The spiral model of software development

Out of the many models of the software development life cycle, the spiral model is one of the most effective models of development. This model combines both the waterfall and the iterative models. This is a risk-driven model which outlines a process that is to be followed to overcome a given risk thereby ensuring an efficient and full-proof development process.

Like in all other SDLC models, this model also comprises of various phases and every phase has its own design goal and has a client review to gauge the progress. First mentioned in the year 1986 by Barry Boehm, in his paper, this software development approach is now being widely adopted by many organizations to ensure timely delivery of projects ensuring all last-minute requirements are met with ease.

Let us now look at what this model actually is in the next segment.

Phases of spiral model of software development

This approach of development begins with the collection of a small set of requirements as stated by the clients. These requirements are then made to go through each of the SDLC phases. In the process, the software development team adds functionality to each of these requirements with every increasing spiral until the application is ready to be migrated to production. This model is a combination of the iterative process and the linear development process that is the waterfall and risk analysis model. This model makes room for incremental release along with incremental refinement with every iteration through the spiral.

Coming to the various phases, this model has the following four phases.

Planning

This phase begins with gathering of business requirements to form the baseline of the spiral. And then in the next iterations or subsequent spirals, as and when the product is being created, identification of system needs, unit requirements, subsystem requirements, etc is completed in this phase. The planning phase also involves continuous two-way communication between system analysts and the customer to ensure all the requirements are met as asked by the client. The planning phase is the most important phase of any SDLC. In this phase since the system analysts are in touch with the clients continuously, it helps them in fine-tuning the requirements and getting better clarity on what the customer is expecting out of the product and how the system is required to be designed around those expectations.

In this phase, if you are a part of the development team, or a system analyst working with the client, it is important to ensure you maintain seamless communication with the client and if you are a client it is important to ensure you stay in touch with the system analysts and provide them with all the requirements in detail.

Risk analysis

In this phase, the potential risks that may arise while developing the system are identified. And then a risk mitigation plan or strategy is finalized around those risks. This phase is about estimating, identifying, and monitoring the management risks along with the technical feasibility. This includes slippage from the schedules or overrunning the costs. Along with the system risks, the risks involving the solution are identified and are included in the risk mitigation plan. At the end of this phase a prototype is released, that is risk-free and is probably the best possible solution.

When the risk mitigation plan or strategy is finalized, if you are a part of the development team, you need to ensure you have a plan B just in case your plan A does not work. Put in as many ideas as possible to mitigate the potential risks so that in case of risks you do not fall short of action points. Ensure that the

Engineering

This phase is the design phase of the SDLC. This phase starts with using the conceptual design or prototype as the baseline spiral. With the prototype design as the baseline, the solution design is being prepared which includes architectural, logical, physical, and final design in the subsequent iterations. After the design is concluded, then comes the coding, testing, and deploying of the solution. In this part, a finished software version is deployed at the client’s site for review, and feedback is sought on the first build or software version.

This phase is again one of the most important parts of the development. This involves designing, coding, and deploying the changes which means a lot of technical expertise in place. If you are a part of the development team, you must ensure the team that you have is skilled enough to tackle all challenges faced in this phase.

Evaluation

After the software is released to the client in the first iteration, the client then reviews the software and provides his/her feedback. The clients are asked to test all functionalities to ensure that anything wrong with the product can be identified. The testing team at the client-side records all their observations and feedback which is being sent to the software development team. This client feedback acts as the input or the baseline for the next iteration of the software which will be to further fine-tune the finished product or software based on the client evaluation or feedback provided.

Once the software is released after the first iteration and is ready to be tested, if you are a part of the development team, you will need to ensure you send a detailed test plan to the clients so that they can test the required modules and provide you with feedback. If you are at the client-side, you need to ensure you or your testing team has tested all functionalities thoroughly, so that relevant inputs can be provided for the next iteration.

Now that we have seen the various phases of this SDLC model, let us now look at the advantages and disadvantages of this model.

The spiral model advantages and disadvantages

The spiral model is known for its numerous advantages. Whereas, there are certain disadvantages as well. Here are certain examples of both.

Advantages:

This approach allows the addition of elements to the existing product once the product is ready without hampering the original product design.

• This model is also a consistent approach to various builds that lets transition and maintenance activities a lot easier.
• This approach requires the involvement of the client at an early phase, thereby ensuring the development work is in line with the client’s requirements right from the beginning.
• Additional functionalities can be added at later stages with every iteration.
• Since the prototype is made in small fragments in every phase, estimation of costs gets easier as you just add up costs for additional features with each iteration and do not start all over again.
• Since it is an iterative development method, potential risks can be effectively managed with every phase. Risky parts can be developed early which implies better risk management.
• This is one of the fastest approaches to the development of software and functionalities can be added gradually in a systematic manner with every iteration.
• Requirements may change mid-flight and this approach allows the accommodation of changes with ease.
• This model allows extensive use of prototypes as baselines for every iteration, thereby making development easier.
• In this approach, customer feedback at the end of every iteration acts as the baseline or inputs for starting the next iteration. As such, there is ample room for customer feedback which in turn means continuous improvement of the product.

Disadvantages:

• Management of the metrics and the plan in the spiral model can be more difficult.
• The project may go on in an indefinite loop as the client may never be satisfied, and the project end may never be known.
• This approach is not suitable for small or projects that involve fewer risks.
• The process of development is pretty complex.
• May involve a lot of documentation after every step – test plans at each iteration or when new functionality is added along with documentation of known issues.

Now that we have seen the advantages and disadvantages this approach may have, let us look at when this model can be used.

When to use the spiral model?

By now you know that the spiral model may not be fit for all projects. Here is a list of instances when this model can be used:

• When the size of the project is large and the project can be broken into fragments.
• When the product release needs to be very frequent.
• When a prototype of the product is required to be made.
• When risk mitigation and cost evaluation are extremely important.
• When the level of risk in the project is high.
• When requirements are complex and subject to change.
• When a long-term commitment to a project is not really feasible.

The bottomline

To conclude, and summarize, the spiral model is one of SDLC is one of the most effective software development models which help in a fast software development process and has room for continuous customer feedback and improvements.

Use this piece, to get a comprehensive idea about how the spiral model works and what are the risks and disadvantages involved.