The Waterfall model in project management

Choosing the right project management approach is not always easy, as it often depends on the project, the company and the industry. One method is the Waterfall model, which is a linear process model divided into phases that are worked through sequentially to deliver a finished project.

What is the Waterfall model and how does it work?

The Waterfall model is one of the classic project management methods that assumes a defined end state is known at the start of the project and is often referred to as an “outdated method”. However, this view is inaccurate, as it does have its advantages, as will be shown below.

The model works with sequential phases that are planned down to the smallest detail, with a clear start and end point that flow into each other. In order to define these clear points, at the beginning of the project the exact requirements of the companies, customers or?stakeholders ?are identified. With this information, the content of each phase can be precisely planned and successfully delivered. This increases?efficiency , but also makes the process very inflexible and rigid. As the model is sequential, there is usually no feedback between the individual phases, i.e., the planned phases are carried out until the end of the project without any return. For this reason, it makes sense to use this method if few changes are expected. However, it has advantages and disadvantages, which are discussed below.

The phases of the Waterfall model

It is not only the result that is planned, but also the way to get there. Depending on the type of project, there may be a different number of phases. It makes sense to use as many phases as necessary, as the model needs to be planned as accurately as possible. However, it can be roughly said that a project can be divided into four to six phases, depending on the?type of project . These phases include at least:

• start,
• design,
• execution and
• completion.

Each phase can be of different length and can be designed individually. The example of an investment project in the life sciences sector illustrates this: The start phase (P0) includes the basic planning of the project, while the design phase (P1) includes the actual project design. The execution phase (P2 and P3) is characterised by the implementation of the project plan, and the final phase (P4) is the completion and handover of the project. Although these phases can be described as standard phases, it is useful to adapt the phase model to the project and make further subdivisions to ensure successful implementation.

At the end of the description of each phase, an example of an investment project is given to show how each phase could be designed. For example, a new plant is to be built for the production of a certain material.

Start

In the start phase, the first step should be to work out how and in what timeframe certain project results will be of use to the customer, i.e., an investment assessment, also called a business case. A project file should also be created, which is a collection of documents that serve as the basis for the project to be planned.

The next step is the stakeholder and?environment analysis . This involves systematically recording the interests and influence of stakeholders. Not all stakeholders have a positive attitude towards a project or have varying degrees of influence. Once this is known, each stakeholder can be dealt with accordingly. But the internal and external environment can also influence the project, for example through?communication with stakeholders . Therefore, it makes sense to link these two points together.

In addition to stakeholders, customers also have expectations of the project, which should be captured in a requirements analysis and prioritisation. It should also be checked whether these expectations can be met.?

It is helpful to have a complete specification, which contains all the necessary requirements from the customer and is given to the project manager. This should be as detailed as possible to enable accurate planning. Once the expectations and requirements are known, a risk analysis can be carried out to assess, evaluate and prioritise the risks.

In the event that problems arise during the course of the project, measures to prevent or detect errors should be defined in advance. To do this, it is necessary to plan the project as precisely as possible so that it does not happen in the first place. If it does, it can trigger a cascade that can lead to time delays and additional costs, which will be discussed later.

It is also important to repeat the?risk analysis , because even with the most careful planning, new developments can always occur.

Since the project is only tested at the end, change management should be established from the outset to know how to deal with any errors. Even with the most accurate planning, there will always be unforeseen developments that need to be responded to. It is particularly important that changes are clearly communicated to everyone to avoid duplication of effort.

This step also considers the tools, technologies and working conditions required to complete the project, as well as a documentation strategy. This addresses how documents relevant to the project will be managed. Documentation includes creating, identifying and recording, summarising, editing and updating, distributing, archiving and destroying documents. The underlying strategy includes decisions about how the documents produced will be stored and managed, and who will have access to them. The documentation strategy should ensure that all relevant information is accessible and understandable to the responsible person so that the documentation supports and does not hinder the project.

Example:

Phase P0, the basic planning, includes the project plan and the plant concept. For the planning, all people involved assume two months, with a budget of 40,000 CHF.

Design

The design phase is about planning the project. The first step is to determine what kind of project it is. This includes innovation projects, investment projects, organisational projects and mixed projects.?

The definition is important because depending on the type there are different approaches, and each has different risks or requires different documents.

The distribution of roles can be as follows: A project leader, who takes on many responsibilities for the duration of the project to make it a success. The project team, which may consist of a few or many team members who contribute to the success of the project through their required expertise. The composition of the project team may vary from phase to phase. For example, at one stage the team may consist only of developers, at another stage of engineers, and at the next stage of builders. And the steering committee, made up of members to whom the project leaders report.

The next step is to agree internally how the project will be organised. This includes the composition of roles and who has what authority. The Waterfall model is often associated with traditional project management, with three organisational structures. In the staff organisation, the project manager has a very limited function. He has no authority to issue directives and is limited to informal exchanges. In the matrix organisation, the project participants have two superiors, the project manager and the head of department, because the competences of the departments overlap. However, this form of organisation is only recommended for larger companies, as it allows for specialisation in several areas. Team members report to both the project manager and the head of department, so the information gets where it needs to go. The third is the autonomous organisation. Here the project manager is solely responsible for the project team, the implementation and the outcome of the project. He is ultimately accountable only to the steering committee. It is advantageous to have a defined communication structure, as the way of communication has a direct impact on the organisation.

Of particular importance is the definition of the project objective, which contains the result to be achieved and is a prerequisite for the fulfilment of the overall task.

The specifications prepared by the contractor are helpful as they contain all the requirements. Once all this is known, the implementation concept can be approved, and the project-specific phase can be planned. The individual milestones of each phase are defined and additional acceptance criteria for changing phases are set. It is also important to create a Work Breakdown Structure (WBS). The WBS is divided into sub-projects and work packages. Work packages are the smallest unit in the WBS and contain the activities required to complete a task, as well as an estimate of duration and cost. This estimate can be used to combine the costs of the sub-projects into the project costs in a bottom-up manner, or to allocate the available budget to the work packages in a top-down manner. By dividing the project into work packages and phases, a schedule can be created that leads to the critical path. This is the longest process chain and is a process without a buffer.

Example:

Phase P1, which represents the kick-off. In this phase, details have been carried out, building permits have been obtained and the budget has been discussed. This phase can take up to three months and has a budget of CHF 60,000.

Execution

In this phase, all the requirements listed in the specification must be implemented. These can also be divided into prototype and final product, followed by a test phase. This is used to check that the project objectives have been met and to gain knowledge for the rest of the project, or to gain experience for future projects and to create an action plan for future implementations. If changes are made because a phase could not be completed, the change should be justified and documented, as it will ultimately require the client’s approval. After all, the project stands or falls with them, so open and good communication about approval is important.

If errors occur, they should be evaluated and corrected. However, if the test went well, a product presentation can be made.

This should be followed by a project review with all stakeholders. This will assess whether the objectives have been achieved, whether there are any outstanding issues, etc.

Example:

P2, the implementation phase, and P3, the construction phase, which includes tendering and contracting, which can take up to three months, and implementation, which takes nine months. The budgets are CHF 100,000 and CHF 500,000 respectively.

Completion

Afterwards, the project is handed over to the customer with a product presentation, a joint review, acceptance of the project and lessons learned. It is evaluated what went well in the project, what did not go well and what could be adapted or possibly improved in a similar project. It is useful to do this together with the customer to get a different input.

If the product requires maintenance, this is also discussed.

Example:

Phase P4, the project closure. The project is handed over.

Advantages of the Waterfall model

Precise planning of each phase ensures a high level of planning certainty in terms of costs, duration and deliverables, as well as what exactly needs to be done in each phase. This results in scenarios that are easy to work through and clear for everyone to see, and also defines clear responsibilities.

As each phase must be completed before the next one begins, no work is left undone. This makes the method clear, intuitive and requires no special training. Documentation is easy to follow, as all information is documented, making it easy to trace each phase.

Disadvantages of the Waterfall model

A major problem, however, is the strict delineation of phases, as there is usually no feedback between phases, so changes can rarely occur. If they do, they must be incorporated into the project as early as possible, otherwise project costs will skyrocket because the project may already be too far advanced. This also means that the customer needs to communicate changes as early as possible. So, you need to be sure from the beginning that there will be no changes from the customer.

It is only at the end that you find out if a project is successful, and if it is not, the worst that can happen is that the project is extended, and the costs explode. In addition, in long-term projects, the final product may already be out of date, and an established change management process may be required to deal with the situation.

Examples for the use of the Waterfall model

As mentioned above, the model is particularly suited to projects where the requirements and processes are well understood, so that there are not too many changes as the project progresses. For example, these could be government projects such as building a new railway line or building a new school or even renovating it.?

Another example of projects that should not take too long because they cannot be changed is software projects, such as the development of a website, because any changes will not add too much time to the overall schedule.

Best practices when using the Waterfall model

Best practice means that certain methods are used to try to achieve an optimal result or to determine the best approach to a solution.?

The model already contains many approaches that are important for implementation. These include precise project plans with defined start and end points for each phase, risk management that is established at the beginning of the project and updated as necessary throughout the project, regular communication with all stakeholders to keep the project on track and everyone informed, and good documentation so that each step can be traced.

Conclusion

The Waterfall model is particularly suitable if you have a tight budget and want to know from the outset when the project will be completed. However, as we have seen, this is not always easy to specify, and delays can occur despite precise planning. This is especially true as errors are discovered towards the end of the project. Careful consideration must be given to whether this model is the right one for the project. If you want the strict separation of phases, but still need a little more flexibility in planning and execution, you can extend the model and add iterative loops to be able to correct errors immediately by returning to a phase if necessary.

Originally published at?https://www.iapm.net .