What are project phases, the reasons for phasing projects and what is project lifecycle?
Project lifecycle is an important concept in project management. A project is any work that produces unique results in a certain timeframe. This work may proceed in phases for better management and control. For example, a long duration project may be broken down into smaller work of shorter duration. Take a city metro rail project for example. The first phase may consider one route within the city, the second phase another route, and so on. In other words, a series of phases may exist during start to end of a project. This series of phases that a project passes through from initiation to closure is known as project life cycle.
It is not necessary for multiple phases to exist; a project may consist of just one phase. In our metro rail example the phasing was done based on partial functional objectives, or intermediate results and deliverables per phase. The same example can also be treated as a milestone based phasing. Sometimes phasing is done due to finance availability reasons.
In our example, it was assumed that phasing was sequential, meaning the second phase was initiated only after the first phase was completed. This is known as sequential phasing. This approach reduces uncertainty due to doing one thing at a time, but also reduces the options to compress the project duration.
It is quite possible for second phase initiation and even planning to be done while the first phase is being executed. Naturally, this saves time. This is known as phase overlap. This helps to reduce project duration. Doing work in parallel (full overlap) in such a manner is also known as fast tracking. But this approach is prone to risk and rework because next phase work is started without getting full information from a previous phase that is not completed yet.
Note that activities within a phase will be logically related to each other so that certain deliverables or results are produced through properly coordinated work. So phase 1 of the metro rail project will include several related activities which will deliver the outcome of phase 1, say route 1. Activities within a phase, and therefore each phase, are time bound. Else the project as a whole will not progress. The same processes will be used in several phases. For example, planning processes will be used in every phase of a project.
Project lifecycle structure
The general structure of a project lifecycle(stages) as per PMBOKv5.0*is:
- Starting a project
- Organizing & preparing
- Carrying out the project work
- Closing the project
Thus the metro rail project in our earlier example will pass through these phases. Also, processes related to starting, organizing, performing work and closing may repeat in all phases. The project will be initiated prior to phase 1 or in phase 1. Phase 2 initiation will include a review of information of project initiation (project charter) which will contain phase 2 high level objectives. During phase 2 initiation, objectives of phase 2 may getupdated if need be, and then the phase 2 charter will be developed before commencing phase 2 execution.
PMBOK maps the above generic lifecycle structure to 5 process groups (process domains, or project stages namely initiating, planning, monitoring & controlling, closing). Due to such a lifecycle structure, it is natural for certain activities and processes to be repeated in various phases. All this depends on project complexity and size, organizational policies and project objectives. For example, an organization may consider feasibility study as pre-project work, the first phase of a project or a project in itself.
Each phase focuses on a different work aspect and can involve a different set of organizations, locations and skills. Unique controls are established to manage every phase. At phase end, deliverables are handed off or transferred from the project to concerned stakeholders. An assessment is carried out after which the next phase is initiated, or changes are made to the project, or the project is terminated. Phase closure is done based on approval. This phase end point is also known as stage gate, phase gate or kill point.
Product lifecycle is different from product lifecycle. Product lifecycle represents the progress or evolution of a product from its concept stage to delivery, further growth in the market, reaching maturity and eventually retirement or end of life.
Thus both phases and lifecycle stages are part of the project life cycle.
The two basic lifecycle types – predictive and adaptive
Project lifecycles are either plan driven (predictive) or change driven (adaptive). In predictive lifecycles, scope is defined at project start and changes are managed carefully. In adaptive lifecycles, product is developed in multiple iterations and detailed scope of each iteration is defined at the start of every iteration. In general, cost and staffing levels are lowest and risk is highest at the start in both types of lifecycles. However, while cost of changes and error correction is lowest at the start of predictive lifecycles, it is low throughout in adaptive lifecycles. On the other hand, ability to influence product without impacting cost is highest at the start of predictive lifecycles, and high throughout in adaptive cycle.
The type of lifecycle used in a project depends on the organization, industry, technology and other factors.
Digging deeper into the lifecycles, and the iterative and incremental lifecycle
Let us understand predictive and adaptive lifecycles in more detail, and also introduce the concept of iterative and incremental lifecycles.
In this type, project parameters like cost, schedule and scope are established early in the project, whatever practically possible. As applicable to any type of lifecycle, phases in the project may overlap or be in sequence; with work and skills differing in every phase. Once the scope is defined at the start, it is planned in detail and executed. Any change in scope will lead to re-planning. This approach is preferred when clients want a product or result at the end of the project which is delivered in full that is 100%. The product must be well understood in the beginning before planning and execution can begin. Predictive lifecycles originated in government projects where mandates were tight, deviations were not easily allowed and nature of work was long term and complex. Note that rolling wave planning can still be done wherein immediately forthcoming work must be planned in detail and future work may be planned in detailed later as and when sufficient planning information becomes available.
Not undermining the strengths of this approach, there were some disadvantages when applied to certain projects due to the nature of those projects. To list a few:
- Project plan once developed could not be changed
- Changes to requirements were not easy
- Lack of communication between project and business leading to dissatisfaction at the end of the project
Iterative and incremental lifecycles
Iteration is a repetition of activities which result in an output or deliverable. Using several iterations, several deliverables may be developed. These individual outputs may be part of a larger product. Output of each iteration may or may not be able to function on its own. Each phase of a project is an iteration. The same set of processes or process groups will be used in each iteration. Feedback from a preceding iteration may be implemented in the next.
An increment adds functionality. For example, constructing one operational lane of a highway. Then adding another operational lane, and so on. The same processes are used to construct each lane therefore iterative.
Without increments, each iteration may or may not add functionality. Without iteration, each increment may use a different set of processes.
The high level vision is developed initially along with high level scope of future iterations, but detailed scope is derived during individual iterations. Changing the scope of an iteration in progress must be done with caution.
Iterations may be in sequence or overlap partially. Planning for the following iteration may overlap with current iteration work. Team members may change between iterations as nature of work also changes.
This approach is preferred when scope and objectives are prone to change. Large and complex projects iterate and increment to reduce risk and to facilitate application of feedback in future iterations due to lessons learned. Also in cases where interim phase deliverables provide value.
These lifecycles have a high level of change and stakeholder involvement. These are also iterative and incremental, but iterations are rapid (2-4 weeks usually) where cost and time are fixed. A product backlog is created and maintained throughout the project, in which scope is decomposed into detailed requirements and work. Though the planning work is more during earlier iterations, each iteration or phase goes through detailed scope definition, planning and execution. Feedback is provided by the customer for every iteration’s deliverable(s).
Priorities are set at the start of each iteration, that is, the high priority backlog items that can be delivered in that iteration. At the end of an iteration, the product functionality developed so far are made ready for customer review. The backlog is updated based on customer (and/or sponsor) feedback so that current needs are reflected in it.
This approach is preferred in rapidly changing environments, or when scope/requirements are not clear at the start, or when even small increments deliver value through PMP Training.
Project lifecycles give a structure to a project and help to breakdown projects into phases if need be. Phases may be in sequence or overlap. Product lifecycles differ from project lifecycles. Project lifecycles are predictive (plan driven), iterative & incremental, or adaptive (change driven).The nature of projects and organizational factors influence the lifecycles.
*PMBOK is a registered trademark of Project Management Institute, Inc.