Six Sigma Principles

Organizations use Six Sigma principles to improve quality, decrease costs, and better meet customer needs. In the book "The Six Sigma Way", define Six Sigma as "a comprehensive and flexible system for achieving, sustaining and maximizing business success. Six Sigma is uniquely driven by close understanding of customer needs, disciplined use of facts, data, and statistical analysis, and diligent attention for managing, improving, and reinventing business process".

Six Sigma's target for perfection is the achievement of no more than 3.4 defects, errors, or mistakes per million opportunities. This target number is explained in more detail later in this section. An organization can apply the Six Sigma principles to the design and production of a product, a Help Desk, or other customer-service process.

Projects that use Six Sigma principles for quality control normally follow a five-phase improvement process called DMAIC (pronounced de-MAY-ic), which stands for Define, Measure, Analyze,Improve, and Control. DMAIC is a systematic, closed-loop process for continued improvement that is scientific and fact based. The following are brief descriptions of each phase of the DMAIC improvement process:

1. Define: Define the problem/opportunity, process, and customer requirements. Important tools used in this phase include a project charter, a description of customer requirements, process maps, and Voice of the Customer (VOC) data. Examples of VOC data include complaints, surveys, comments, and market research that represent the views and needs of the organization's customers.
2. Measure: Define measures, then collect, compile, and display data. Measures are defined in terms of defects per opportunity.
3. Analyze: Describe process details to find improvement opportunities. A project team working on a Six Sigma project, normally referred to as a Six Sigma team, investigates and verifies data to prove the suspected root causes of quality problems and substantiates the problem statement. An important tool in this phase is the Fishbone or Ishikawa diagram.
4. Improve: Generate solutions and ideas for improving the problem. A final solution is verified with the project sponsor, and the Six Sigma team develops a plan to pilot test the solution. The Six Sigma team reviews the results of the pilot test to refine the solution, if needed, and then implements the solution where appropriate.
5. Control: Track and verify the stability of the improvements and the predictability of the solution. Control charts are one tool used in the control phase.
   

Using Work Results for Project Quality Control

Have you ever had a quality plan, but wondered how the plan provided quality control information that could be used for analysis? The quality plan has reporting procedures and feedback that provide this information. This information is called work results. The two most common work results are performance reports and change requests. Each will be examined in turn.

Performance reports
Performance reports provide data that can be turned into various visual charts for analysis. Three of the most useful activities that performance reports measure are:

• schedule adherence
• cost adherence
• quality standards adherence.

In performance reporting, the simpler the visual projection of the data, the better. This makes it easier to understand the requirement for action. Four of the most common performance reports are: Gantt charts, S-Curves, histograms, and tables.
In histograms, data is shown as a vertical bar graph. This illustrates major problem categories. It forms the basis for control charts and Pareto diagrams.

Tables are often used to convey more complex data in its raw data form. Tables can convey observational data.

Performance reports can have a direct impact on quality management. Simple visual images can easily emphasize the need for immediate change. They can also indicate the project plan is working well.

A project manager for a microchip project submitted a weekly performance report after a difficult week. The visual report showed that most of the department's problems were coding errors. This indicated that changes were needed to prevent errors from reaching the final product. The project manager will have to look into this problem in more detail.

Change requests
Almost every process or project encounters some difficulty that requires a change in the conditions of the project such as time, cost, or quality objectives. Project managers may request additional time or money to ensure the project meets its original definition and expectations for quality.

Change requests, another form of work results, ask for the alteration to the project's objectives or quality. This can occur if the product is urgently required or additional expenses for quality will not result in increased profits or sales.

Change requests can also require alterations to quality methodology. Changes in the handling of data used to measure the project, the measurement process and techniques of data collection, or the evaluation of data, can be requested.

The handling of data may change in its method of collection, depth of detail, or type of data. For example, a company introducing new machinery may need to develop new measurements to reflect the change in machinery or technology.

Changes to the process or techniques of data collection are meant to ensure reliability, consistency, standardization, review, timeliness and rapid access to data. For example, a company may adopt an advanced database to aid inputting and calculations.

Change requests can also require new methods of analysis and improvements to the quality of data. For example, a manufacturing company using new equipment may want more precise data and more sophisticated analysis.

The value of the work results in the form of performance reports and change requests lies in their identification of a potential problem. The data these results provide points the way for more in-depth analytical treatment.

Results of Quality Control Measurements

Results of quality control measurements are inputs to project quality assurance. These results are the testing and measurement records of quality control, and include acceptance decisions, rework, and process adjustments.

From a quality assurance perspective, you can use the results of quality control measurements for comparison and analysis purposes. Specifically, you can compare the results found with the expected results of the product or process, pinpoint exactly where non-conformities took place, and examine the reasons for non-conformities. Details about these important inputs to project quality assurance are provided below.

1. Acceptance decisions
How many work-related decisions do you make in one day? The number is likely to be much higher than you realize. Throughout the duration of a project, the number of quality issues that require a decision may seem endless.

One type of quality assurance decision is an acceptance decision. This results from an inspection of a process stage, or a final product or service. The acceptance decision is simply this—the item will either be accepted or rejected.

2. Rework
When an item or process is rejected as the result of an acceptance decision, it may require rework. The purpose of rework is to make non-conforming or defective items or processes meet requirements or specifications. Companies try to minimize the need for rework, as it causes projects to miss deadlines and budgets.

3. Process adjustments
Immediate corrective or preventative action should take place when rework is required as the result of an acceptance decision. This course of action is called a process adjustment.

The changes that are made as a result of a process adjustment will help prevent the defect from occurring in the future. More details about process adjustments are provided below.

• Find the source of the problem. Process adjustments involve more than fixing non-conformities. They include discovering the source of the problem itself.
• Take corrective action. Process adjustments may be small adjustments, or major undertakings, depending on the nature of the problem.

The results of quality control measurements are acceptance decisions, rework, and process adjustments. These quality assurance inputs are an important part of discovering the reasons behind non-conforming projects or processes.

Operational Definitions and Project Quality

Individuals working on a project should have an eye for detail, as it is an important aspect of quality. What kind of quality details do you keep in mind throughout a project?

Operational definitions focus on many kinds of details. Perhaps these include details you have also considered. Operational definitions describe what something is and how it is measured by the quality control process.

Operational definitions, sometimes referred to as metrics, are used to measure quality. They are quality assurance inputs that project management teams used to indicate the specifics about the quality of their projects.

For example, to define the testing process of a new pharmaceutical drug, the operational definition must describe the drug's composition, the type of study conducted, the variables tested, and the subjects used.

Defining how something is measured can be challenging, as there are many details to consider. To simplify this process, an operational definition answers the following three questions.

1. What is measured?
For example, A.L. Textiles manufactures a wide variety of textile products for industry and commerce. To be sure each material conforms to its intended purpose, the operational definition details the makeup of the textile and indicates how it is measured for quality.

2. How is it measured?
A.L. Textiles measures the durability of the medical uniform style AJ347. The product is 65 percent polyester and 35 percent cotton. It's available in adult sizes, small to extra large, and comes in colors white, gray/blue, and green.

3. When is it measured?
The durability of uniform style AJ347 is measured after initial manufacturing by washing the garment 300 times. The condition of the garment is recorded on a check sheet. In addition, the durability of the uniform is measured after any necessary product rework.

Remember, operational definitions are used to measure quality. They indicate the specifics about the quality of a project's product or service. In particular, they describe what something is, describe how it is measured, and describe when it is measured.

Project Cost Performance Measurement Techniques

One of the most important aspects of project cost control is cost performance measurement. You can use a number of performance measurement techniques to measure cost performance, including cost variance, earned value management (EVM), and the cost performance index. Details about these three cost performance measurement techniques are provided below.

1. Cost variance
Cost variance (CV) is the most basic performance measure. Simply stated, cost variance is the difference between the earned value and actual costs. A positive variance indicates that the project is running under budget, while a negative variance means that costs are overrunning. For the purpose of tracking over- or underrun percentages, you may want to use tables, Gantt charts, or bar charts.

Cost variance is typically expressed as a ratio or percent. You can calculate CV by comparing the actual cost of the work (AC) to the earned value (EV). Follow the steps below to calculate cost variance.

• Calculate the difference between the earned value of the project and the actual costs.
• Divide this amount by the earned value.
• Multiply this figure by 100 to obtain a percentage.
• Keep the negative sign for cost overruns.

The project manager's goal in calculating variances is to provide the basis for earned value management. You must understand the problems behind variances and take action that will correct any problems.

2. Earned value management
Earned value management is perhaps the most useful activity in cost control because it combines costs and the schedule into one indicator. It tells you how much the project is physically accomplishing in terms of both cost and time, giving management a more accurate and timely report on project progress.

The concept of earned value management multiplies the project budget (planned value, or PV) and percent-complete figures to arrive at a budgeted dollar value of the work that has actually been completed so far. The main difficulty in using earned value data to measure cost performance is in determining work completion. How does one accurately measure how much of a task is complete, while avoiding subjectivity in measuring performance as much as possible?

There are five methods you can use to assess work completion. They are described below, from the most conservative and least accurate to the most accurate.

• The zero/100 rule. Many companies do not assess percent complete incrementally. This removes any subjectivity. A task is assessed as either not done (zero percent complete) or finished (100 percent complete). This method works well for activities with a short duration—less than a month, for example.
• The 20/80 rule. This method is almost as conservative as the zero/100 rule. When it is started, a task is considered to be 20 percent complete, and 20 percent of the PV is charged against its account. When the task is complete, the remaining 80 percent of the budget is applied to the task.
• The 50/50 rule. This is probably the most popular method. You assume that once a task has begun, 50 percent of its budget is used. When a task is complete, it has used the other half. For a project with a large number of tasks, this method provides a fairly accurate way to calculate earned value.
• The milestone method. This is used for long work packages that are broken down into distinct milestones. A budget is assigned to each milestone instead of to the task as a whole. Value is earned when each milestone is completed.
• The percent complete. This method is usually used for long-duration work packages (for example, ones that last three months or more). Your project may not have identifiable milestones, but you are still able to estimate the percentage of the task that has been completed.

3. The cost performance index
You can use the earned value figure to establish another important performance indicator. Calculate the ratio of earned value to the actual costs to find out how efficiently your team is accomplishing the work. This ratio is called the cost performance index (CPI). The formula for calculating CPI is as follows:

CPI = EV ÷ AC.

When the CPI is measured periodically, you can plot CPI figures in a line graph to see the trend over the life of the project. This is called a trend analysis.

You will usually see the cost performance index reported along with its "companion" indicator—the schedule performance index (SPI). The SPI is the ratio of earned value (EV) to the planned costs (PV).

Project managers use the CPI and SPI to rate the cost and schedule performance of their projects. A poor rating provides a warning signal, allowing for corrective action to be taken before it's too late. These indexes fall into three categories:

• If equal to 1.0, performance is exactly as planned.
• If greater than 1.0, performance is better than planned.
• If less than 1.0, performance is poor.

Evidence shows that without corrective action, most projects will continue to perform at their cumulative CPI rate. Once the project is about one-third complete, you will have difficulty recovering from a CPI of less than 1.0 without aggressively managing the remaining tasks.

Remember, it's important to measure the cost performance of your projects. By using the three techniques described above, you can control project costs and ensure the project comes in on-budget.

The Project Cost Baseline

What do you get when you cross a project budget with a schedule? It's something that gives you an idea of the total budgeted costs of project activities per time period—otherwise known as the project's cost baseline.

A cost baseline, which is an output of the cost budgeting process, is a time-phased budget that is used to measure and monitor cost performance on a project. It is developed by summing the estimated costs by period. The cost baseline takes the form of a cumulative cost curve, or an S-curve.

Why does a project's cost baseline look like a curve? It has to do with the typical spending pattern, or "burn rate," of project funds. A cost baseline usually takes an "S" shape, when plotted on a graph, for the following reasons.

• Costs in the planning or design phase are usually low.
• Once production begins, costs increase quickly relative to the passing of time.
• Costs decrease toward the end of the project during wrap-up and delivery.

But why stop at just one type of curve? You also can plot and display the values of expected cash receipts and actual project costs. Plot these curves next to the cost baseline for comparative purposes.
Plot expected receipts next to the cost baseline to display a forecast of the project's cash flow. You may have several periods of time during the project where you have significant expenses but have not received payment for that work, especially if you are paid only for deliverables.

If this creates a need for short-term financing to cover bills, you will want to know the likely time frame in which this will occur. Then you will be able to predict the terms for any necessary loans.

Plotting budgeted costs and actual costs side by side enables you to quickly see whether cost performance is good, evaluate the size of cost variances, and easily report cost performance to stakeholders.

Before you can plot planned costs over time, you have to establish the coordinates—that is, what costs should be per time period. If you are using a computerized scheduling tool, simply input the data and generate a cost report by time period. The software will probably create the cost curve for you.

In the absence of computerized tools, you can use a bar-chart of your schedule to establish cumulative costs over time. Follow these steps.

1. Draw a chart of activity durations.
2. Assign a budget to each activity.
3. Measure the accumulated costs per time period.
4. Plot cost-per-time data on a graph with duration on the x axis and cost on the y axis.
5. Connect the dots to see the curve.

The cost baseline for your project is the plan against which you will measure, monitor, and control costs. Use the cumulative cost curve as a graphical representation of the baseline when comparing planned costs to actual costs.

The Project Cost Management Plan

Besides the cost estimates themselves, the most important output from estimating project costs is the cost management plan. This is the document that outlines how project costs will be kept on track.

As excellent as your cost estimates may be, you must have an adequate plan for managing cost variances and dealing with problems. A good cost management plan (CMP) clearly outlines how to implement corrective action in order to reduce or eliminate cost variances.

The main purpose of a cost management plan is to provide direction to project management teams. It must contain guidelines to follow, outlining what to do when positive or negative cost variances occur. Cost problems are usually handled according to their severity. Variances that pose no real threat to overall cost performance may require no action at all.

Cost management plans can be structured differently depending on the performing organization and the needs of the project stakeholders. The points below describe a broadly-based plan and a detailed plan.

• In a broadly-based plan, costs are estimated, monitored, and reported at the high-level work breakdown structure (WBS) elements. Tasks are not complex.
• A detailed plan will deal with costs at the lowest level of the WBS, making it easier to obtain timely and accurate information.

You also must decide whether your cost management plan will have a formal or informal structure. Bureaucratic structures are much more formal than project-based companies. Design a CMP that is consistent with the organizational culture of your company and with the authority that project managers hold. More details are provided below.

• Bureaucratic. Does your company have a bureaucratic and inflexible structure? If so, you likely will require formal forms, processes, approvals, and reviews for every step of the cost management process.
• Mixed. Your project team may work within a formally structured organization that answers to upper management but has the authority to manage project costs at its own discretion and without all the red tape.
• Project-based. The cost management plans of small, entrepreneurial companies typically reflect the increase in decision-making authority held by project managers. Processes are flexible, depending on particular projects and situations.

With the four options available—broad-based and detailed, bureaucratic (or formal) and project-based (or informal)—for structuring cost management plans, a type of matrix is formed from the combinations.

• Formal-detailed. This type of plan would be appropriate for a large institution with rigid accounting procedures.
• Formal-broad. The processes are vigorous but costing categories are general.
• Informal-detailed. The cost of every team activity is outlined but loosely monitored.
• Informal-broad. Costs aren't closely monitored or very detailed.

The two extreme situations are easy to identify. For the others, you need to look at organizational structure, level of detail, degree of decision-making authority, and the important of auditability.

How will you know if your project's cost management plan is a good one? Assess whether or not it meets the needs of the project stakeholders. Stakeholders are the people involved in producing the project or who are affected by the project's end result. They include the project team, investors, and the customer. Stakeholder needs may vary from one project to another.

Remember, as you develop cost estimates for your project, keep in mind that your project also needs a cost management plan to help you manage project costs after the budget is developed and project activities are under way.

Measuring the Impact of Approved Project Changes

If you are familiar with performance measurement techniques, you know that one of them—earned value—is used to forecast estimates before work is started. Did you know you could also use it to forecast the effect of a project change?

When a change is verified, the work must be incorporated into the project plan, which usually increases the cost and lengthens the schedule. As the work for the change progresses, earned value (EV) analysis can be used to forecast how the change will affect the project completion estimates for budget and schedule. This will allow the project manager to better control the change.

• Planned value (PV) - is the approved cost estimate for a change during a specific period of time. It answers the questions: "How much will the work for the change for this time period cost?" and "How much work should be done by now?"

• Actual cost (AC) - is the real amount it costs to perform the change in a given time period. It answers the question: "How much has it cost for the work on the change so far?"

• Earned value (EV) - is the value of the work actually completed on a change in a given time period. It answers the question: "How much work is done and what was the original budget to complete that work?"

To use EV analysis to measure the performance associated with the change and the project as a whole, you will need to calculate two performance indicators:

• schedule performance index (SPI)
• cost performance index (CPI)

The SPI is the ratio of earned value of the accepted change to the planned cost of the change at the present point in time. It indicates how the change is affecting the schedule right now. To calculate the SPI, divide the earned value by the planned value.
Project managers can use the information resulting from the SPI calculations to make changes to the final product delivery schedule.

• If SPI is equal to 1.0 - no changes to schedule are necessary.
• If SPI is greater than 1.0 - project will finish ahead of schedule.
• If SPI is less than 1.0 - additional days will be needed to complete the project.

The EV for Carla's textile project is $188,000 and the PV is $200,000. To calculate the schedule performance index for her project, Carla divides the EV of the change, $188,000, by the PV of the change, 200,000, to equal 0.94. Since the result is less than 1.0, Carla will need to add additional days to her project schedule because of the change.

Another factor that you must look at when a change has been accepted is how the change will affect the final cost, or cost upon completion, of the project. To do so, calculate the cost performance index (CPI). It divides the sum of all individual EV budgets by the sum of all AC budgets. When calculating CPI, round your answer to the first decimal place.

CPI = Total of EV/ Total of AC

The results of the SPI and CPI are then used to forecast the project completion estimates for the changed project.

Earned value analysis provides a way to measure project performance and determine where your project is heading after the approval of a change to the project. This helps project managers to determine if the finished project will be on time and within budget.

Dealing with Performance Problems

Leaders can help their employees improve their performance by using a progressive discipline system.

To help an employee with a performance problem, you should:

• get to the heart of the problem
• respond to the problem by preparing an action plan
• use discipline to deal with the problem.

Get to the heart of the problem
When there is a problem with an employee's performance, you should get to the heart of the problem before you attempt to correct it. This is important because the cause of a performance problem is not always apparent.

To correct the problem, you need to address the cause. To determine what is causing a performance problem, you should ask the following questions:

• Does the employee understand the goals and standards of the job?
• Does the employee get timely feedback on his or her performance?
• Did the employee get adequate training for the job?
• Does the employee demonstrate competency in other tasks?
• Is the employee capable of doing the work?
• Does the employee have the resources to do a good job?

Respond to the problem by preparing an action plan
Once you have determined the cause of a performance problem, you should meet with the employee involved to discuss possible solutions.

When you do this, don't get personally involved in the problem or apologize for having to deal with it. You want to help the employee succeed. Find out what the employee would like to see as an outcome. Then set up a plan of action.

Begin with the assumption that any corrective action you take should first and foremost help improve an employee's performance.

The following gives more about preparing an action plan:

• Make sure that employees with performance problems have a clear understanding of their jobs.
• Present clear information about how the job should be done, clarify each step, and emphasize the major responsibilities.
• Provide training if necessary.
• Make your employees responsible for the outcome of this process. Ask them to check back with you frequently.
• Encourage your employees.

Use discipline to deal with the problem
When it becomes necessary to take disciplinary measures, you should use a four-step procedure. Because these steps sometimes lead to termination, you should ensure that you have all of the facts and use the same procedure for all employees.

It's also important that you do not allow your judgment to be clouded by prior, unrelated problems. There is more information on the steps below:

1. In step one, verbally remind the employee about the problem, and state that the problem is now in the formal disciplinary stage.
2. Go to step two if the problem continues. Give the employee formal written notification of the continuing problem.
3. If these steps fail, move to step three, in which the employee is given one day of paid leave to decide whether to stay and improve or leave. This demonstrates the organization's desire to give the employee control over whether to stay or leave.
4. Step four is termination. Having documented all the previous steps in the employee's file, set a specific time for an interview. Inform the employee of your decision. Keep the meeting simple and short. The employee had every opportunity to improve. There is no need for further discussion.

By following well-thought-out steps in correcting performance problems, a leader can keep employees motivated and effective. Do you have procedures in place that make discipline a progressive measure?

Project Objectives

Project objectives define target status at the end of the project, reaching of which is considered necessary for the achievement of planned benefits. They have to be formulated as S.M.A.R.T.

• Specific, with
• Measurable (or at least evaluable) achievement,
• Achievable (recently Acceptable is used regularly as well),
• Realistic and
• Time terminated(bounded).

The evaluation (measurement) occurs at the project closure. However a continuous guardance on the project progress should be kept by monitoring and evaluating.