CW PMP Reading-Sample
CW PMP Reading-Sample
CW PMP Reading-Sample
CHAPTER 2 PREVIEW
• Distinguish among predictive, iterative, incremental, and agile project life cycles as
well as hybrid approaches.
• Discover how to split projects into phases to enable incremental value delivery using
any type of project life cycle.
• Explore what is done during the overlapping Process Groups of Initiating, Planning,
Executing, Monitoring and Controlling, and Closing.
• Understand the Agile Manifesto and the principles that underly agile project
management methodologies.
• Discover the benefits of learning about unfamiliar project methodologies.
This chapter addresses the basics of the project life cycles, project phases, and
Process Groups. Topic 1 introduces project life cycles and phases. The second and
third topics address life cycles in more detail. Topic 2 presents core concepts of
predictive projects, including introducing the Process Groups, while Topic 3 presents
the core concepts of agile and hybrid projects.
Note that this chapter is intended to give you just enough information to put the
discussions in this and the next module into a project management context. The more
detailed discussions of project management processes are reserved for Modules 3
through 5.
Most projects, even those with just one phase, are typically planned and executed
using what is called progressive elaboration. For this reason, this topic starts with a
discussion of progressive elaboration. We’ll then look at project life cycles and
project phases in more detail.
Progressive Elaboration
The PMBOK® Guide defines progressive elaboration as “the iterative process of
increasing the level of detail in a project management plan as greater amounts of
information and more accurate estimates become available.”
You need to start planning somewhere even though you know information from other
planning areas will influence the part of the plan you are working on. For example, when
you are working on the initial schedule and cost estimates, you haven’t yet worked on the
quality plan or the risk register. Quality control activities and risk responses take time and
money, which will either directly impact the schedule and budget or take the form of
reserves. Therefore, you will go back and revise these initial plans as you go through the
various planning steps.
You will not start out with perfect estimates. As the relevant events get closer or earlier
phases are completed, a more detailed level of information will be available and estimates
will be more precise. Since the phases of agile projects are typically very short, the
progressive elaboration in this case is meant to occur between each phase. (Changes to
plans during a phase are actively discouraged.)
Project managers view plan refinement as a natural part of planning that allows more
precise baseline specification, measurement, and management as the project evolves.
• Cost and staffing curve. In general, costs and staffing for a predictive project life
cycle are low during the earlier Process Groups—Initiating and the first pass at
Planning—because few team members are on board and the project is using fewer
resources. Costs increase quickly during the later Process Groups of Executing and
Monitoring and Controlling, when most of the project work is being done. Costs
begin to decrease at the end of Executing and Monitoring and Controlling as
deliverables are completed and resources are released. Costs decrease more quickly
during the final Process Group, Closing. On an agile or hybrid project, early phases
may likewise have lower costs if teams start small, and costs could ramp up if the
work shows promise and additional resources are devoted to the project. Teams may
also be kept the same size, in which case the cost and staffing curve will be more flat
and, as a side benefit, predictable. Stable teams could be small or large (e.g., to
represent many cross-functional interests).
knowledge and experience increase, and steps can be taken to manage risk more
effectively. For example, a predictive project aiming at producing a new type of
manufacturing equipment has more risk during Initiating and Planning, because the
sponsor and the project team don’t know how or if the objectives can actually be
achieved in proposed designs and if the designs can be translated into production
within scope, budget, and time frames. Once a project is accepted by the customer, no
further risk exists. On an agile/hybrid project, as phases are completed and their
deliverables are accepted, the level of risk and uncertainty gradually is reduced.
• Cost of changes curve. For a predictive project, the least expensive time to change
scope or product characteristics is early on, before work has been done and resources
spent. As the project work progresses, the ability to make changes without
significantly affecting cost and schedule (and possibly other measures such as quality
and project team motivation and engagement) decreases. Changing course may mean
that investments made as a result of earlier decisions have been wasted. Work based
on a mistake or incorporating a flawed element has to be redone. Changes may have a
“ripple effect” on work that has already been completed. A team working on new
manufacturing equipment could handle this issue by proving their design concepts
through a prototype test before proceeding to building the equipment. (Prototyping is
an example of an agile mindset.) This would decrease the risk of later changes to
project baselines. Agile, by design, works to keep the cost of changes lower
throughout a project so that changes can be accommodated even late in the project.
The relationship between the anticipated curves for risk and the cost of changes (for a
project on the predictive end of the spectrum) is illustrated in Exhibit 1-3.
Exhibit 1-3: Risk and Uncertainty Decrease, but Cost of Changes Increases as Project Progresses
Source: Project Management Institute, A Guide to the Project Management Body of Knowledge (PMBOK®
Guide)—Sixth Edition, Project Management Institute, Inc., 2017, Figure 1-3, Page 549. Material from this
publication has been reproduced with the permission of PMI.
Not all projects are alike, however. In some projects, costs are front-loaded or rise
quickly—for example, when a project requires very expensive architectural planning up
front. In some projects, uncertainty is very difficult to manage. For example, a new
building may include features that have never been built before or materials that have
never been used in construction of this type. To apply heightened levels of managerial
oversight, organizations can divide complex predictive projects into more phases, as is
discussed at the end of this topic, or they can use iterative, incremental, or agile life cycle
types, as is discussed next.
Agile Dynamic Repeated until correct Frequent small Customer value via
deliveries frequent deliveries
and feedback
Source: Project Management Institute, Agile Practice Guide, Project Management Institute, Inc., 2017, Table 3-1,
Page 18. Material from this publication has been reproduced with the permission of PMI.
Each of these project life cycles may manage uncertainty by using rolling wave
planning. This is a planning technique in which the work to be accomplished in the near
term is planned in detail while future work is planned in a more general manner. The first
wave of project work appears on the horizon, and the team busily prepares to manage it.
While that wave dissipates, the team turns its attention to planning its response to the next
wave. It can see this wave more clearly now, and it may understand how to manage it
better because of the experience and deliverables generated during the previous wave.
Agile projects in particular rely on rolling wave planning between phases.
Note that the four project life cycle types work on a continuum, so the iterative and
incremental types tend to overlap with other life cycle types, as shown in Exhibit 1-5.
Source: Project Management Institute, A Guide to the Project Management Body of Knowledge (PMBOK®
Guide)—Sixth Edition, Project Management Institute, Inc., 2017, Figure X3-1, Page 666. Material from this
publication has been reproduced with the permission of PMI.
Predictive
In a predictive life cycle, also called a fully plan-driven cycle, “the project scope, and the
time and cost required to deliver that scope, are determined as early in the life cycle as
possible.” A predictive cycle can also be called a waterfall cycle, because each step flows
down to the next without repeating prior steps. (However, progressive elaboration will
still occur during planning.) The ability to develop detailed plans in advance of execution
allows the project manager to predict an unfolding sequence of activities that will take
place in each phase, therefore avoiding risk and reducing costs. One can associate
predictive life cycles with the image of a falling line of dominoes. If you’ve seen a falling
dominoes display, you know that individual dominoes are first carefully laid out in an
exact pattern. The “domino master” (or project manager, for our purposes) tips the first
domino. Unless there has been a fault in planning or some unexpected and uncontrollable
external force intervenes (imagine the effect of an earthquake), the dominoes will fall in
the anticipated order and produce a satisfying result. (Clearly, a project manager needs to
stay more involved than this, but it is just an analogy.) Similarly, in a predictive project
life cycle, the phases unfold and develop in the anticipated manner. Complex domino
displays and projects with predictive life cycles can also unfold in sequential and
overlapping phases.
Each phase focuses on work that is essentially different from that in the preceding phase
and may involve different team members. The deliverable is not in a form acceptable to
the customer until the end of the project.
Although the scope is well-defined at the beginning, planning for later phases may not be
as detailed. As the project life cycle proceeds, rolling wave planning may be used to plan
the next phase in detail. Scope changes are controlled carefully, however. Changes in
requirements that occur in late phases can create significant changes to a project’s
schedule and budget. As a result, any changes to the project scope are carefully managed.
Predictive life cycles are best suited to projects that have easily defined, well understood
deliverables. Think about the manufacturing of a new school bus model. Although the
project may include some new engine or dashboard technology and new kinds of seats or
seating configurations, the customers know the requirements for a school bus and the
suppliers know what it takes to manufacture one.
Source: Project Management Institute, Agile Practice Guide, Project Management Institute, Inc., 2017, Figure 3-2,
Page 21. Material from this publication has been reproduced with the permission of PMI.
the project scope is generally determined early in the project life cycle, but time and
cost estimates are routinely modified as the project team’s understanding of the
product increases. Iterations develop the product through a series of repeated cycles,
while increments successively add to the functionality of the product.
The reason this life cycle is called iterative is because the repeated cycles are the key
element used to drive value by ensuring that this is the correct solution. While there
are increments, the increments are not distinct deliverables but rather features or
refinements of a single deliverable per phase. Repetition (or iteration) of one or more
project activities deepens the project team’s (and possibly the customer’s)
understanding of the product, thereby allowing them to progressively develop plans to
avoid risk. Rolling wave planning may be used to add detail to plans for the next
iteration. A project with one or more prototypes but only one final deliverable is an
example. The prototypes could be used to add functionality or take the project in an
unexpected direction.
an adaptive project life cycle in which the deliverable is produced through a series of
iterations that successively add functionality within a predetermined time frame. The
deliverable contains the necessary and sufficient capability to be complete only after
the final iteration.
Another advantage of iterative or incremental life cycles is that the project manager can
plan the next phase while the project team executes the current phase. In this way, only a
high-level vision of the scope of work needs to be determined initially. Elements of the
plan may be only tentatively assigned to a particular iteration. However, once work
begins on an iteration, scope changes to that iteration are subject to integrated change
control.
A project can be iterative or incremental and still be closer to the predictive end of the
spectrum. This might be the case if the project phases are long and each phase involves
detailed planning based on feedback gained from the prior phase, including the client’s
reaction to the partial deliverables received so far. When a project is iterative and
incremental and has other features like very short phases, then it is likely an agile project.
Agile (Adaptive)
While an iterative life cycle is a little incremental and an incremental life cycle is a little
iterative, an agile life cycle, also known as the change-driven or adaptive method, is
equally iterative and incremental. Both are vital to success. The increment to produce is
approved before the start of a given iteration. The iteration is used to minimize
requirements uncertainty and the potential for plans to get rapidly out of date by only
planning for the upcoming iteration in detail. Agile life cycles differ from iterative and
incremental life cycles that are closer to the predictive end of the spectrum in that
iterations are very rapid (usually two to four weeks in length), which allows this approach
to address risk and costs as issues emerge. The agile approach is intended to facilitate
projects that expect a high amount of change. It requires a high degree of ongoing
stakeholder involvement.
Planning for a single iteration means that planning is a regular event rather than being
addressed up front. When the team meets in their iteration planning session, they
determine how many of the highest-priority requirements can be completed in the current
iteration. Projects with agile life cycles collect ideas for improvements and new features
and feed them continually into the next iteration. Predictive projects with some degree of
iteration will collect these ideas throughout the project but hold the most disruptive of
them for a new, subsequent project—version 2.0.
Exhibit 1-7 compares a waterfall (predictive) project, on the left, with an agile project
that includes features of both iterative and incremental life cycles, on the right. The path
of the agile project is like a series of loops, so it is iterative. Iterative projects can be
thought of as a spiral that starts with a central idea at its core that gets broader and more
complex each time it loops around through the Planning, Executing, and Monitoring and
Controlling Process Groups. For example, testing conducted during Executing and
interpreted during Monitoring and Controlling can influence the next round of Planning.
The project also produces valuable deliverables or results in increments after each phase.
Because iterations are brief and feedback is gathered continuously, the agile life cycle can
handle change even late in the project in a cost-effective manner.
Exhibit 1-7: Waterfall (Predictive) Life Cycle versus Agile Life Cycle
The exhibit shows how project phases intentionally iterate or repeat certain activities as
the project team develops a deeper and deeper understanding of the product while
incrementally releasing better and better deliverables or new deliverables per phase. An
example could be a billing system for a large utility company. The first phase could
provide the company with a system with core deliverables such as the framework plus
basic functionality. Subsequent phases could make the system more user-friendly or
decrease transaction times or increase the number of interactions that end users (the
company’s customers) can perform. Note that the example shows just one Initiating and
one Closing Process Group, but these could also be repeated in each phase. (These
Process Groups are explained more in the next topic.)
The agile cycle is best for situations when scope and requirements cannot be determined
or must remain flexible to adapt to changing enterprise environmental factors such as
competitor actions. Agile projects use a product backlog to control scope and prioritize
activities to be performed in a given iteration.
There are two very different approaches that agile frameworks can take:
that the same resources are used in each phase). The iteration is done at the end of the
set time period, meaning that while the plan is to start and complete all work (by
selecting just enough work to likely fill the iteration), some planned tasks may not get
done and will need to be completed in the next iteration.
• Flow-based agile. In flow-based agile (such as the Kanban Method), the process is
still iterative, but work is not restricted to a particular iteration cycle. It is called
flow-based because there is a continuous flow of work as features are started/
completed on demand rather than new work being added only at the start of an
iteration cycle. The primary control is instead a strict limit on how much
work-in-process (WIP) can exist at any time. The limit is typically fairly low to help
identify issues quickly. As a task or feature (a deliverable that represents a chunk of
value to the customer and may require completion of several tasks) is completed, new
capacity is created for the team to add another task/feature. The team convenes
whenever there is available capacity to discuss what to add next. The plan is to
always refill the WIP limit to its maximum as long as there is work to be done. For
example, if a team is working on three features at a certain point and one feature
(including testing) is completed, then the team meets to select and add another feature
to fill up the queue even though the other two features are still being worked on.
Exhibit 1-8 illustrates an iteration-based agile life cycle. The example shows Scrum, an
agile framework. After the project is initiated, the team meets regularly in a series of
sprints, which are often as short as two or three weeks. The team plans and estimates for
just that period and then implements that plan. The first half of the regular Scrum meeting
for a sprint is dedicated to replenishment (a retrospective of the prior sprint). The second
half of the meeting is devoted to planning and estimating for the upcoming sprint. The
plan is executed between meetings, and the process is repeated until the final deliverable
is ready for release. Each sprint results in a more developed product with more features or
better functionality.
Hybrid
A hybrid life cycle is a combination of the predictive and agile life cycles. Instead of being
assigned to the whole project, predictive and agile processes are assigned to different
elements within the project, based on their qualities and requirements. It is up to the project
management team to determine which is the appropriate life cycle for a project and to
ensure that it has the requisite flexibility to encompass all project elements.
Project Phases
Projects can be divided into phases to control the work and to integrate the project’s final
product, service, or result into the ongoing operations of the organization for which the
project is being performed. With the exception of iterations for agile projects, a phase is
usually a major subdivision of a project.
One or more project phases comprise the project’s life cycle. Simple predictive projects
often have only one phase. In a predictive project with multiple phases, the phases could
each occupy a fairly long duration, such as several months, but each phase will produce
something that can be delivered to the customer. In an incremental or agile project, there
are almost always multiple phases, but these phases will often be very brief, such as a week
or two.
Phases are always appropriate for an agile project. When are phases appropriate for a
non-agile project? Apply the following criteria to decide:
• There is a distinct goal or objective to meet in the phase that can impact later phases.
• The work to be done in a phase is mostly or fully completed in that phase, resulting in
a distinct deliverable.
• Deliverables are produced and turned over to customers before completion of the entire
project.
• One of the deliverables has a different delivery date than the others.
• The people, organizations, or resources for the phase differ from that for other phases.
(When the work required in each phase differs significantly, a good portion of the
project team will also differ from phase to phase. For example, one phase may involve
architects while another may involve electricians and structural engineers.)
• Unique processes or controls are needed in each phase.
Each phase has its own full or partial set of Process Groups. For example, during a
feasibility phase, there would be an Initiating phase to authorize the feasibility study,
Planning to determine feasibility scope and so on, and then Executing, Monitoring and
Controlling, and Closing, resulting in a completed feasibility study. If the results look
promising, then the next phase would begin. Other phases could include concept
development, milestone reviews, or examination of lessons learned.
Phases can be identified using a variety of attributes. For example, a simple name or
number (Phase A, Phase 1, Sprint 1), a duration (Week 1, Month 1), or a resource
requirement (which building it will take place in, specific equipment that is needed) can be
used. Entrance and exit criteria can also be used to identify project phases, such as the
completion of certain deliverables, including any required documentation and successful
quality or integration test results.
At the end of a project phase is a phase gate. The project manager and the team will assess
their work and progress. Sponsors or product owners will analyze accomplishments, the
ratio of benefits to costs, and changes in risks or the organization’s own strategic priorities.
The decision will be made to continue with or cancel the project. If necessary, remedial or
corrective action can be taken before beginning the next phase in the life cycle. Different
organizations call this point by different names, for example, milestone, kill point, stage
gate, or phase review.
What are the phases of the project life cycle? The answer is, it depends. There is no
consensus in the project management profession or even in a given industry.
Organizations select or develop their own life cycles and phases and may choose different
ones for different types of projects. (From this perspective, an organization’s guidelines
about how to structure a project are an example of an organizational process asset, or
OPA.)
With sequential phases, the next phase begins only after the prior phase is
complete—deliverables are accepted and the phase has been closed. Sequential phases
can be used when a deliverable is needed as an input to a later phase, when additional
staff or resources are unavailable, or when risks need to be minimized. When using
sequential phases, there are fewer opportunities to compress the schedule.
With overlapping phases, work in one phase may overlap another phase or be done
simultaneously. This is possible if the teams, resources, and deliverables are independent
of one another or only partly dependent on one another. Overlapping allows project
managers to use schedule compression tools such as fast tracking (simultaneous
activities). Overlapping may require increasing the size of the project team and resources.
It increases some risks, primarily the risk that progressive elaboration in an earlier phase
will lead to rework in a later phase. Another risk is added complexity due to the need for
greater integration of the work.
An agile project with a fixed project team will typically use sequential phases because all
the team is occupied on the same iteration. This makes sense because when a project is
subject to high degrees of uncertainty and possible change, the end of a sequential
iteration presents a point at which adjustments can be made, before the project moves to
the next phase. Agile projects can use overlapping phases, but this will generally be
possible only on larger agile projects with multiple small project teams. Each subteam
will perform tasks on its own, and the overall work will need to be integrated. This
maximizes the amount of time spent on execution and reduces management overhead. An
overall integration team can then focus on overall improvements as opposed to just those
defined by the scope of an individual iteration.
CASE STUDY
Earlier we introduced Quanta Pharmaceuticals. They are initiating a project to
develop a new anti-scarring agent currently called B71, a drug for use in hospitals and
private practices such as for plastic surgery. Later parts of this case study will discuss
the rationale for the project and how the project fits into the strategy of the
organization, but here we will just discuss the project’s phases and life cycle choices.
This project will have a number of distinct phases. Pharmaceutical development is
extremely complicated, so we’ve simplified it somewhat and condensed Quanta’s
project into eight generally overlapping phases:
• Preclinical investigation and testing
• Submission
• Creation of product profile
• Formulation and packaging
• Clinical investigation
• Construction of manufacturing facility
• Preparing for launch
• Launch
Because of the nature of drug development and the need to fulfill specific regulatory
requirements at stage points, a predictive project life cycle will be used for the
majority of these phases. A few of the phases, however, will use an agile life cycle. In
our case study in this Learning System, we will focus on two of these phases,
preclinical investigation, which will use an agile life cycle, and testing and construction
of the manufacturing facility, which will use a predictive life cycle to illustrate the
predictive project Process Groups, from Initiating to Closing.
The Gantt chart in Exhibit 1-9 shows the phases proposed to management along with
their associated milestones. Note that the chart hides the activities within each phase
and focuses primarily on the phases and their relationships.