Quality in project management--a practical look at chapter 8 of the PMBOK® guide

Abstract

Project quality management is a vital aspect of any project, yet it is often misunderstood or improperly applied. Chapter 8 of A Guide to the Project Management Body of Knowledge, Third Edition (PMBOK® Guide), addresses the various aspects and importance of the topic, however, it doesn’t really tell project managers how to apply the tools and techniques effectively and with confidence. This paper attempts to make the topic of “quality” easily understandable and applicable to any project. Included within this article are practical advice, hints, and suggestions on the three aspects of project quality: Planning, Assurance, and Control. A project manager must have a firm grasp on how to effectively utilize data and measure results to effectively communicate with various stakeholder groups. If a project manager and a project team understand the various quality tools as well as how and when to use them, they will ultimately make better decisions, move the project along faster, and be much more successful with project recommendations and implementation.

Introduction

For most people, crunching numbers, dealing with data, and making charts is simply not too much fun. Most likely the data you want or need is either hard to access or nonexistent. If you actually can access the data, getting it in the format you need can be another hassle, altogether. To make matters worse, analyzing data, especially large data sets, can be confusing and downright monotonous. Yet few would argue having good supporting data makes project life easier, especially when communicating with stakeholders and top management. The proper use and application of data and data analysis can help just about any project be more successful. Often times, unfortunately, project managers struggle with how to effectively use data, and various analysis techniques, to make better, more informed decisions. “Quality management” is one of those ambiguous topics that can be quite confusing and maybe just a little bit scary. Project managers who struggle with the quality aspect of project management need some straightforward and practical advice on how to apply those quality tools and techniques mentioned in Chapter 8 of A Guide to the Project Management Body of Knowledge, Third Edition (PMBOK® Guide) The following article attempts to give readers several suggestions and guidelines on incorporating quality concepts, tools, and techniques into the successful management of any project.

The PMBOK® Guide summarizes the tools and techniques associated with quality management in the table shown in Exhibit 1. Project quality management is broken down into three main processes: Quality Planning, Quality Assurance, and Quality Control. At first glance each process group has an imposing list of inputs, tools and techniques, and outputs. Keep in mind these tools are not new. They have been used in business settings for decades. Most of the analysis and charting techniques listed in the table can be done in a basic spreadsheet. The key for project managers is to simply incorporate the right tools from each process during the course of a project.

Project Quality Management Overview (PMI, 2004)

Exhibit 1 – Project Quality Management Overview (PMI, 2004)

Quality Planning

One of the most important aspects of Quality planning is the establishment of quality metrics. Project managers must go beyond the traditional metrics of scope, time, and cost. It is imperative to link a project to the strategic objectives of the company, organization, or business unit. As mentioned in the article, Project Management & Six Sigma, Use Six Sigma Methods for Better Project Results (Rever, 2006) the project manager’s efforts on any project he or she is managing should result in some kind of improvement for the business. Ultimately, this is why a project is in existence – to make some part of the business better. To show improvement requires measurement. If a project manager wants active interest, participation, and support from key sponsors, it is essential to link the project to some metric of importance to the business. A measure the sponsor is responsible for is an easy way to approach this linkage. Perhaps a metric the sponsor owns, such as order accuracy, processing time, or turnaround time will be improved by the project. If the project manager can link his or her project to a measure the sponsor must report out on at the inevitable monthly results meeting, he or she is likely to have all of the support they need throughout the life of the project. A key aspect of quality planning is for the project manager to understand the processes his or her project is impacting. The project manager must then develop process measures for the project so as to measure the impact of recommended changes to those impacted processes.

It’s fairly common for people to struggle with metrics. They are a challenge, no question about it. A great way to think about metrics is to categorize them into three buckets: business measures, customer measures, and process measures. The Balanced Scorecard by Kaplan and Norton (2006)es customer and business measures in their framework of translating strategy into operational terms. Business measures are important to the business. Attendance, safety, profit margins, and sales are all examples of business measures. Targets and goals could also fit into this category. Customers don’t care too much about an organization’s business measures. Customers, whether they are internal or external, usually only care about a few things. Timeliness, accuracy, and meeting requirements are examples of customer measures. Some kind of customer measure should be incorporated into every project. The third bucket, process measures, is the key link between a project, associated processes, and what is important to an organization. The reason projects are in existence is to improve either business measures or customer measures. The way to improve those measures is by improving the process measures. As depicted in Exhibit 2, an excellent and straightforward approach to establish strategic metrics for a project is to do the following:

  1. As a group, the project team should flowchart the process their project touches as it is today. During this exercise, pinpoint which steps are the bottlenecks, constraints, work-arounds, or problem areas. The entire flowcharting exercise should take no more than 2-3 hours.
  2. Next, determine where the logical intervals or handoffs are within the process. These could be handoffs between departments, time stamps, milestones, or any other logical process break. Gather data on these interval or process measures and chart them.
  3. Finally, link each process measure to the ultimate customer or business measure the project is supposed to improve. Those process measures with the best linkage, or correlation, are where the project team should focus to improve results.
As-Is Flowchart and Associated Metrics

Exhibit 2 – As-Is Flowchart and Associated Metrics

Other suggestions for project managers to consider regarding linking their metrics to what is important to the organization have to do with cost of quality metrics. In his book Principles of Quality Costs, Jack Campanella (1999) suggests quality costs can be categorized into one of four groups: external failures, internal failures, appraisal, or prevention. Failure costs are the most expensive forms of poor quality costs for a company. If the aim of a project is to reduce failure costs such as rework, reinspection, scrap, returns, warranty claims, or refunds, those metrics should be tracked and linked to the ultimate recommendation and process changes of the project.

A very important tool under the quality planning process is design of experiments (DOE). DOE is an extremely powerful tool which, unfortunately, is often overlooked or not utilized by project managers. A DOE allows for the simultaneous testing of multiple factors or variables to determine which test factors improve results. DOE is often associated with the manufacturing environment. It is, however, a robust tool which can be utilized to improve results on just about any process in any industry including service, healthcare, finance, retail, and communications. To be sure, an experimental design does require a certain amount of expertise and know-how to run effectively. However, any project manager should be on the lookout to when, not if, but when they can utilize this tool to verify, with data, what works and what does not work before final project recommendations and implementation. As stated in Improving Performance Through Statistical Thinking (ASQ, 2000), it is often tempting to declare success after implementing a proposed solution prior to obtaining evidence that the problem has actually been solved. Doing this can strain people’s credibility (p. 102). A designed experiment allows for statistical verification and validity in determining what changes truly do make a difference in the key metric of interest. The message is clear: before making changes to a process, take the time to verify, through piloting and solid statistical analysis, if improvement ideas do indeed work. DOE is one of the best tools for that very purpose. As Exhibit 3 demonstrates, the main idea behind a DOE is to list test ideas, describe each factor in terms of “levels” (generally the low level is the current procedure and the high level is the new idea), and then select a test design to fit the number of test ideas. Project managers, who also have expertise in Six Sigma, perhaps as a Green Belt, have an advantage in terms of familiarity and comfort level with the DOE tool.

Test Factors, Levels, and Design of Experiment Matrix

Exhibit 3 – Test Factors, Levels, and Design of Experiment Matrix

There are many tools associated with quality planning. Decisions must be made as to which tools and techniques are needed to advance the project. Every project, however, should be better off if the project manager takes the time to link their project to what is important to the organization, flowcharts the processes impacted by the project, establishes appropriate metrics, and utilizes powerful verification tools such as design of experiments.

Quality Assurance

The process of quality assurance is associated with continuous improvement and process analysis. Before quality levels can be verified, it is imperative to have accurate data; as the old saying goes, “garbage in, garbage out.” Therefore, every project team should conduct a thorough measurement system analysis to verify the accuracy and integrity of the measurement system and the data. As described in The Lean Six Sigma Pocket Toolbook (George, Rowlands, Price & Maxey, 2005, pgs. 87-89) there are several components to a good measurement system:

  1. Accuracy – data reflects the true value of the property or what is being measured
  2. Precision – data is precisely measuring what it is supposed to measure
  3. Repeatability – successive measurements by the same appraiser should be the same
  4. Reproducibility – different appraisers measuring the same item get the same result

Time and effort should be made by the project manager and project team to ensure the accuracy and credibility of the measurement system. The credibility of future decisions is based on this vital quality assurance step.

Process analysis is another key aspect of quality assurance. Process analysis includes the topics of root-cause analysis and value-added analysis. Most project managers are familiar with root-cause analysis, in particular the use of a cause and effect or fishbone diagram. What is important to remember about root-cause analysis is to include the five major categories: people, methods, machines, materials, measurement system, and environment when investigating the sources of problems. It is easy to focus the majority of improvement efforts and corrective actions on the people. But remember, as the quality guru Edwards Deming (1982) has suggested, 85% of the problems in business can be associated with management. After all, management decides on the methods, procedures, materials, and processes so be sure to investigate root causes in each of those categories, as well.

Value-added analysis is another easily applied, yet effective, approach to continuous improvement. Exhibit 4 shows a basic swim-lane process map where each step is coded as either adding value (green) or not adding value (red). The rule of thumb for determining if a step adds value is if a customer would be willing to pay for that particular step. Actual processing time can be added to each process step and from there basic process efficiency calculations can be made. Usually the area of opportunity for improvement is the waiting or hold time between each step.

Value-Added Analysis

Exhibit 4 – Value-Added Analysis

Quality assurance is all about continuous process improvement. This includes the investigation or root-cause analysis of issues within processes as well as continual assessment of which steps in a process are adding value.

Quality Control

The last process under project quality management is quality control. Quality control has to do with monitoring the project metrics, identified in the quality planning phase, to ensure those metrics are performing at satisfactory levels. Quality control also includes understanding the concept of variation as well as how to effectively communicate with data. Metrics were identified in the quality planning stage while gathering accurate data for those metrics was part of quality assurance. In the quality control process, graphical analysis tools are used to display the data so decisions can be made easily and quickly concerning the quality of the output of the process. It is easy to get into the “analysis paralysis” mode with this step so the key is to keep the tracking as simple as possible. For the most part, simple tools such as run charts, control charts, pareto charts, and scatter plots can be used to monitor performance.

As mentioned in the article, Effectively Communicating with Data, (Rever, 2007) every project measure should have either a run chart or control chart. Run charts, often referred to as line charts, answer the question, “How are we doing over time?” When looking at data over time, the appropriate tool is a run chart or control chart, not a bar chart. Run charts are natural predecessors to control charts; they just don’t have control limits. Save bar charts for categories or “buckets” of information. Pareto charts are the appropriate simple tool for categories of information such as: reasons for service order errors, results by region, or types of customer complaints. Pareto charts answer the question, “What things are impacting the key metric?” Every run chart should have an associated set of pareto charts to help explain what things are affecting the key metric of interest. The run chart and pareto chart are a great “one-two” combination and go a long way towards simple, yet effective communication with stakeholder groups. A median line can be added to a run chart so process shifts can easily be detected. Exhibit 5 shows examples of a run chart and pareto chart combination as well as a control chart and scatter plot. Scatter plots are the appropriate tool to visually show if there is a correlation between two variables. Remember, putting two measures on one chart implies there is a relationship between those measures even if there isn’t one. Keep charts simple; one metric per chart is a good rule of thumb.

The Simple Tools

Exhibit 5 – The Simple Tools

Other suggestions for charting and monitoring results include:

  • Avoid becoming a “chartoonist.” Keep the charts as simple as possible.
  • Charts should speak for themselves. If you have to explain the chart it is not serving its purpose.
  • Avoid adding “trend” lines to charts. It is too easy to make an incorrect interpretation of a “trend” which, in reality may not even exist. Instead, use the rule of six or seven points in a row in one direction as an indication of a trend.
  • Try not to use only monthly data on charts. Plotting daily or weekly data makes it much easier to pinpoint what is happening in the process. Process shifts are also detected much faster than with monthly data.

Understanding the difference between normal variation and unusual data points is, perhaps, the area that could help improve decision making in business and on projects the most. Understanding the different types of variation is actually quite simple, yet it is extremely difficult to apply in the day-to-day business setting. People are natural problem solvers. In a business setting, managers have this inherent need to act even if, in actuality, there is no need to act. Project managers can go a long way in reducing knee-jerk reactions with their projects when they differentiate between common and special cause variation. The control chart will do this for you. The control chart is an extremely powerful tool that should be used in the monitoring process to determine if processes are in control.

When special causes are identified, root-cause analysis is the correct response using cause and effect diagrams discussed under quality assurance. When only common cause variation is present, yet results are not where they need to be, perhaps a design of experiment will help determine how to shift the process in the right direction. If a process measure is in control, that is it only has common cause variation, yet the results are not good enough, this indicates the process must be changed or improved. Asking people to explain differences in data between the control limits is not only frustrating but a waste of time. A better approach is to recognize the current process is not capable focus on improving the process. The best way to improve processes is through a process improvement project team. Understanding variation makes decision making much easier. Investigate special causes, reduce variation, and decide if the process is capable. If it the process is not capable, instead of reacting, form a process improvement project team to study the process and shift the process in the right direction.

Process monitoring is an integral part of quality control. The quality tools used in this process can assist project managers in numerous ways including: differentiating between normal noise in the process and something unusual, communicating effectively with easy to understand graphical analysis, and making better and less emotional decisions based on sound data and facts.

Closing

Projects are almost always emotional endeavors for more than just a few stakeholders. To take that emotion out of decision making and to communicate more effectively with top management, it is essential to incorporate quality processes into every project. To successfully improve results over the long term, project managers must develop a continuous improvement mind-set. This can be done by becoming familiar with and utilizing the various quality tools and techniques on every project. Project managers who take the time to establish good project metrics, analyze the processes affected by the project, and understand the concept of variation should improve the effectiveness of the management of their projects significantly.

ASQ Statistics Division. (2000) Improving Performance Through Statistical Thinking. Milwaukee, Wisconsin: American Society for Quality.

Campanella, J. (1999) Principles of Quality Costs. Milwaukee, Wisconsin: American Society for Quality.

Deming, W. E. (1982) Out of the Crisis. Cambridge, Massachusetts: Massachusetts Institute of Technology, Center for Advanced Engineering Study.

George, L. M., Rowlands, D., Price, M., & Maxey, J. (2005) The Lean Six Sigma Pocket Toolbook. New York, New York: McGraw-Hill.

Kaplan, S. &Norton, P. D. (1996) The Balanced Scorecard. Boston, Massachusetts: Harvard Business School Press.

Project Management Institute. (2004) A guide to the project management body of knowledge Third Edition (PMBOK® Guide) (2004 ed.). Newtown Square, PA: Project Management Institute.

Rever, W. H. (2007, July) Effectively Communicating With Data. www.AllPM.com July, 2007 newsletter.

Rever, W. H. (2006, March/April) Project Management and Six Sigma. Use Six Sigma for Better Project Results. www.AllPM.com March/April, 2006 newsletter. 1.

This material has been reproduced with the permission of the copyright owner. Unauthorized reproduction of this material is strictly prohibited. For permission to reproduce this material, please contact PMI or any listed author.

© 2007, Harry Rever
Originally published as a part of 2007 PMI Global Congress Proceedings – Cancun, Mexico

Advertisement

Advertisement

Related Content

Advertisement

Publishing or acceptance of an advertisement is neither a guarantee nor endorsement of the advertiser's product or service. View advertising policy.