares 0. Time Car 1 2 3 ∑ Driver A 7.13 7.0003 wdp consulting six sigma .19 7. e.6849 – (64.53)2 / 9 = 0.51 64.1082 F 1. The factors here are cars and drivers.0041 Between Cars SS = (21.9481 2.14 21.(64.49)2 / 3 + (21.1082 = 0. As for the one-factor ANOVA we can have several groups of data for each factor.42)2 / 3 + (21.18 (0.42 ∑ B 7.18 21.53)2 / 9 = 0.3289 ………… the sd com from the t-test Two-way ANOVA Here we are considering two factors.18 7.19 21.56)2 / 3 + 21.53 Total sum of squares = ∑ x2 – (∑ x)2 / n = 462.(64.55 C 21.20 7.

0003 0.0 is > 6. Drivers B and C are similar but A is faster than both of them wdp consulting six sigma .4 (0.0001 F 21.776 (√(0.776 critical difference = 2.0021 0.0002 0. Note: S12 / S22 ∼ F distribution Since there is a difference between the drivers we need to compare the means.0 F 2. there is no difference between cars. Driver mean A 7.94 there is a difference between drivers.0048 MS 0.0041 0.0185 ∴B = C > A In words.19 If we simply compared two groups H0 : B – A = 0 H1 : B – A ≠ 0 From the t test B – A ± t .Two-way ANOVA Table Source Drivers Cars Residual Total df 2 2 4 8 SS 0.14 B 7.0001) for any comparison we can calculate the critical difference required for significance.0 is < 6. s( 1/n1 + 1/n2) s comes from the ANOVA table s = √ (0. t from tables df = 4 (na + nb –2) (0. But since the f value of 2.18 C 7.05) = 6.94.05) = 2.94 Since the F value of 21.0 2.0001) (1/3 + 1/3)) = 0.0004 0.

Distance travelled over Time 70 60 50 Distance (miles) 40 30 20 10 0 0 5 10 15 20 25 30 35 40 45 50 55 60 Time (minutes) 60 mph 30mph However. This can be shown as a linear relationship signified by: Y = a + bX Where a is the point at which the line crosses the y-axis (the intercept) and b is the slope of the line. A simple example would be the relationship of distance travelled with time it travelling at a constant speed. (other common forms are . Y = b0 + b1X). Y = mX + c . This fits the line that has the smallest squared distance between the observed values and the fitted line. wdp consulting six sigma .Regression This is a methodology to analyse the relationship between two or more quantitative variables. To do this we use the ‘method of least squares’. Thus if we know a and b. we can solve for any value of X. all data does not give rise to a perfect straight line and we have to ‘fit’ the best line to describe the data.

b. ∑ Yi = n.Y = a + bX X X d5 d4 X X X d3 X 2 d6 d1 d Minimise ∑di2 The calculation of the line of best fit is derived from two ‘Normal’ equations.a + b ∑Xi ∑ Xi Yi = a ∑Xi + b ∑Xi2 By re-arranging these equations we have: Slope b = Sxy / Sx2 And Sxy = ∑XY – (∑X ∑Y / n) and Sx2 = ∑ X2 . X wdp consulting six sigma .((∑X)2 / n) Intercept a = Y .

84 x 30 = 31.25 + 0.6 b = 0.25 + 0.84 a = 29.5 Regression can be taken further by looking at how good the fitted line explains the data and how precise any estimate from the equation is in predicting the Y values.84 x 28.2 a = 6.84 x French score We can estimate the German score if the French score is 30 : German score = 6.8 – 0.8 ∑X = 282 ∑Y = 298 n = 10 / (9312 – ((282 x 282) / 10)) ∑X2 = 9312 ∑Y2 = 10066 ∑XY = 9539 b = (9539 – ((282 x 298) / 10)) b = 1135.Example: Individual 1 2 3 4 5 6 7 8 9 10 Exam scores (%) French (X) German (Y) 10 11 10 22 18 22 25 19 28 35 33 27 34 33 39 40 42 42 43 47 Can the German scores be predicted from the French scores? X = 28. wdp consulting six sigma .4 / 1359.25 Equation: German score = 6.2 Y = 29.

occur at random Any patterns in the plotted residuals indicate that the fitted line is not explaining the data. These values can be plotted to test various assumptions that lie behind regression analysis. are not related to the X values.The deviation of each observed value from the fitted line is known as the ‘residual’.((∑Y)2 / n) SSreg = ∑ (Y – Y )2 where Y ∧ ∧ ∧ = predicted value from equation a + bX SSresidual = ∑ (Y –Y )2 wdp consulting six sigma .Y )2 = ∑Y2 . stable (do not change over time). Assumptions: Residuals are : from a Normal distribution. Source Regression Residual Total df 1 n-2 n-1 Sum of Squares SSreg SSresidual SStotal Mean Square F Where: SStotal = ∑ ( Y. ANOVA The significance of a regression line can be tested via the use of Analysis of Variance.

32 The regression is significant.707 F 31.97 34.632) = 237.05) = 5. German results can be predicted from French scores. 8 (0.35 0.81 0.37 Y .77 33.01 41.97 -1.37 27.65 14.84(X) Predicted 14.65 7.91 F 1.63 SS residual = (-3.65 21.Y )2 = ∑Y2 .6522 SStotal = ∑ ( Y.25 29.((∑Y)2 / n) = 10066 – ((298)2 / 10) = 1185.53 42.948 29.352 + ……….4.25 5.652 + 7.47 4.23 -6.6 Source Regression Residual Total df 1 8 9 Sum of Squares 947.predicted -3. wdp consulting six sigma .Using the previous example X French 10 10 18 25 28 33 34 39 42 43 Y German 11 22 22 19 35 27 33 40 42 47 = 6.652 1185.25 + 0.63 -8.948 237.600 Mean Square 947.99 0.81 39.

8) / √ (9312 – (282)2 / 10) (10066 – (298)2 / 10) r = 1135.8943 x 0.6) (1185.6) r = 0.7997 wdp consulting six sigma . Warning: a high correlation does not mean causation.10 x 28. X. Y ) / √( (∑X2 – ((∑X)2 / n) (∑Y2 – ((∑Y)2 / n) ) From French – German data X = 28.9478 / 1185.Correlation This is related to regression and provides a dimensionless measure of the relationship between two variables.4 / √ (1359. It is also known as Pearson’s Correlation coefficient.6 = 0.8943 = 0.8 ∑X = 282 ∑Y = 298 n = 10 ∑X2 = 9312 ∑Y2 = 10066 ∑XY = 9539 r = (9539 .8943 r2 = SSreg / SStotal r2 = 947.2 x 29.n. The value(r) ranges from –1 to +1 r = -1 : perfect negative relationship r = 0 : no linear relationship r = +1: perfect positive relationship The squared value of r measures the percent of variation in the Y-values that is explained by the linear relationship with the X-values.7996 compare with 0. Mathematically : r = ∑ ( X –X) (Y – Y ) / √(∑ ( X – X)2 ∑ (Y – Y )2) r = (∑XY .2 Y = 29.

04 5.74 7.08 5.50 0.26 10.73 7.25 12.5X And the same correlation coefficient (r) B 1 2 3 4 5 6 7 8 9 10 11 average r X 10 8 13 9 11 14 6 4 12 7 5 9.0 Y = 7.0 Y 8.6 Exercise (Part 2) The following data has the same regression equation as the previous data Y = 3.96 7.84 4.33 9.46 6.74 8.51 n = 11 ∑X2 = 1001.0 D Y 6.56 7.11 7.58 5.0 + 0.50 0.10 9.14 8.74 7.0 C Y 7.5 ∑X = 99 ∑Y = 82.71 8.76 7.10 6.82 5.84 8.77 9.81 8.82 What is different? How do you know? wdp consulting six sigma .42 5.47 7.14 8.91 6. 1 2 3 4 5 6 7 8 9 10 11 X-bar Y-bar X 10 8 13 9 11 14 6 4 12 7 5 9.13 7.5 X = 9.84 6.50 0.0 ∑Y2 = 660.68 7.Exercise (Part 1) Calculate the regression equation and correlation for the following data.26 8.39 8.24 4.81 8.58 8.77 12.82 X 10 8 13 9 11 14 6 4 12 7 5 9.95 7.0 Y 9.173 ∑XY = 797.13 3.04 6.26 4.89 7.15 6.50 5.82 X 8 8 8 8 8 8 8 19 8 8 8 9.

“experienced leaders within Toyota kept telling me that these (lean) tools and techniques were not the key to TPS. Lean Thinking has five main principles: specify value from the point of view of the customer identify the value stream o physical flow o information flow o new product introduction improve value-adding activities eliminate non-value adding activities make value flow (remove obstacles / queues) pull at the customer’s rate of demand (use one : make one) seek perfection through continual improvement - There are two stages to the introduction of lean operation: Stage 1: Prepare the organisation for flow. TPS is essentially about the culture and way of working. Rather the power behind TPS is a company’s management commitment to continuously invest in its people and promote a culture of continuous improvement” (The Toyota Way. like Six Sigma. doing more with less to such an extent that the organisation becomes ‘anorexic’. It is interesting that the emphasis of Lean. However. For example TPS focuses on 3 Ms: Muda (waste) Mura (unevenness – levelling) Muri (overburdening) All Lean teaching emphasises the removal of waste. flexibility. Despite the above teaching from Toyota. In the West Lean Thinking has become synonymous with being ‘lean and mean’ i. but few ever mention the concept of not overburdening people or machines. has been on the teaching of the tools. short lead time and high quality.e. Lean promoters still emphasise the tools.Lean Operation Lean Thinking has developed from the Toyota Production System (TPS). Such action does not fit in with the TPS concept of Muri. some talk of levelling flow. Jeffrey Liker). Areas involved: Demand management (stability) Standards Maintenance (TPM) six sigma wdp consulting .

but work should have been started. capacity. Training) This stage is about preparing for the creation of a lean operation. Some of the Stage 2 techniques may not quite as applicable for low volume manufacturing. control) Inventory management (Batch size. Buffers. It is not the purpose to cover every aspect of lean operations. 5S) Layout (cell concept. They need some of the Stage 1 ideas to be in place. Rewards and Incentives. machines) Improvement (process mapping) Master scheduling (stability. Stage 1 does not need to be completed before starting on Stage 2. Stage 2: Produce with agility and no waste.- Quality (Prevention. Areas involved: Team involvement (actions. opportunities) Visibility and Simplicity (management by walking about) Process data collection (inventory. Flow control) Pull and synchronisation (Kanban) Measurement (wastes and improvement) Stage 2 are the techniques for operating in a lean way. The areas focussed on will be: 5S Waste On-going improvement wdp consulting six sigma . workstations) Change-over time reduction (SMED) Team preparation (Policy.

” Use shadowboards. throw away. Sustain Or CANDO : Cleanup. Straighten. Sweep – physical tidy-up of anything out of place. Neatness. Scrub. Standardise – set standards and audit Self-discipline – participation and improvement The S’s are sometimes written as : Simplify. Five minute clean up each week. Arrange. Discipline. Stabilise. Red tag – label with date – if not used within a time period. Sort – classify everything by frequency by use. Simplify – “a place for everything and everything in its place. Ongoing improvement wdp consulting six sigma .5S This is basic housekeeping.

double handling too fast. people not productively employed. materials. batch sizes. not moving. excess walking. too big. too variable. What Ohno actually meant. which are ‘absolutely essential’” (To add value to the product) Taiichi Ohno defined seven wastes Waste : Muda Examples of waste too early. expediting all material movement. unnecessary inspection. exertion. the workers resented this remark. bending. excessive turns to loosen Rework.Seven Wastes : Muda (Taiichi Ohno) Waste is defined as: “… Anything other than the minimum amount of money. parts. too much. wdp consulting six sigma . control systems Reaching. Non-value added work Those operations or activities that take time and resources but do not add to the value of the product which is sold to the customer. however was “Will you do your value-adding work for a least one hour a day?” So what exactly is meant by ‘value-adding’ work? Value-added Work A primary operation that transforms. buffers. converts or changes a product towards that which is sold to a customer. space and workers’ time. one big machine rather than several smaller Stores. equipment. consequences of not doing the right thing first time Overproducing Waiting Transporting Inappropriate processing Unnecessary Inventory Unnecessary Motions Defects Taiichi Ohno said to the workers “may I ask you to do at least one hour’s worth of work each day?” Believing themselves to have been working hard all day long. just-in-case materials queuing. rejects.

failures. This leads to additional storage and may eventually result in this material being scrapped. Overproduction: extra work put into making more items than are needed. Transporting: this can often be overlooked as we see forklift trucks. conveyor belts as part of the production process. How much of your material in the warehouse has been there for more than 12 months? Will it ever be used? Attitudes that create overproduction are: Produce as many as possible because the cost per unit drops Encourage each line to focus on productivity alone Produce more to increase machine utilisation Such attitudes do not focus on the overall process and usually end up with having to deal with excess production. usually just in-case there is a problem. The question should always be can this transportation be eliminated. This waste can also be seen when someone is simply monitoring a machine that is doing the value adding work itself. how much of your time is spent moving around or waiting for something to happen or arrive. At each step of the process value should be added to the product. in many processes much that we do does not add value to the product. For example a small circuit board arrives in one company inside a plastic bag separated in the box by dividers. Inappropriate processing: this can be seen in a number of areas such as running machines knowing that they will produce rejects or simply adding additional steps in the process that add no value. How much of the work that you do adds value to the process or the product? You may even be employed in a department responsible for putting things right. unload and logistical problems. In a chemical company the output from an extrusion process was transported via a tractor from one building to another for further processing. Often when someone follows a product or piece of paper around a factory they are astonished at the distance it travels. Waiting: in a production environment this is when the operator can do no valueadding work because they are waiting for additional material or a machine breakdown to be repaired. information or service and then passed onto the next step. We can often fill the time with other things and therefore do not notice how much time is spent waiting. In a administration this can be waiting for information or waiting for a decision. However. This involved delays time to load. However the need for such transporting is due to a process being built around existing structures. Think.If the ‘value-added’ definition is applied to any process then it will be discovered that most of what goes on within the process is in fact non-value adding. rejects. Other boards arrive wdp consulting six sigma .

It is here that significant improvements can be made. rework. wdp consulting six sigma . semi-finished products and finished products do not add value.simply separated by dividers without problems. Taiichi Ohno used the term ‘rejects’ for one of his wastes. It is important to identify how much waiting time is involved how far things are moved.g. requiring a method and people to record and manage the storage. Rather than fix a process we stock more. sort. Eighth Waste In addition an eighth waste has been added to the original seven. This has in some companies been reworded as ‘correction’. How far does an operator have to walk to get a tool or how far does an administrator have to walk to get to a filing cabinet or a phone? This movement delivers no added value to the process. High inventory will often hide problems within a process. This is probably the largest waste of all. Unnecessary Motion: this aimed primarily at an individual. Elimination of waste: many of the areas of waste appear to be simply part of the way we do business and few people think about removing them because they are not aware of the cost. They have taken resources to produce and now occupy space. Waste elimination and good housekeeping (5 S’s) go hand in hand. Defects do not only occur in products. It is estimated that in manufacturing 15-20% of turnover is wasted but for Service industries this can be as high as 40 – 50% of turnover. In one company it was prepared to live with a defect rate of 15% and employing 10 people to repair the rejects rather than fix the process. or return to the supplier of the boards. Often when a process is mapped these wastes do not appear. This is the waste of not fully utilising human resources. the additional cost of dealing with the defect e. Unnecessary Inventory: raw materials. how many errors occur on an invoice leading to delayed payment and additional handling. The use of the bag adds steps into the process that appear to add nothing and leaves a company with lots of small bags to dispose of. scrap. i. Defects: rejects interrupt production and require resources for either rework or disposal.e.

Improvement Make improvement a way of life. Books such as ‘The Toyota Way’ by Jeffrey Liker will give more information.” (The Toyota Way. Be aware of the seven/eight wastes Question why things are done Kaizen (on-going improvement) Mapping (identify problems / opportunities) Materials – reduce inventory Methods – go to gemba (work place) to collect data set standards and audit (adopt best practice not police) Summary Creating a Lean Operation requires more than simply introducing 5S or Kaizen. Liker) wdp consulting six sigma . J. At the centre of TPS is people. “these are just technical tools and they can be effective only with the right management and philosophy – the basic way of thinking.

E.Leading Teams Part 1 – Self Team leader: Competencies and Behaviours Self-Assessment Questionnaire The Team Leadership Role The Facilitator’s Role Presentation Skills Part 2 – Individual Giving and Receiving Constructive Feedback The Intervention Descriptive versus Evaluative Feedback Handling Feelings and Conflict Part 3 – Group P.R. Team Performance Review Team Behaviours wdp consulting six sigma .P.

Individual. the group and the individual. As a team leader. wdp consulting six sigma . notably Self. Group and Task. that is down to an individual’s style. It should be used as a reference point and as a guide.Introduction These notes have been compiled to give the reader an introduction into team leader and facilitation skills. The balls they have in the air relate to the task. and will be catching the balls and throwing them up in the air again so they do not fall in the ground. What it does do is give some key pointers under the main aspects. It cannot tell you exactly how to lead a team. you will make various interventions to the group you are facilitating to ensure the meeting is a success. A good juggler will have an awareness of their own ability and the mood they are in when juggling. Team Leadership is like juggling. At all times the juggler must be aware of where the balls are. as they juggler hopes his / her juggling act will be a success.

SELF Team leader Competencies And Behaviours Question No Individual Score Total Score Average Score 7 20 31 14 Spot opportunities for individual or departmental improvement Turn good ideas into action Try to change things rather than just live with them Take prompt action when opportunities arise Initiative Average 26 1 41 19 Change current thinking for the “way we've always done it” Adapt to new ways of doing things Engage in idea generating or "what if" discussions Generate innovative ways to deal with situations Creativity Average 30 3 40 22 Exercise confidentiality when dealing with sensitive issues Deal with people consistently Keep promises once they are made Inspire the trust of others Personal Integrity Average wdp consulting six sigma .

Question No Individual Score Total Score Average Score 2 42 29 32 'Reads' and interprets people's behaviour accurately Has a flexible style in dealings with others Appear to be approachable and easy to talk with Ask questions to learn more about (or to clarify) what you are saying Give praise / recognition where appropriate 17 Skills With People Average 12 21 10 39 Identify the process being used to deal with a situation Focus on the key issues in a mass of detail Think of practical ways of implementing ideas Demonstrate an ability to think on your feet Process and Logical Thinking Average 4 36 27 6 38 Describe what took place when giving feedback Remain uninvolved when observing situations Make insightful comments on behaviour in a group situation Identify trends and patterns in situations Actively listen and note what others say Observation Average wdp consulting six sigma .

Question No Individual Score Total Score Average Score 13 33 5 Keep appointments and arrive at meetings on time Plan meetings in advance as appropriate Organise activities and timescales to achieve objectives on schedule Establish contingency actions in case things go wrong 23 Planning and Organising Average 9 37 25 16 Choose an appropriate time to interrupt a meeting Keep the overall picture in mind when looking at details Show a willingness to change their mind Assess the possible consequences of decisions Judgement and Decision Making Average 28 11 35 15 Make effective formal training presentations Express ideas or information clearly when speaking Presents a good first impression Answer questions specifically and to the point Presentation 18 Act as an effective facilitator between people who have different points of view Persuade people to change their minds Anticipate objections or counter-arguments and have solutions ready Influence people not under your direct control 34 24 8 Influencing Average wdp consulting six sigma .

However. These behaviours are shown as a set of competencies that fall into certain categories. The second is the table. Both of these are shown below. and the following model is designed so that team leaders can assess what they are strong at and which skills require some development. Each competency is defined in the table and two other columns are also shown. people who possess these naturally are likely to be those who emerge as leaders of the group. The first of these highlights the typical behaviours that one would expect the leader to exhibit. The last column shows what the outcome or impact of that particular behaviour should be on the team. and sometimes the leader will emerge as the team begins to work together. Good leaders will have a certain set of behaviours. which shows everything in more detail. Sometimes these are elected or pre-chosen. • • • • initiative and risk taking vision planning and goal setting creativity / innovation • personal integrity • performance under pressure • diversity Looking Ahead Personal Values The Team Leadership Model Managing People • • • • • • • Managing Outputs • quality • problem solving and decision making empowerment communicating delegating motivating coaching team building mentoring wdp consulting six sigma . The model is split into two parts. without the formality of being chosen. most people will need to develop their team leadership skills.The Team Leadership Model All teams have leaders. showing them in 4 distinct categories. The first summarises the competencies involved.

displaying energy. Team sees leader as its champion. People are committed to continuous improvement. making things happen and continuously looking forward. Team takes a pride in its output. and their own part in the scheme of things. People understand why the team exists. Personal Integrity The ability to gain the trust and confidence of others by acting fairly and honestly. sharing information openly. demonstrating a consistent personal commitment to quality. Outcome/Impact On Team Team has a sense of urgency. standing up for what they believe in. Taking a step back from day to day activities.Competency Initiative And Risk Taking Definition The ability to demonstrate proactive thinking. Team never happy with the status quo. inspiring others. Team has confidence that the leader will deliver on promises. Team members feel they have encouragement to take measured risks and initiatives. persevering. operating in an evidence based manner. People feel that they can speak up without fear of reprisal. Quality The commitment to produce high quality results via high quality processes. performing work that exceeds expectations. acting in a way that is consistent with the vision. Team feels a sense of purpose. describing and discussing the purpose and strategies of the team. dealing with people fairly and consistently. but within the context of the wide picture. showing a willingness to take the lead. Team are comfortable in their working environment. people are encouraged to regularly improving their performance. representing their team to the organisation and the organisation to their team. Keeping their promises. drive and accept responsibility for their actions. Members feel that they will be supported and the team defended where necessary. balancing attention to results with attention to work and interpersonal processes. taking personal responsibility. The team has its own identity. Producing what is required first time round. keeping people focused on the future rather than the past. where the team is going. reviewing the purpose periodically. showing a willingness to take risks. Vision The ability to create and describe an ideal state or condition and focus people’s attention on it. people are encouraged to view mistakes and problems as learning experiences. dealing with sensitive issues in a confidential way. displaying charismatic behaviours which support the vision. and have a sense of equality. . Typical Behaviours Spotting and acting on opportunities.

using delegation as a development tool. Team communication with others is clear. involving people on key projects. providing a sense of belonging to the team. assigning responsibilities that match individual capabilities. asking pertinent questions. Setting an appropriate example. People feel that they can approach team colleagues for help. Everybody understands the same thing. allowing people to finish what they have to say. answering questions specifically and to the point. Typical Behaviours Allowing people to use their judgement when making decisions.Competency Empowerment Definition The willingness to create a work environment in which people are encouraged to develop their full potential. team members are prepared to admit their mistakes. Delegating The ability to identify and assign work. Team members feel committed to individual and team performance. encouraging team members to learn from each other. Giving constructive performance feedback to individuals or the team. Motivating The ability to create a satisfying work environment that encourages others towards achieving team goals. people are willing to take responsibility for their actions and are given authority for actions. Tasks are perceived to be allocated fairly. Misunderstandings about issues are uncommon. Communicating The ability to express oneself clearly and to listen to others. aware of others feelings and concerns. and defines how individual performance will be measured. Team members receive interesting and challenging work to do. ensuring people receive training for tasks. people feel that they have the freedom to develop and express themselves through their work. Expressing ideas clearly. Team as well as individual performance is known and understood. People have a positive attitude towards development. The ability to provide feedback and offer support when people are confronted with performance problems and/or development requirements. Explaining new assignments clearly and patiently. Coaching . showing an interest in what others are saying. presenting information in a well-organised way. They understand what is expected of them and how they will be measured. clarifies expectations. people enjoy working in the team. Team members understand what is communicated to them. They feel listened to. praising individuals or the team for good work. performance problems are dealt with quickly and effectively. creating a learning organisation. Skill levels and coverage are raised within the team. allowing people to initiate actions. letting people know what is expected of them. there is an identifiable team spirit. Team output is high. recognising contributions to team performance. taking time to develop people’s effectiveness. Outcome/Impact On Team Team members feel that they have worthwhile work to do and worthwhile responsibility. looking for opportunities to create learning.

decisions are made in a timely fashion based on facts as well as opinions. contributions from individuals. involving team members in problem solving. interactive processes and team focus. listening to the ideas of others even when these do not coincide with their own. Plans are changed as a result of environmental influences and tasks completed on schedule. measurable objectives. harmonious group working. Many ideas are generated within the team. setting clear. Problems are shared amongst the whole team. not just the symptoms. Many ideas are adopted. The team functions effectively as a unit. Customers feel able to give feedback that will be listened to and result in action. Creativity And Innovation The ability to develop and apply new and innovative ideas and practices. There is an appropriate balance between team and individual decision-making. and make appropriate decisions. dealing with the cause of problems. using time effectively. Making sense of complex situations. Typical Behaviours Developing realistic plans and timetables. People feel valued for their creativity. The team is involved in the planning process. making sure that good ideas are shared within the team. identifying problems at their early stages. adapting to new ways of doing things. engaging in idea generating or ‘what if’ discussions within the team. The team creates a disciplined environment that does not inhibit the freedom of its members. The team is seen as adding value to the organisation. Encouraging team members to be creative in their work. Team Building The ability to ensure that the value and effectiveness of the team is greater than the sum its individual members. Difficult decisions are not put off. Objectives and schedules are clearly communicated. Team members share ideas with each other. Team members feel that they have been involved in solutions. and set realistic. . Problem Solving And Decision Making The ability to analyse a situation. identify root causes and alternative solutions. establishing priorities. encouraging democratic decision making.Competency Planning And Goal Setting Definition The ability to plan and organise work in a structured and visible manner. The priorities of the team are in line with those of the organisation. Paying attention to the team’s internal processes such as purpose and commitment. ensuring that the team and its members enjoy excellent relationships with other teams and with its customers. demonstrating a willingness to take tough decisions. measurable objectives for individuals and the team. Outcome/Impact On Team There is congruence between individual and team goals. Team members feel that the team is good to belong to.

demonstrating strength of character in adversity. The team are compatible with and able to encompass and synergise the many different ideas and approaches of its members. increasing levels of responsibilities to prepare people for a higher level position. bringing together people with varied talents and perspectives to resolve work group problems. Team members feel that positive thought is being given to their future and feel able to request only support needed. Mentoring The willingness to support others in their personal and professional development.Competency Performance Under Pressure Definition The ability to maintain high levels of performance and personal conduct in challenging or unpredictable situations. Team members work well with each other and with the team leader. Team can take changes on board. Team members develop during the life of the team and receive feedback on their performance at regular intervals. Diversity The willingness to work with individuals and integrate the differences that exist among others. acting with professional detachment and objectivity. being able to adapt to changing circumstances. Making the effort to deal with others in a cooperative fashion. providing suggestions for personal and professional development. including those in other teams. demonstrating a willingness to be flexible and open minded. negotiating differences of opinion openly and fairly. Team is aware of steadiness of purpose. Identifying career and development opportunities for team members. Outcome/Impact On Team Team members feel that external pressures are being managed. Typical Behaviours Remaining calm when confronted with problems or crises. Team’s output is not unduly affected to external pressure. . sharing relevant personal insights or experiences.

The message may have been delivered but the relationship or contact between the team members has not been enhanced. we should take this further” may sound positive and supportive. but from time to time. but the speaker has not actually made any personal contact with Joe. The use of someone’s name suggests that the speaker is focused on the person spoken to (rather than rambling in his or her own head) and therefore can help to establish contact. A statement like: “I thought Joe’s proposal was excellent.Presentation Skills Structure Opening impact Tell them what you are going to tell them Main body of presentation Tell them what you have told them Close Techniques Start with the listener Appearance Eye contact Use examples Distracting mannerisms Visual aids Presenting yourself Talk from experience When a team member illustrates a proposal with examples of someone else doing something else somewhere else. Speak to. but watch for it as this frequently happens. Address the person by name Not always. attention becomes short. not about people present It may seem inconceivable that team member A will speak about team member B or something that team member B has done by addressing everyone else but team member A. When the same team member gives a personal example. wdp consulting six sigma . eyes swivel back and attention is complete. something that happened to them.

While recognition and acceptance of differences is fundamental for individuals to make contact. In such cases contact will be maintained if the speaker catches the eye of the person spoken to before and after launching into an exposition. The statement will always clarify the question and make a response much simpler. If your interpretation is correct some other team members will usually be experiencing the same feeling. Flush hidden positions out with the question: “Is there a statement behind your question?” Trace opinions back to observations When giving feedback to a team member on performance. Describe. Feelings result from our interpretation of what we see happening. ‘people’. Only specific instances. Confusion. if not lack of interest. Feelings contribute Even apparently negative feelings can help the team become more aware of what it is doing. frustrating everyone except the individual who is holding back his or her ‘ace’. Differences establish how people are separate from each other. etc. Make statements before questions Behind every question lies a statement. members who take responsibility for their own opinions make strong contact with the rest of the team. difference should not be used as a means to compare and judge the relative world of others. Team members who habitually talk whilst looking out of the window may discover they are not heard very well. ‘one’. etc. often results from the use of ‘we’. ‘you’. ‘you’. Say ‘I’ not ‘we’. wdp consulting six sigma . described as if by a video camera. Recognition of this can instigate an immediate positive change of focus. When a team member repeatedly challenges an idea or asks question after question. don’t judge Judgement implies evaluation.Look at the person you are talking to Many people attempting to express a complicated idea will look up to the ceiling or out of the window. Evaluation involves comparison. Again. Opinions are often then disguised as established facts. Comparison established difference. can avoid an argument or further discussion. This process often prolongs unnecessary discussion. it is frequently because they already have an opinion and are searching for information that will confirm their opinion.

wdp consulting six sigma . is followed by an opinion. we create an environment in which the content of our communication is given the meaning we intend by the context we create. Summary If we frame information clearly and speak appropriately to our team-mates and colleagues. is followed by a statement. Saying ‘I would like’ is clear and can drastically shorten meeting time.Say ‘I would like’ rather than ‘You should’ or ‘We should’ ‘I would like’. The circular causality of content and context in communication should be apparent: good communication fosters good contact amongst colleagues and vice versa. ‘You should’. Right relationships are enhanced. If we foster good contact amongst team members and in our relationships with others. which will frequently lead to a lengthy argument. we understand each other better and generate more genuine contact.

EXAMPLE 1: EXAMPLE 2: Judgement. Descriptive. Be descriptive not judgmental Describe behaviour not personality Be constructive not destructive Own your own feedback 5. it is rude. so that each person’s development is supported. Be timely Be specific Give your feedback as soon as possible after the event. 4. not for others.INDIVIDUAL Giving And Receiving Constructive Feedback For most of us. Why are you giving the feedback? Is it to make yourself feel better or to help the receiver? Speak for yourself. 3. Guidelines for Giving Constructive Feedback 1. speaking for self. Describe specific behaviours and reactions. speaking for others. 6. not what you think they are. in an appropriate setting. You will have many opportunities to give and receive feedback from others. Behaviour is easier to change than personality. do not make assumptions about why Focus on what someone does. 2. and it got everyone upset”. It is important to make your exchange constructive. personality focused: “You should not be so aggressive. particularly choose those they should keep and those they should change. Describe what you see. wdp consulting six sigma . I felt as though there was no point in explaining my idea”. feedback is an essential ingredient of learning and developing. behaviour focused: “When you interrupted me several times. hear and feel.

Do not defend or justify your behaviour. and explain the kind of feedback that will be especially helpful to you. or how they see it. quietly discard the rest. Use what is helpful. Tell people what you are trying to find out. Thank you for telling me. Listen to what people have to say. 3. Accept it EXAMPLE 1: Defensive. React positively 4. Ask them to be specific and descriptive. Ask for it Direct it Be selective. not listening: “No. without having to apologise. you are completely wrong there. accepting: “Yes. I did it like that because I had to and I know I am right. perhaps I need to explain it in more detail”. I can see from your point of view that it might look like that.Guidelines for Receiving Constructive Feedback 1. EXAMPLE 2: wdp consulting six sigma . find people whose views you will value. Accept criticism by acknowledging that there may be some truth in what they say. Positive. ask for it. and I am not prepared to listen to anything else”. 2. If you need information.

the facilitator might try to rectify matters by supplying the missing behaviours himself. and Reacting. Typical Problems Caused By Dominant or Missing Behaviours When a meeting is unusually High on INITIATING • • • Characteristic Problems are too many ideas and alternatives to handle lack of attention to detail “up in the clouds” feeling Low on INITIATING • • • • • • meeting becomes backward-looking lack of enthusiasm and excitement undue attention to detailed analysis meeting becomes emotional misunderstandings become more frequent people take sides and issues become entrenched tendency for repetition people withhold important information meeting is awkward and forced very time-consuming obsession with minor issues “swimming in syrup” impression meeting becomes disorganised hasty decisions are made people cannot agree afterwards on what has been decided High on REACTING Low on REACTING • • • • • • • • • High on CLARIFYING Low on CLARIFYING For example. You should be constantly aware of the balance of meeting behaviours between Initiating.e. Balance. then the meeting will become ineffective. in a meeting that was low on clarifying. For guidance on how to do this read again the notes on Descriptive Feedback and Ways of Speaking. interrupt) the meeting when he/she has something important to which to draw the group’s attention. WHEN TO INTERVENE The facilitator should intervene (i.Intervention 1. If one of these is lacking for any length of time. Thus he/she might spend time on seeking information bringing out the details of proposals under discussion and encouraging group members to follow his example wdp consulting six sigma . Clarifying.

Jean. He/she could also bring in other group members encouraging them to give information on the reasons for their initiating and reacting behaviour. this subject is one in which you have some experience.He/she could also test understanding and summarise to make sure that these important activities are not neglected and the correct degree of understanding is achieved. DEALING WITH CONFLICT AND DISAGREEMENT Encourage people to concentrate on the arguments rather than personalities. Jim. Test for consensus. What do you think about it? The important thing is to give people the OPPORTUNITY to speak. He/she could shut out group members who attempted to overload the meeting with further initiating and reacting behaviour. 3. 2. Angela? Tom. I think you should get people’s reaction to what you have just said. although whether they take it is up to them. wdp consulting six sigma . perhaps you might like to ask the others what they think. We can always come back to it later when people have had a chance to digest it. Here are some examples:Before you continue. Use expressions like: What do you think. Harry? You seem to have gone a bit quiet lately. ask each team member to give their view individually. BRINGING IN AND SHUTTING OUT Bringing in involves identifying those members who are low contributors (either because they are naturally so or because they are being shut out by high initiators). If the disagreement is between two people check to see what the other team members think. Shutting out is basically an interruption and needs to be done very carefully. Encourage people to look for the good points even if they disagree with other parts of the proposal (“3 likes and a wish”). Are you sure people understand what you have been saying? Let’s leave that for now and move on. Shutting out can be done deliberately by the Facilitator in order to give others a chance to speak. is there something you want to say? Do you agree with this.

Descriptive Versus Evaluative Feedback Descriptive feedback describes what you see or hear the other person is doing or saying.. “BEHIND THE SCENES” Sometimes the meeting is not the most appropriate setting for the facilitator to carry out his/her full role. Descriptive feedback can also be used to describe your internal feelings: thus “I feel happy when you do that” wdp consulting six sigma . Descriptive feedback is more likely to result in change because: it is more impartial it is not emotional it is difficult to argue with. or is over-contributing... you’ve been talking non-stop for the last three minutes” are both forms of descriptive feedback.. 4. Based on evidence ‘I imagine this is . Try to make the protagonists see that merely re-stating their positions will convince nobody. Thus “I notice that you look out of the window when I’m talking to you” or “Bill. and can act as a “go-between” between individuals or sub-groups. “I think your performance is bad” and “Your man management skills are weak” are both forms of evaluative feedback.. Some things are best dealt with on an individual or more private basis before or after the meeting. involves giving a judgement on what you are seeing or hearing. Evaluative feedback is more likely to produce a negative or defensive reaction because the person themselves feel criticised. this can be pointed out to them in private rather than in front of the whole group.’ Evaluative feedback... For example. on the other hand. or during a break. The facilitator can also use this time to talk to members privately about their performances as team members..“Park” the item for tabling later after a cooling off period. The facilitator can use these occasions to take soundings about what people feel about the meeting. if someone is not contributing.

Which of them are descriptive and which are evaluative? How might you convert the evaluative ones into more descriptive language? Your presentation skills are poor Your presentation skills need developing In your speech I did not notice any mention of costs I feel uncomfortable when you discuss my performance as a facilitator I feel that you should look at Peter when you talk to him I notice that you always disagree with what Tom says and never with what Harry says Your methods of man management are very good Your methods of people management need improving I like the way you get to the heart of a problem People come to you for advice You are very reliable wdp consulting six sigma .Exercise on Descriptive Feedback Look at the list of statements below.

Whilst sometimes uncomfortable. 2.Handling Feelings and Conflict Handling feelings and conflict is an important part of a team leader’s role. Encourage people to express their wants. These skills often involve separating:issues from solutions facts from opinions feelings from logical reasoning. communicating that you feel issues are present and affecting the performance of the group. Search for and evaluate alternative solutions . Face up to the issues and acknowledge that these exist. The solutions generated may happen in a variety of ways that are as follows: wdp consulting six sigma . 5. A framework for tackling difficult situations can be helpful. Ensure there is common understanding of the different positions. This is shown below as a series of steps. encourages creativity and interest. It can provide a medium for problems to be shared and tensions released.ensure everyone knows who will do what by when. Gain agreement on the best solution . 3. challenging the status quo often causes difficulties and skills are needed to deal with the situations that arise. 1. and reduced group effectiveness.maximise collaboration on common ground and separate creating alternatives from evaluating them. wishes and preferences. particularly if people find dealing with feelings and conflict unpleasant. 4. Define the issues. trying to reach a mutually acceptable definition of the problem. There can of course also be destructive outcomes of conflict such as discontent. Encourage people to be open. Avoid a combative approach and continually reiterating the positives of each position. and restate the positions expressed. diminished communication. conflict can be positive when it improves the quality of decisions.

All of these blocks can be reduced by the use of facilitation skills.a party may recognise that a conflict exists but reacts by ignoring it. guarded or aggressive.people's frustrations very evident In addition blocks to resolving conflict situations may also stem directly from people asking poorly thought through questions which can lead to greater anxiety. being too ready to criticise Misuse of language . However much you desire to handle feelings and conflict.• Competition (assertive and uncooperative) . or ‘because I could not find you and had to deal with a problem which was your responsibility. • Collaboration (assertive and co-operative) . • Accommodation (unassertive and co-operative) . withdrawing or suppressing the conflict.not listening to others point of view. • Avoidance (unassertive and uncooperative) .people failing to acknowledge differences Excessive talking . and want to work towards a mutually beneficial outcome. inevitably blocks to this will arise such as:Wrong language .combative rather than collaborative Not confronting .‘people .when each party has to give up something important to them. wdp consulting six sigma . For example if your boss asks you ‘where were you yesterday?’ it could either be followed by: ‘because I wanted to invite you to lunch’. high level of emotion. without regard to the impact on others in the process. and repetition of views Poor anger control .when each of the parties in the conflict wants to satisfy fully the concern of all parties. and therefore the outcome is compromised. rather than reducing it.when one party seeks to achieve certain goals.blaming’ rather than ‘I’ statements.when parties seek to appease their opponents so much that they may be willing to place their opponents’ interests above their own. Questions thus spring from a variety of directions or motives and if those on the receiving end are unsure about these they may be defensive. • Compromise (midway on both assertiveness and co-cooperativeness) .

R.E. How many meetings have you been to where one of the following has happened? • Some people do not know why the meeting is taking place.P. PLACE RELATIONSHIPS END RESULT PLAN P. R. E. P.P. or have different views • Some people do not know each other wdp consulting six sigma .

R. The process outlined below is a means of ensuring that important matters are attended to before the agenda starts in earnest.P.R.• Attendees do not know why they are there • There are not enough chairs to go round • The room is too dark • People are unsure about the agenda • The meeting never really seems to ‘get going’ • Little. The process is called P.P and should be carried out at the beginning of every meeting. appropriate? Will everyone be able to hear what is going on? Is the venue ok? Or should an alternative be found next time? Relationships wdp consulting six sigma .E. the shorter the meeting the less time is needed for P. It should be conducted by the Leader of the meeting. This will depend on the mood of the meeting. Usually. A little time invested at the start of the meeting can pay big dividends when it comes to achieving meeting effectiveness.E. or more subtly by simply ensuring all four steps are covered. may be done very overtly by saying that ‘P.P. Most people have experienced some of these problems.R. but it should always be done. Is the room appropriate? Is the arrangement of furniture as it should be? If visual aids are to be used can everybody see them? Are the temperature. yet they are quite easy to avoid.E.P. Place Pay some attention to the physical environment in which the meeting is taking place. if anything.. lighting etc. They are allowed to happen by everyone getting immediately involved with the content of the meeting or their own agenda. Introducing P.’ will be used. and all members should be actively involved. is achieved.E.R.

Consider the meeting attendees. Does everyone know everyone else? Has this group worked together before or are they new to the meeting and each other? Are there any special roles involved? For example. Where specifically do you want to be at the end of the allotted meeting time? What should have been achieved? Plan / Process This involves agreeing a means of getting the team to the end results specified. When did you last meet? What has happened in between? How do people feel about what is going on? How is the team relating with people external to it? End Result Everybody must be clear about the purpose of the meeting. there is less of a need for introductions. To review performance the team needs to break away from what it is doing and undertake the following three key stages:1. whereas the opposite may be true. a Time Manager or a Recorder? Does everyone know what is expected of them? For teams that have been meeting for a while. with limited emphasis placed upon opinions. If the group is to meet several times. and requires time for the team members to review what is happening within a team and assess the performance being achieved. conversely. For example. Team Performance Review All teams go through a process of development whereby they move through four stages from forming to storming to norming and then performing. telling someone they are dominating is more likely to cause resentment and ill-feeling. As far as possible this should be based upon facts. it becomes easier to establish different norms of behaviour. This process does not happen automatically. which should have been sent out beforehand. who is the Chair. By stating the fact that other participants of the meeting observed the person talking more than others.e. but often a need to establish outstanding issues. if someone dominates the discussion. who should ask if everyone understands it and thinks it is feasible. should there be a Facilitator. a Scribe. outside themselves. the purpose of each specific occasion needs to be clarified at the beginning of each meeting.i. should be clarified by the Team leader. Time slots should be allocated. Each individual should spend time reflecting on what is going on within the team . There is a series of issues that could be thought about:wdp consulting six sigma . including time for warm up and warm down. although these do of course have their place. others perceptions and opinions may be that the person does not want to listen to what others have to say. The agenda.

although recognising the value this can have. Each individual should then assess how they feel about the meeting in terms of their own level of comfort and emotion i.the place and environment 2.Achievements and Assessment of processes used Roles and relationships Other people .attitudes.for example if the team recognises that it consistently generates a lot of ideas but does not actually manage to do anything with them.e. actions and body language Making decisions Ambience . assess what is working well in the meetings and what could improve. wdp consulting six sigma . Each time the team meets they should focus on one thing about their behaviour they are going to develop. and will help the team and the individuals within it get a greater sense of satisfaction. By going through this process the team will develop closer and more productive working relationships more quickly. Based on assessments made by the individuals. As far as possible the developments should be based on facts and descriptions. then some change of process is necessary. 3. with limited emphasis placed on perception and gut reaction. Once this has happened the team should look for any patterns that may emerge . inside themselves. the team should then share some of their thoughts.

Team Performance Review es & Rol nships atio Rel R O Other People Ac h Aw iev ar eme Pr ene nts oc ss & es of s M De akin cis g ion s M e nc bie Am A A Outside Inside of Self of Self How well is the team working? What can we change? wdp consulting six sigma .

Reacting Clarifying Clarifying This behaviour involves testing understanding (establishing whether an earlier statement was understood). opinions or clarification from others) and giving information (offering clarification to others). To ensure all members of a team understand a new concept. and clarifying. and inputting new ideas. and defending/attacking (attacking another person or behaving defensively). building on or modifying someone else’s proposal. summarising (re-stating previous discussion in a more succinct form) seeking information (seeking facts. the facilitator should ensure that all three behaviours are brought out. Initiating. Reacting. wdp consulting six sigma .Team Behaviours When teams are working together there are three types of behaviour that may be observed. disagreeing (stating disagreement or raising obstacles or counter arguments). especially when ideas are being brainstormed and developed. reacting. Initiating Reacting This behaviour involves supporting (either verbally or non-verbally giving support for another person or their concepts). Clarifying Initiating This behaviour involves making suggestions or proposals. These are initiating.

GROUP PERFORMANCE By Simon Haben wdp consulting six sigma .

’ Once a group has formed. However.. They are permanent or temporary.INTRODUCTION & DEFINITIONS In this document I explain some of the theory relating to groups and performance and then follow this by my own experiences.a work group.They have a common objective. carrying out. both in terms of quantity and quality. Before examining the factors affecting the performance of a group it is worth defining the two key words. group and performance. doing. an awareness of group identity and 'boundary'. wdp consulting six sigma . or the best way of putting them into practice. the cabal.the ad hoc meeting or discussion. They can be a most effective device for blocking and obstructing new ideas.to the informal .. outlining improvements that could be made. the clique. The first part of this document defines groups and performance." The definition of `performance' according to the Oxford Dictionary is `the execution. performing of play. the third examines the influences on group performance. the second illustrates different types of groups and their uses.. it is not too hard to expect some level of performance according to that definition. Performance is something that can be measured tangibly. According to Mullins (1989) "a group consists of a number of people who have: a common objective or task.. They are liked by their members or regarded as a waste of time. a project team." This definition of a group is rather a 'clinical' one and I feel Handy (1993) expresses the definition of a group more practically:"Groups vary from the formal ... what I am explaining in this paper is how effective the performance of a group is. notable feat. a committee. etc. a board . a minimum set of agreed values and norms. which regulates their relatively mutual interaction. the luncheon group.

Formal groups are created to achieve specific organisational objectives and concerned with the co-ordination of work activities. not necessarily related to the task being undertaken. There are two different types of groups: formal and informal. People are brought together on the basis of defined roles and the structure of the organisation. The nature of the task to be undertaken is an important feature of the formal group with goals being identified by management and certain ground rules. Both may be different to that of the organisation from which it originates.GROUPS Groups are an essential feature of the work pattern of almost any organisation. It is very unusual for an informal group to be exactly the same as a formal group. wdp consulting six sigma . The membership and roles of informal groups can cut across formal structure. Once in a group. These groups tend to satisfy psychological and social needs. The activities of the group can also be closely associated with the process and style of leadership adopted by the leader / facilitator of the group. relationships and perhaps norms of behaviour established. Group pressures can have a major influence on the behaviour of individual members within that group and their performance to the group's overall aims. Understanding the informal group network can be criticised in determining how to facilitate formal groups. the people within it influence each other in many different ways. The group may even develop its own hierarchy and its own leaders. These groups tend to devise ways of attempting to satisfy member’s affiliation and other social motivations that are perhaps lacking in the work situation. Informal groups are based more on personal relationships and agreement of group members than on defined role relationships.

This is influenced by:(1) (2) the givens i. the leader / facilitator of the group should try to maximise performance by encouraging people to adopt a role slightly different to the one they would normally play.e.e. If a group becomes too large it may split in two sub-groups with friction developing between them. However. As group size increases. Belbin (1981) has done some study into this and defines a variety of roles within a group. The more homogenous the group in terms of background. the finisher will check details and deadlines. the easier it is to promote cohesiveness. When setting up a group it is not often possible or practical to test prospective members to find out which role they are likely to play. wdp consulting six sigma . and under perform. they can also cause difficulties and conflict may arise when members are in competition with each other. the intervening factors i.e. In my experience groups with more than 7 or 8 permanent members starts to split into sub groups. an understanding of Belbin's roles and people’s background is useful. Where a group has been pre-selected. Differences in personality or skills in the group may actually complement each other and help the overall performance. the task. productivity and member satisfaction. However. problems arise with communication and co-ordination. attitudes and values. the group. An assessment can then be made of what different members the group is likely to need (ensuring a good mix thereby overcoming any problems of 'group think') and of people most likely to meet those characteristics when recruited to the group. the plant will be introverted but have many ideas. The individual roles in a group can therefore be important.GROUP PERFORMANCE The performance of a group is essentially its effectiveness. leadership / facilitation style. where people are so in tune with what the others are thinking they do not always see the broader picture and may overlook something incredibly obvious. Below I have examined the givens and the intervening factors that lead to the outcome. processes and procedures. Compatibility of members of a group is also important. For example the Shaper will be the task leader and is needed to spur action. interests. I believe that the optimum performance of group size is of five to seven people. this is often not the case. While this may be beneficial if both groups are focusing on different aspects of the task. leading to:(3) the outcomes i. The Givens The size of the group is an important factor in its performance. the environment. If groups are too cohesive. they can suffer the phenomenon of `group think'. motivation.

hidden agendas can be overcome by careful allocation of task. no focus. and the outcome was poor. It is my experience that this can be a painful process. This enables greater demands to be placed upon the group. The timescale set should be realistic so that good performance can be achieved. I once worked with a group that had no plan of what it was trying to achieve. impressing the boss. Performance will be improved the task is made clear and a plan formulated with the groups involvement. although very senior in the organisation. Physical barriers can hinder cohesiveness and often act as communication barriers. performance in terms of quantity may be a lot higher with group members motivating each other. or preferably encouraging members to admit their own issues to the group. but not too long as procrastination may occur. It is important for the group leader / facilitator to reveal hidden agendas. The type of task will naturally affect the kind of group that is formed. On a project I was working on. For as long as hidden agendas exist. Hidden agendas may be protecting the interest of one sub-group. or even covering up past errors. the more important it is for the group to work well together and the need for consensus on a focal person or leader.It has been found that the more complex the task is. Hidden agendas can give rise to conflict. In my experience. This can result in work of a lower quality than if the timescales were broadened. group performance is likely to be hindered. Good performance therefore requires that a leader / facilitator should be picked because of ability to direct and influence rather than status. and will prevent trust from building. Once recognised. however the long-term benefits are worthwhile. The benefits of task allocation are that it allows a structured approach. groups at a very minimum need a vision. This will assist the group through the stages of group development. making a particular alliance. therefore having implications for the leadership of that group. The more important the task is to the individual within the group. offering almost no leadership. then this can lead to the formation of different roles within the group and also increase stress. the more committed they are to achieving it. Group cohesiveness will often be enhanced if members are situated in the same location or within close proximity of each other. the group leader. It is important that the task is clear and unambiguous. However. The timescale of the task also reflects how the group will work together. wdp consulting six sigma . If the timescales are very tight. much time wasting and duplication. However. The group that is formed to formulate ideas about the detail of the task needs a more supportive management style and may not initially have a structured approach to its work. and if this is not specific the group must make it so as soon as possible. motivation was low. either by talking to individuals separately. Another important factor is the 'salience of the task'. was very weak. then the group may not develop to deal with individual needs. this resulted in fragmented group. if the task requires ambiguity.

If a group is formed to work on a task that is not part of there everyday work, it can be useful to have a 'project room', so that members can work in the room when working on the project to avoid interruptions. I have worked on a project where this has not been the case, and performance has suffered due to the level of interruptions and distractions on everyday matters. However the needs of the project would dictate the need for this facility. Communication is important in developing group cohesiveness. The more easily groups communicate with each other the more cohesive they become. People have varying levels of belonging needs. Groups of people that have higher belonging needs are likely to require greater access to communicate with each other. Regular meetings ensure a certain level of communication and make people feel part of the group, leading to better performance. However it is important the meetings allow discussion. One group I worked with held regular meetings, but it was essentially a briefing, and discussion was discouraged by the leader of the group. The result was that people did not feel as though they belonged to the group. Technology can affect the way in which work groups function. The nature of technology can be unfavourable for the creation of work groups and a source of alienation, especially for manual workers. The impact of information technology is likely to lead to new patterns of work organisation, and affect the structure of groups. For example, physical location may cease to be a problem in forming a group if information technology links can easily be made. However this does not allow much interpersonal contact, which may negatively affect the performance of the group. If technology is likely to lead to alienation, mechanisms should be put in place to overcome this. External threat can serve to dramatically increase the cohesiveness of a group, with members uniting against the common threat. This may continue once the threat has been removed. If this occurs, the group should be managed to make best use of the increased cohesiveness, ensuring effort is focused in the right direction. However threats should not be over-exaggerated and the group leader must be honest about the threat.

The Intervening Factors
These factors can be changed in the short term to help the performance of a group. Leadership / facilitator style can affect a group substantially. A change in style can have tremendous rewards for group performance, and therefore the style has to be 'right' for the groups and the situation. Conversely, poor leadership / facilitation is likely to result in poor performance. Guidance, support and encouragement must be tackled according to the group needs, ensuring good employee relations. To improve performance the leader / facilitator must be constantly aware of group and individual needs and try to satisfy these. In my experience the more successful groups have leaders who are
wdp consulting six sigma

willing to listen, observe and react according to the situation, and thereby maximising participation. Poor performing groups often have leaders who 'bulldoze' on regardless of the group. Motivation is a key influence to group performance. If individual motivational needs are not recognised and met, individual satisfaction and performance will diminish which will inevitably lead to poorer group performance. Performance will improve if individual’s needs are clearly recognised and met. Motivation theory suggests, and my own experience supports this, that individuals have differing needs. Some are motivated by money, but many by recognition and the sense of achievement. It is therefore important to reward people satisfactorily, give praise where it is due - both privately and publicly, and ensure individuals have from which they will get a sense of achievement from completing.

wdp consulting

six sigma

Group motivation will be enhanced if individual motivation needs are satisfied and if the members like each other; if they approve of the task; if they wish to be associated with the standing of the group in the organisation; and if members know the expected results and are given feedback. Motivation will show itself in the actual behaviours of an individual that will be related to the role they play in the group. It is therefore important to regularly clarify these. The diagram below shows the factors affecting positive and negative performance.

Negative Performance

Expected patterns of behaviour

Positive Performance

ROLE CONFLICT Role ambiguity Role incompatibility Role overload Role underload

Role perception Role motivation Role capability

Role sanctions Role Stress

ACTUAL PATTERNS OF BEHAVIOUR

Various processes and procedures have to be undertaken in order for any group to be effective. These can be categorised into two functions - task & maintenance. The task function is often focused around problem solving, and the way in which a group tackles this may affect the solution. The maintenance functions are important to ensure the general well being of the group and they include - encouraging, arbitrating, peacekeeping, clarifying, summarizing and standard setting. The methodology used for undertaking the processes and procedures will affect performance. For example it is important to involve the group in standard setting; otherwise people will not accept and adopt them. I once led a pre-selected project team to a specific goal. The team had the same line manager who was not on the team. In this case it was not just my responsibility to undertake some of the maintenance functions, but also the line manager. He declined to do this, and as a result I believe performance was adversely affected. It is therefore important to recognise and ensure that the maintenance functions are dealt with by the key stakeholders, and the group leader / facilitator.

wdp consulting

six sigma

Groups. This core is moderated by the personality characteristics of the group’s members. and a status hierarchy that defines its structure. leading to the desired outcome. Communication and decision-making. the size of the group and the stage in its development. Motivation and productivity. 1. wdp consulting Developing mutual acceptance and membership. six sigma . Many of these factors can be seen in the following group behaviour model:- Group decision making Leadership CONTINGENCY VARIABLES Member personalities GROUP STRUCTURE Roles Norms Status Group Performance Communication Member satisfaction Development Stage Size Group cohesiveness Power & Politics Conflict The core of the model is the group structure. and performance expectations must be altered accordingly. like individuals. The group will function differently at various stages of development. norms. While leadership provide direction. have a growth cycle. Control and organisation. it is also important to consider group development.Good management of the given and intervening factors will enhance group performance. One model identifies four distinct stages in group development. 2. Every group has roles. Group Development When examining the components of group performance. 3. group members interact through communication and collective decision-making. 4.

which inevitably must occur at some time or other. 3. This will also help overcome the 'mourning' phase of a group.group behaviour and standards are developed. Another model is also split into four stages:1. Storming . and that is the destruction of a group. 4. Neither of the two models illustrates the later stage in group development. and teambuilding exercises should not just be saved for the beginning of a group’s task. This model assumes the group does not perform well until the last stage in development.This model shows that a group build up to achieving a high level of performance by the third stage by understanding and communicating with each other. as appropriate. Forming . but undertaken at various intervals.bringing the group together. Performing . Teambuilding exercises may help with this. leading to the 'mourning' stage. It is also important to recognise that teams need to be maintained. and should not be overlooked even if the team already know each other. with members working more independently. The fourth stage allows for a more flexible group. throughout the life of the task.individual views become more open and conflict may occur. when performance will naturally diminish. 2. where performance may fall when members leave the group. and does not recognise more independent working in the later stages. Performance will improve if groups can 'get through' the first stages as quickly as possible. wdp consulting six sigma .the group effectively performs the task set. Norming .

wdp consulting six sigma . There are differences between formal and informal groups. motivation and some of the processes adopted.CONCLUSIONS There are many differing factors that can affect the performance of a group. The factors that can be altered include leadership style. the leadership / facilitator of the group which should be a positive role model but contribute to the task and the group. It is these factors that I recommend are initially examined when seeking performance improvements. The group development cycle is an important consideration when examining the level of performance to expect. while others can be changed. The given factors may include the members of the group. the members of the group and their willingness to share ideas and communicate. As informal groups tend to be voluntary. In my experience the most influential factors are:• • • the interest level of the task to the team members. and the potential tangible outcomes. communication and openness may be easier to obtain. Some of these factors are given. the objective of the group the resources available to the group and any external influences. but formal groups are more likely to be more task orientated.

10 2.2000 81.00 0.4 5 8 10 20 30 40 70 100 150 230 330 480 680 960 1350 1860 2550 3460 4660 6210 8190 10700 13900 17800 22700 28700 35900 44600 Yield % 94.81 5.50 1.3200 88.3200 91.51 0.70 4.20 wdp consulting .9992 99.76 5.0000 31.31 5.9990 99.61 5.8000 75.60 3.0000 14.00 1.0000 54.22 1.9997 99.8140 99.40 1.9980 99.8000 72.60 2.00 4.0000 12.62 0.8650 99.9040 99.50 4.70 1.92 5.51 5.73 0.9320 99.0000 22.40 2.30 2.91 4.70 3.1300 96.40 3.3790 99.7300 97.5000 86.2200 97.1810 98.11 1.0000 28.12 5.5340 99.80 2.9670 99.10 3.8000 58.80 1.22 5.32 1.9850 99.9970 99.9960 99.7450 99.70 2.40 4.Sigma Table Yield % 99.80 4.92 0.5400 3.80 3.0000 43.30 Defects Per Million Opportunities 3.10 4.30 4.0000 19.0000 50.6000 61.00 2.0000 46.42 0.9300 98.6100 98.44 5.0000 39.9900 99.0000 35.00 5.50 3.9200 90.6540 99.2000 65.5000 84.60 4.00 3.9520 99.90 1.9930 99.6000 69.83 0.0000 Sigma 3.9770 99.0000 25.50 2.90 2.60 1.33 six sigma Defects Per Million Opportunities 54800 66800 80800 96800 115000 135000 158000 184000 212000 242000 274000 308000 344000 382000 420000 460000 500000 540000 570000 610000 650000 690000 720000 750000 780000 810000 840000 860000 880000 95.20 4.0000 16.5200 93.20 2.9995 99.90 3.4100 Sigma 6.6000 78.

Symbols for Process Mapping / Flowcharting System Approach Process step (activity) Decision End of process Start of process or link Information block (inputs or outputs) Direction of flow wdp consulting six sigma .

decision Incoming material Planned process delay Storage Operator Planned movement where potential control is required Different products may take different routes Raw materials / subassemblies Eg: curing time for rubber Product is intentionally placed in a bin / rack/ pallet Used to show where people are found in the process 10 20 6 30 Operation Multiple operations Operation with inspection Inpection 40 Batch 001 50 60 wdp consulting six sigma .Process Approach Symbol Name Description Inputs are transformed / changed Operation 20 makes 6 products at a time Eg: grinding operation with automatic gauging Manual or automatic inspection Movement in the process Primary flow Transportation ? Process path .

number of people. reject level A process in which material or information is flowing Description Movement in the process wdp consulting six sigma .Process Flow Approach Symbol Name Manual information flow Electronic information flow Information Manufacturing process description Push movement arrow Inventory Data box Where information or material accumulates Shows information such as cycle time.