Software Development Life Cycle Models

Software Development Life Cycle Models

1. Sequential / Predictive

   • Waterfall model
     • One of the first formal SDLC models.
     • Linear-sequential
       • Follows simple structure of phases in a cascade
     • Clearly defined start/end points for each phase, making progress easier, reduces "slippage"
       • Costs can be estimated with a fairly high degree of accuracy once requirements defined
         • Test scenarios are already detailed in the functional spec of the requirements phase
     • Needs can be difficult to define, clients may change their mind once they see the end product
       • Not as flexible as other models.
         • Longer delivery time
   • Prototyping model
     • Development of prototypes which displays limited functionality and abstracted logic
       • Allows users to test and evaluate units of the final product as it is developed
     • Increased client / user involvement before implementation
       • Reduced times and cost as defects detected earlier
         • Missing, confusing or difficult functions can be identified early
     • Risk of insufficient requirement analysis / too much dependency on the prototype
       • May increase the complexity of the system as scope may expand beyond original plans
         • Effort invested into building prototypes may be excessive if not properly monitored

2. Iterative / Adaptive

   • Spiral
     • Combines iterative development with the systematic elements of Waterfall
       • Waterfall method with an emphasis on risk analysis
         • Incremental releases / refinement through each iteration
     • Changing requirements can be accommodated
       • Users see the system early and can feedback allowing requirements to be changed
         • Development can be divided into smaller parts; the risky parts treated earlier
     • Managements is more complex
       • Complex process
         • Excessive documentation
   • Agile
     • Collaborative effort of self-organising and cross-functional teams
       • Adaptive planning, evolutionary development, continual improvement
         • Rapid and flexible response to change
     • Reduced risk as output of devs is reviewed early and repeatedly
       • Highly flexible
         • Lots if communication between devs and clients, resulting in better understanding of requirements
           • More collaboration between functional teams, improving chances of success and higher product quality
     • Easy to understand, hard to implement properly
       • Less predictable
         • Requires a lot of collaboration / must be managed effectively
           • Less of a 'blueprint' of what the final product will be