Feasibility Study
A study leading to a preliminary report to the end user to be advised on:
Technical Practicality, Cost Effectiveness, Time Scale, Budget.
To provide information required to support a decision to proceed
Stages of Program Production
- Feasibility Study
- Analysis
- Design
- Implementation
- Testing
- Changeover
- Maintenance
Abstraction
removing layers of complexity to reduce the problem to essential features
Analysis Stage
Investigation into a problem and how to solve it
- identify the main requirements
- identify IPSO from system
- describe system using DFD
- develop ER diagrams
- identify constrains
- specify hardware/software required
Decomposition
breaking down a problem into smaller sub problems
Interview
- Suitable when requiring lots of info from few staff
- choose key stakeholders
- Gather large amount of information.
- Judgements on validity – Body language
- Analyst can respond and ask follow-up questions to get detailed responses about key parts
- Time consuming and expensive
- Need trained interviewer
- Difficult to analyse large amount of information
- Difficult to analyse large variety of information
Questionnaire
- Cheap to produce for large number of people
- Distributed worldwide
- Completed online
- Need to be designed by experts
- May be ‘too busy’ to complete
- May provide false answers
Document Inspection
- Find out current data storage requirements
- Can see data flow and document trail throughout, from start to finish
Observaition
- Shadow emploees
- Gather information first hand
- Confirm findingings from other modes of investigation
External Entity DFD
Entering Data (2 Boxes)
Process DFD
Calculation, sort, filter (rounded box)
Data Store DFD
Provides data to/from a process (box with no right line)
Data Flow DFD
Must be labelled with data, direction of flow of data (arrow)
Design Stage
Design team will create accurate technical documents to instruct the programmers.
Implementation Stage
Programmers consider:
- Level of language
- Type of language
- Choice of language
-IDE
- Translation method
- Rules for writing/debugging
Alpha Testing
Testing carried out in house by employees to test functionality and takes placebefore beta testing
Beta Testing
When software close to finished product is released to an audience outside of the company and their comments are recorded
Acceptance Testing
Testing performed by the customer in their own environment
Direct Change Over
- Sudden change to a new system
- Simple but dangerous
- Used if there is no old system or is it broken/unusable
- All new systems liable to have bugs
Phased Changeover
- System is replaced feture by feature
- Projects are often split into modules
- Replacing modules of old system with new system
- User can get used to new system gradually
- Training done in stages
- Difficulties managed in one area can be in the next area
- Slower to get new system started
Pilot Changeover
- Whole system is replaced departmetn by department
- Large organisations tend to have more than one site
- Tested on a limited number of sites
- Whole system can be fully trialled
- Only part of the system is affected by failure
- Pilot staff can help train other staff
- Difficulties identified by one department can be fixed
- Slow
- No backup if failure
Parallel Changeover
- Both systems running together for a time
- Every action is duplicated on each system
- Safe, not catastrophic
- New system available to all
- IO can be compared
- Expensive as needs staff to operate both systems
- Confusion having 2 systems
Perfective Maintenance
When system functionality is improved/ enhanced
EG search time reduced
Adaptive Maintenance
When the system is changed to run on new hardware or software platform
EG switch from windows to linux
