Earlier you see signs, earlier you can take actions

This capability works in tandem with another skill, that of foresight. The project manager would need the capability to look ahead and discern warning signs. A pattern recognition skill that lets you deduce what a given set of warning signs mean is useful. Proactiveness is about looking out for such warning syndromes and then to take action early. Sooner deviations are brought under control, higher is the probability of a project completing on time and budget. Being able to detect signs early gives everyone a little more reaction time and set up the preventive and corrective actions necessary. This little bit of extra time is often the difference between success and failure. One thing is obvious, frequent formal monitoring points are useful. The project team gets to review the performance data at these points and the formal analysis undertaken at these times bring out any problems that are coming up. Reaction times to deviations are then decided by how frequent these monitoring points have been set up. These reviews being formal affairs involving many people can only be arranged at significantly long time frequencies. Besides, the formal analysis and the reviews take time.

However, many a times there are straws in the wind, some indirect signs that not everything is going properly. If one were to be on the lookout for such tell-tale signs, they usually are noticeable early. As soon as you see a pattern that something is going wrong, you could act and try and prevent the problem. The project manager will need to raise red flags to concerned department or group, if it can be associated with the work of some specific group or a department. This ability to discern a patter, would obviously come from a combination of knowledge and experience. Someone who has been there and done that would spot such events almost immediately some early warning signs start to appear. Thus it is imperative that the project manager is knowledgeable and experienced, particularly in the same domain area and relevant technology.

But, that’s only the hard part of the skills necessary. One needs to have these sensitivity about anything out of the ordinary, ability to detect a pattern and possibly the premonition of danger coming up are necessary as well. The indicators are not hard facts. The numerical or hard data about deviations from baseline are only available after the analysis on the work performance data has been made. So the project manager will have to depend on indirect indicators and the so called “gut feel”.

The items on the watch list of the manager are not only the direct issues such as a noticeable excess use of time compared to what has happened at that time. Lots of secondary effects need to be taken into account as well. For example, if there is excessive conflicts within a team or the team leader is unduly egotistic and commits mistakes in his decision making then these are sure to lead to time slippages. Costs go up in direct proportion too. So, what a proactive manager needs to do besides being watchful, is to resolves these conflicts and ensure project results do not suffer as a result. Waiting for the next formal review that will tell you how much the slip actually is, it is better to take action early.

Project manager’s job is not an easy one. There are a whole range of issues that a project manager has to manage. But carrying out all of these responsibilities what is required of a project manager is that he act with integrity in everything he does and personally he is through professional. Integrity means that he works towards doing the right thing always. Even that he also strives to do them the right way. Thus it is important that he should do what professionalism demand but also the means to achieve some end is also the right way to do it.

Project manager acts professionally and with integrity

Integrity requires that the project manager acts completely lawfully in the activities related to the project. He must attempt to carry out his actions that are morally right too. There should not be a moral conflict between end and means. If the end is required for the project the means to achieve that is also through professional and lawful processes. These compliance with law should cover all aspects of the law, regulations, safety requirements of the region or country he/ she is operating in for the sake of the project. If the customer or any other stakeholder is from another country laws and regulations related to such countries also need to be complied with fully.

If complying with the laws and regulations of this larger context is important, it is equally important to comply with the rules and regulations of the organization of which the project manager is a part of. All the regulations and policies that apply are religiously adhered to by a conscientious project manager. This professional honesty also reflects in the way the project manager interacts with customers, stakeholders and above all his team members. The communication tasks particularly the task of reporting project performance is usually tricky. The project manager being the overall in-charge of everything in the project, his performance is directly linked to that of the performance of the project. Thus reporting of project performance can easily get colored by subjectivity; that of the project manager looking good. However, a lot depends on professional reporting of this data. If the project is on a drift, that needs preventive/corrective actions that should be taken in time for these actions to be effective. This will get jeopardized if the reporting manager is not a professional and a person of integrity.

Demands of professionalism

Whether it is the hard aspects of time, cost etc. or soft aspect of the human resources it is absolutely essential that the project manager act professionally. It is also necessary that he be perceived as a professional too. Particularly with respect to human resources issues he ought to be a very objective one to be perceived by others as a just person, who does not take sides. This needs to be really irrespective of race, sex, age and all the rest. He needs to be respectful of multiculturalism. All the rest of common professionalism such as being punctual, honest, ability to maintain confidentiality are essential too.

As far as the domain areas or the knowledge areas are concerned, the manager needs to be fluent in the theoretical basis for his actions and be a great believer in the processes that are advocated for project execution. This professional faith in the process groups needed for a project at different stages is the sign of a project management professional. Thus religious implementation yet intelligent customization of the same would be expected from such a professional. These processes ensure that the project is initiated, executed, monitored and controlled and closed appropriately.

Ethical responsibilities of a project manager

The project management institute has a code of ethics for professional managers titled “ Code of ethics and professional conduct”, approved Oct 2006. This code sets the standards of ethics expected of professional managers. It talks of mandatory standards on many of the ethical aspects. These are to be definitely met. It also lists aspirational standards that every professional manager must strive to meet. One thing is certain, professional responsibilities of a manager include not only doing the right things but also to do them the right way and not cut corners/compromise when the pressure to meet performance goals are high.

The goals of the code are described as follows. “ This Code of Ethics and Professional Conduct describes the expectations that we have of ourselves and our fellow practitioners in the global project management community. It articulates the ideals to which we aspire as well as the behavior that are mandatory in our professional and volunteer roles.” The code is expected to enhance confidence in the profession and would make better persons out of the management professional by following these standards. Credibility and reputation of the project management profession is built through how everyone of such professional behaves.

The code defines four categories or aspects of the professional behaviour that is expected to meet certain standards. These apply to aspects of behaviour that include responsibility, respect, fairness and honesty. Each has mandatory standards that must be met and additional aspirational standards that every individual aspires to meet and  tries continuously to meet by improving self. Mandatory aspects of responsibility includes such issues as whistle blowing. Since the code aims to improve the standards of a community some means of complaints against persons not meeting the standard is essential. However, such whistle blowing cannot be frivolous. Violations must be borne by facts and attention of appropriate authority should be drawn to such incidents. It is wise to alert the persons affected by such violations. What individuals themselves must meet are that individuals are responsible to be aware of and uphold the policies, rules, regulations and laws about the volunteer or professional work carried out. Aspirations need to be directed at making decisions and taking actions based on the best interests of society, public safety, and the environment. It require s that the person take on assignments that fits his background, experience, skills, and qualifications.

Respect related mandatory requirements cover negotiations in good faith, not to exercise power to influence decisions for personal gain, not to be abusive and respect property rights of others. What the manager then must aspire are trying to understand others, discuss directly with whom conflict/ disagreements exists and behave professional even when they may not be reciprocated. Mandatory fairness standards require full and pro-active disclosure of any conflict of interest and a voluntary action not to participate in decision making in such situations. Favouritism, nepotism, bribery, discrimination are all strict no-nos. Apply rules of the organization without any bias. Aspirational aspects require complete transparency and constant examination of self to improve the sense of fairness. Honesty in all its manifestation, including not to do things for personal gain or at the cost of others, is demanded of a project manager. Seeking truth, being truthful in all communication and conduct, providing accurate and timely information, making commitments and promises in good faith and to create an environment where people feel safe to tell the truth are the aspect the project manager should aspire for. Overall that defines the ethical responsibilities quite clearly.

We discuss here the balance of the two of Ishikawa’s seven tools of quality and a few more of tools and techniques used in QC processes. Statistical sampling, inspection and review of approved change requests are the other tools & techniques used in QC processes and are discussed here.

Run charts are similar to control charts. The difference is that in does not show any of the limits displayed in control charts. For example the upper and lower limits as well as upper and lower control limits are not indicated. Thus it is a plot of measurements related to output items as the process runs. Thus this plot is able to clearly show variations over time, if drifts are happening or there is a declining drift, process improvements/declines over time etc. Thus this curve is easily amenable to trend analysis and let you mathematically predict future outcomes. Trend analysis will show up technical performance such as how many defects have been identified and how many have been corrected and how many remain etc. Time and cost trends also can be obtained that show how many activities show significant variances.

A scatter diagram is used to plot two variables against each other. The resulting graph can clearly indicate if the two variables are correlated. If they are not, one would expect the points generated from the values of the two variables to be random. The graph formed will then be a completely scattered set of points against the two axes. Whereas if they have strong correlation the graph would resemble a diagonal line, indicate that the value of one variable influences directly the other. This information then can be further used in determining root causes or planning a design of experiment.

Statistical sampling is all about taking samples, conducting measurements on them and drawing conclusions about the complete population from these data. The size of samples and how frequently they are to be taken are guided by statistical theories. Thus the QC personnel will need to be conversant with the theoretical background of such sampling and be able to decide the right size of the samples to be taken from the population of items we want to subject to QC activities. Sample are selected and tested as defined in the quality plans.

Inspections are about checking work products and determining if these conform to standards drawn up for the project. These may be from organizations own quality standards or specifically drawn up for the project. Inspections are carried out throughout the life of the project. How frequently or at what project steps are defined in the quality plan. The inspection level can be at any level and has to be determined a priori. The inspection level could be at a specific attribute level or right up to a set of measurements related to a product. Inspections are used to confirm defect repairs too. Peer review, Walkthroughs, audits etc are other synonyms that may be used in literature or in an organization.

Review of approved change requests is another tool that can help implement QC processes. Changes required are identified during QC activities. These are approved by the QC and the project teams. But to be doubly sure these should be subjected to review to determine if any of the requests are not really needed or an opportunity to improve a process has been missed.

Armed with these tools the QC department carries out the planned QC activities for a project.

There are about ten tools that has been indicated as useful in the PMBOK, for the QC process. The first seven tools are also known as Ishikawa’s seven basic tools of quality. In this part 1 of the articles on QC tools and techniques, the five tools out of the seven are discussed. The balance two and the three other tool/techniques such as statistical sampling, inspections and review of approved change requests besides the balance two of the Ishikawa’s tools are discussed in part 2 of this set of articles.

The Tools & Techniques

Cause & effect diagram helps in root cause analysis. They can also help identify risks in the project. A line drawn leading to major defect is the main focus. You drawing groups of causes along the top & bottom of the line The leading lines are drawn slanted. The overall diagram then looks like a fishbone. You can take any set of causes and expand the analysis by asking a series of how or why questions. Such “how-how” or “why-why” detailed diagrams will help identify the root cause easily.

Control charts that plot a series of desired values that also show a upper and lower acceptable limits are a good tool to determine if a process is within control. The chart also includes the upper and the lower control lines. These are tighter that the limit range and indicate if a process is showing undue variations or drifting one way that is a trend to exceed control limits.

Flowcharting is a graphic tool that clearly indicates the process steps and the decision points. The process and its behavior is clearly understood so that qc activities can discover either a failing step or opportunities for improvements.

Histogram is a vertical bar chart of how often a variable state occurred. For example, one may chart the frequency of different types of defects that occur. These may not be done in any particular order. But as they stand side by side the bars clearly show which the more frequent types of defects are. This can easily focus on investigating the reasons for the types that record a higher occurrence.

The Pareto chart or the Pareto diagram is also a histogram but drawn in a specific way. Staring from left you draw the highest bar first, then the next highest and so on. You also draw a line graph that shows the cumulative frequency as each type of defect is considered. You continue to add another bar graph until you have reached close to 80% cumulative % point. The bars included under the cumulative curve are the most significant causes of defect.

This comes from the Pareto principle that 80 % of problems are usually caused by 20% of the causes. We kept adding the causes and their effects until we accounted for 80% of the defects. These 5/6 or whatever cause we now have under the cumulative curve are the most significant causes cause most of the defects. You thus have arrived at clear issues to investigate further and determine how to eliminate them.

Quality control measures are to be carried out throughout the project. As you carry out these planned measures/measurements/inspections you need to record the outcomes. The activities under the quality control head helps you asses the performance and recommend necessary changes, if any. Like any other process you take a set of inputs, process them using suitable and recommended tools & techniques and then the output documentation reflects the state of quality achieved as well as any process changes that are considered essential.

The Inputs

Quality control processes are based of statistical theories thus the team assigned to work on quality control will need to have a good grasp of applicable theories. Theories related to statistical controls as they apply towards sampling and probability is the vital areas where adequate familiarity is needed. Usually such a group, typically, is named a QC department or something similar. Their job is to carry out the planned QC processes and identify any shortcomings in product or process quality.

There are some terms that are often confused by many people including QC personnel. So they’ll need to be very clear about the distinctions in the terms discussed here. Prevention and inspection is a pair of terms which can be conflicting in their implications. While prevention looks at ways to keep errors out of a process, inspection is the act of keeping errors away from customer deliveries. Thus the context is different though the main purpose is to prevent errors in the product/process/project delivered to the end customer. Both are part of the QC process. Similarly the distinction between attribute sampling and variables sampling need to be understood well. Attribute sampling is a kind of discrete sampling process. Results either meet or fail to meet the success criteria. Variable sampling, on the other hand, checks items on a continuous scale to decide degree of conformance. Another such pair of terms whose implications should be clear are tolerance and control limits. Tolerance represents variations that are inherent. The variation in the output item caused even when a process is completely under control and thus acceptable. The variation threshold that determines the threshold that lets you conclude the process is out of control.

The inputs that you would use for the QC process would include (but not limited to), Project management plan as the overall reference for everything. Quality metrics and checklists will let you check for no-conformance easily. Work performance measurements tell you if the project process is working well. Change requests are generated when any deviations are noticed in quality or process improvement opportunities are found. These approved changes form inputs that add value to the QC process. You are able to judge the overall process better by taking the original baseline and the intended changes together for a bigger picture. Deliverable is the target we are chasing through executing the project. There is always learning involved through all this and you need to capture that in updating the organizational process assets.

Tools & Techniques that You Need to Use

Tools & techniques used are based on statistical and probability theory and some graphic techniques. In totality they form a body of knowledge that has come to be known as quality theory. The tools and techniques include and control charts, Pareto charts, histograms, cause & effect diagrams, run charts, flowcharts and scatter diagrams are various graphic techniques that are based on statistics and mathematics. Flow chart being a graphic tool that makes for easier understanding of a process. Statistical sampling and inspections are tools that help make drawing conclusions about quality achievements easier. Approved change requests are reviewed to make sure there are no superfluous changes and also that nothing significant has been missed.

One of the major responsibilities of QC activities is to determine the root cause of problems seen and thereby decide remedial actions that show up as process change request. The graphic tools from the above list that help do this analysis are the cause & effect diagrams, histogram and the Pareto chart, the run chart, the scatter diagram et al. The control chart helps determine when and where the process is likely to break. Pareto chart is a great tool for focusing on the primary causes for problems. Most of these techniques are also helpful in QA processes and have been discussed in some detail in an article about the tools and techniques for the QA process. Also review the discussions on the tools & techniques as appropriate for the QC processes are discussed in later articles.

Quality control measurements are recorded. These records sort of traces what has been happening about quality achievements over time. Validated changes form inputs to necessary improvements. These may be about confirming if the changes requested have been implemented appropriately. Validated deliverables identify if the deliverables are appropriates as per verify scope and forms the basis for customer acceptance. Process asset updates adds the necessary improvements to the asset documents- Completed checklists and lessons learned are the important ones.

Change requests take care of the necessary preventive/corrective actions for the process and they need to be incorporated into the process. Change management process should control the implementation of changes. Quality standards and any other project documents that need update should be changed after the QC activities at a specific time point is done.

Techniques in the perform QC

These include cause & effect diagrams, control charts, flowcharting, histogram, Pareto chart, run chart, scatter diagram, statistical sampling, inspections and approved change requests reviews. Of these the techniques of control charts, flowcharting, statistical sampling has been discussed in part 1 and elsewhere so they’ll not be discussed again here.

Cause & effect diagrams, fishbone diagrams or the ishikawa diagrams are different names attached to the technique. Different types of potential causes are drawn on a line progressing towards a cause, such as a major defect. The diagram looks like a fishbone, leading to its head, the major problem. Root causes are then discovered by following a set of questions why-why or how-how kind of reasoning and bringing out possible causes. The two diagrams indicated in the PMBOK (figs. 8-12 and 8-13) illustrate the way it is used.

The histogram technique is to create bar charts of frequency of occurrence of some attributes of a problem. The reasons will have different frequency of operation. The attribute that happens most frequently is the most common cause of the problem being studied.

In the Pareto charts the histogram is a ranked one with the higher bars coming first. A cumulative frequency distribution also is plotted. So that it is immediately evident which 20% of the causes create 80% of the problems (as stated by Pareto in his law). That illustrates very starkly the causes to focus on.

Run charts are the control charts without the control limits displayed. It is simply a record of how things are going; so that mathematical trend analysis etc can be carried out of the curve that shows up. The analysis looks at historical data and attempts to predict the future trend. Trend analysis can be used to monitor technical performance as well as cost and schedule performance. You could deduce how many errors/defects have been identified and how many of them remain to be corrected. For cost and schedule performance you could look at how many activities per period were completed with noticeable variations. One could see from the graph, qualitatively what kind of variations are happening, if deviations are increasing or decreasing etc.

Scatter diagrams let you confirm if correlation exists between two variables. When plotted against each other, if the diagram resembles values that are scattered all over the graph area, then they are not related. Whereas a diagram that is close to a diagonal indicates strong correlation between the two variables.

Inspection is a technique of taking measurements and determining if an item does conform to a specified quality level. Inspections could be carried out to determine root causes too.

Review of change requests ensures that the requests have been studied thoroughly and no superfluous changes are carried out. Quality audits are also reviews that are used heavily to do a structured and independent review. Audits help identify best practices being used, gaps/shortcomings, share best practices from elsewhere, help the team to achieve higher productivity as also improve the lessons learnt knowledgebase of the organization.

Lastly, process analysis is another important tool to help identify process improvement opportunities and if some non-value add step could be removed altogether. Process changes necessary to correct and prevent errors are also possible.

Quality Planning

Tools & techniques used during plan quality phase include cost benefit analysis, looking at the cost of quality, control charts, benchmarking, design of experiments, statistical sampling, flow-charting and so on. There are some well known quality processes and some may have been adopted at your organization. Techniques from such processes are useful too. These may be the six-sigma or the quality function deployment (QFD) related techniques. Additional tools used would include brainstorming, affinity diagrams, force field analysis, nominal group techniques, matrix diagrams and prioritization matrices. These techniques have been discussed in details when discussing the quality plan processes. We shall touch upon them briefly here.

Benefits of quality are that of less rework, higher productivity and increased satisfaction with the product/ project/ process item. This is offset with the cost of quality steps needed against the money value of the benefits. Cost of quality is derived by the balance of the cost of conformance versus the cost of non conformance. The cost of conformance includes the prevention costs and the appraisal costs as against the non-conformance cost that includes the internal failures and external failure costs. Essentially is the difference in costs to achieve a level of quality versus the warranty and support costs that will be required to meet the failures due to low quality.

Control charts help in monitoring a process through a set of measurements, including on a sample population, to control the underlying process. The control chart will usually have a set of desired values which are bounded by the upper and lower most variations acceptable. Control lines that are better than the worst variations are used to ensure the limits are never reached. Whenever there is undue variation or a drifting trend, the process needs a tune up.

Benchmarking helps establish good/best practices by comparison to other such best practices while design of experiments establishes what to expect when some significant parameters are varied in a designed manner.

Statistical sampling is the technique of drawing a small population of items, do measurements on them and draw conclusions about the larger population. Statistical theory basis for the whole way of doing the analysis are available and the quality people will need to be aware of the foundations.

Flowcharting is a technique of describing a process graphically so that conclusions about it can be drawn from the understanding derived from the flow-chart. It describes process steps and decision points to represent the process.

The quality assurance or the QA activities are run in conjunction with process improvement activities. Any process improvement that reduces waste and makes the process a little more efficient cannot be overlooked. Processes must be run at the most efficient and effective state given the state of knowledge available currently and in the combined wisdom of not only the project management team but also all the other interested stakeholders as well.

The QA processes are typically run by a specialized QA department. It is also usual to have the department report to authorities to whom the project manager reports. That ensures that not only they are not under the same deadline pressures as the project team as also project manager would not be able to force the QA to accept unless he is satisfied. Thus they would not be tempted to cut corners. The department’s first commitment is to enforce the QA processes that include standards and process control defined in the QA plans. This direct reporting also demonstrates the commitment of higher management, that they are committed to fulfill the quality requirements fully.

The Process

The structure of the QA process itself can be described the way all the other processes in the PMBOK are described. This is in terms of the inputs specific to the process, what tools & techniques are used to process these inputs and what are the outputs generated out of the complete process.

In those terms the inputs that this process uses are the project management plan, quality metrics, work performance information and quality control measurements. The project management plan, as always, is the mother of all related plans so that it is an absolutely necessary input. The QA plan and the process input plans provide you the necessary QA process overview and the process improvement plan provides you the necessary analysis processes to detect improvements.

Quality metrics are the dimensions that define the overall quality to be achieved and the work performance information tells you how much of that is actually achieved. The dimensions of work performance include the numbers indicating technical performance measures, project deliverables status, schedule progress and costs incurred. The performance as measured by the QC measurements is what tells you exactly where we are. QC measurements are the documented results of quality control activities. The formats, usually, would have been specified in the quality planning phase.

The outputs, the documents, generated out of all this processing include projects management plan updates, change requests, organizational process asset updates and other project document updates. Change request take care of the changes to be made to the project activities, quality activities etc. or even a change required in one of the deliverables. If there are any changes required in the project plan as a consequence of these changes they need to be taken care of too. The process assets updates take care of any knowledge that has been generated and are captured in the process documents.

Quality audits and process analysis are the two major techniques in use during the Perform QA process. The tools and techniques used during the plan quality and perform quality control processes.

Quality audits are geared towards identifying the good and best practices being used, identifies all the gaps and/or shortcomings, sharing good practices used in earlier projects, offering assistance to project team to improve implementation of processes and improvements in productivity and updating the process assets through the lessons learnt from such audits. Audits will have to be very structured to be most effective. The review also will have to be independent of the project team to help identify whether project activities are following the policies, processes and procedures as dictated by the organizational as well as project policies.

Correction of deficiencies arising out of audit results should reduce cost of quality and result in increased satisfaction of the stakeholder/sponsor and/or customer. Audits can be used to confirm that earlier changes requested are actually implemented whether for improvements or preventive/corrective actions. Process analysis attempts to identify problems their root causes, problems and constraints experienced. More importantly if a non value added activity has been discovered that can be eliminated.

Process improvement plan is part of the overall project management plan. That the process quality can take care of the quality of the output and continuous improvement is necessary to zero in on quality targets; are well known principles. Thus a plan for improving the process in use has to be constantly looked at for opportunities of improvements possible.

Process Improvement Plans

A typical process improvement plan documents the ways to analyze a process and mark the areas/activities that can improve the process. The areas that should be considered include process boundaries, process configuration, process metrics and targets for improved performance.

Process boundaries are issues that define the processes such as the purpose of a process, how it begins and ends, the inputs/ outputs, data required, the process owner and the involved stakeholders. This is the basic requirement. The items involved with this define the process as it exists now. So that forms the baseline in any improvement activities to be planned. Like any other innovation generating, brainstorming activities, these existing items should be able to trigger thoughts about, changing, modifying, reducing, increasing, changing in some way and any other variation you can think of, to generate what can make improvements to the process happen.

Process configuration is a graphic depiction of the process. It lets you illustrate the interfaces with other processes too. Being graphical this lets you visualize the process accurately, particularly as you are able to take the interfaces into account too. This understanding along with the process boundary information will help with analysis to be as comprehensive as possible. Process metrics and their respective control limits let you analyze the process efficiency accurately. Targets for Improved performance lets you decide the targets to chase in the process improvement plans.

Armed with these details one can embark on the process improvement planning. A well known methodology is the GQM or the goals, questions and metrics. This is discussed here only as an example. This is not the only method available but gives you a general understanding of what needs to be done in creating a process improvement plan. It is one of the simpler methods and may not be the most effective.

To have meaningful measures an organization needs to establish goals. You then link the goals to data; the data that gives operational foundations for measurements. The outcome of a GQM approach is a set of identified measures directed to specific issues. The measure could be conceptual or goals, operational or questions and quantitative values that are metrics. Goals typically would include products, related processes and resources. Attempts at setting the goals should be able to define the objects to be considered, purpose of the analysis and the focus or the quality issue.

Questions raised should be to put the object identified in goals in context of specific viewpoints. The questions may help additional measures necessary to identify answers to the questions. Metrics are the associated set of data that help provide an answer. The metrics could be subjective or objective. The answers related to the necessary changes then need to be implemented to improve the process.