Projects are practically broken down into different phases in order to break down the overall project objective into substages. A further aspect is to achieve a reduction in complexity and the opportunity at the end of a phase to check whether the activities planned for the phase have been completed in the scope of their objectives (costs, schedules, results). If this is not the case, then there is an opportunity to intervene at an early stage, thus avoiding early mistakes or omissions that could be carried through until the end of the project, possibly even increasing dramatically in effect.
Milestones
Each phase ends with a milestone. Milestones are decision points. A choice is made here between three options:
1. All previous activities are within plan, the phase may be concluded, the project may proceed as planned.
2. A number of activities that, according to plan, should have been concluded already, indicate significant deviations in the relevant parameters (costs, schedules, results). Reworking is required in order to complete the phase.
3. Circumstances have arisen that make the practical continuation of the project appear impossible, the project is stopped and, where necessary, discontinued completely or relaunched again under new framework conditions.
In addition to these key milestones it may be sensible to define additional milestones, depending on project size, at which key interim results must have been achieved. However, typically, milestones remain something unusual, the achievement of which may also require the mobilisation of additional resources.
Phase models
Individual project phases are self-contained periods within a project that end with a milestone and supply a key interim result. Simple phase models are built upon one another sequentially:
This simple phase model offers the advantage of being simple, concise and easy to understand. It is ideal for many small and medium-sized projects. The disadvantage in this respect is that it is constructed sequentially. This means that a new phase can only be commenced when the previous one is fully completed. In practice, this is often not possible, because it costs too much time and is therefore uneconomical.
Parallel phase model
In the parallel phase model account is taken of the fact that, in order for a project to continue, it is often sufficient for a large portion of the preceding activities to be completed. For example, it is possible for planning to commence even where the project set up is not one hundred percent complete. If, for example, all project team members are on board, there is space to work and the team is waiting for the joint project office, although this is not optimal, it does not mean that it is not possible to already conduct planning at this time.
This parallel approach can result in great time savings, making this phase model particularly suitable for complex projects that continue over a longer period, where it often offers the sole opportunity to maintain an ambitious and tight schedule.
The danger in this approach lies in commencing things too early without all preconditions having been met to a sufficient extent, with work continuing on an incomplete basis. The consequence of this is often significantly higher effort and expense than would have been the case with waiting.
A further risk lies in failure to ever complete the supposedly unimportant things. Initially because the concentration is already on the next stage, later because they have quite simply been forgotten. In practice, these are often unpopular tasks such as documentation.
Iterative phase model
In the iterative phase model various phases run one after another in numerous loops, thus achieving an improved result with each loop. In this, loops are completed until the required result is achieved. On the one hand, this approach costs a great deal of time, on the other hand it is often the only possibility of achieving an acceptable result in complex framework conditions that cannot be reduced down to a plannable scale.
The advantage of the iterative procedure is that it is possible to establish a continuous process of improvement and to achieve a truly improved product with each new loop. With the rising learning curve in the project it is also often possible to reduce the number of loops required as the project progresses.
The risk with the iterative approach lies in the lack of thorough documentation, that results are not adequately reflected and discussed, with the consequence that no continuous improvement process is established.