Information Systems · Manual vs CBIS · SDLC Phases · Testing · Deployment · SDLC Models
Understanding what a system is — and how data flows through it — is the foundation of this chapter.
Every system — no matter how simple or complex — has exactly three basic components:
Raw materials, data, or information fed into the system to be processed.
The operations that transform the input into a useful output.
The final result produced after processing the input.
This is an information system — it takes raw data (numbers from sensors) and converts it into meaningful information (a weather forecast people can act upon).
Name of student and relevant year
Information systems are classified into two main types:
All processes are performed by humans — writing, filing, searching. No computers involved.
Uses a computer to transform data into information. Automated, fast, and secure.
Understanding the advantages of CBIS over manual systems is essential for exam questions.
Detect whether a book is available and its shelf location instantly.
Use a barcode to maintain book issues and returns efficiently.
Automatically calculate fines for late book returns.
Search for books efficiently by title, author, or subject.
Maintain complete records of everyone who borrowed books.
List students who have failed to return books on due dates.
Online reservation and searching for books from anywhere.
Provide digital e-books to library members.
Networked with school office — confirm library book returns when issuing leaving certificates.
| Feature | Computer Based System ✅ | Manual System ❌ |
|---|---|---|
| Errors | Minimal — computer programs reduce human error | More room for error — all processing done by humans |
| Processing Speed | Data processed more efficiently and quickly | Processing of data is less efficient and slower |
| Storage | Large amounts stored in small physical space; database software used | Requires large physical space; filing cabinets needed |
| Security | Secured with backups and passwords | Data open to many threats; less safe |
The SDLC is a structured process for developing an information system through six ordered phases.
Investigate the existing system thoroughly. Gather information from users and stakeholders. Define objectives, benefits, and efficiency targets.
Design the proposed system's architecture, user interface, database, hardware requirements, and test plans.
Write the actual program code using a suitable programming language. Keep code clean, readable, and well-commented.
Test each part of the system to find and fix errors in coding, planning, and requirements.
Install and launch the tested system. Choose an appropriate deployment approach (Direct, Parallel, Pilot, or Phased).
After deployment, continuously maintain the system to fix issues, adapt to new requirements, and adopt new technologies.
A complete investigation is carried out on the existing system to gather all necessary information before development begins.
At the initial stages of developing a computer based system, the existing system is observed as it is to gather information. This is important towards making accurate assumptions.
Face-to-face meetings with staff and users. Provides an opportunity to discuss matters related to personal attitudes openly. Useful to identify personal requirements and preferences.
A written set of questions given to users. The answers collected are analyzed for understanding the needs. Good for gathering information from many people at once.
Reports or files maintained in an organization under study are examined for information. Reveals existing data structures, formats, and workflows.
A model or prototype of the proposed system is developed and demonstrated to staff and users for feedback and comments. Users can see and interact with an early version.
Phase 2 blueprints the system. Phase 3 builds it using actual code.
The main objective of this phase is to resolve errors in coding, planning, and requirements before deployment.
Each individual unit (module/sub-system) in the system is tested separately. Outputs are tested against expected outputs.
Units that have already been tested and debugged are combined into the complete system and then the integrated system is tested.
The complete system is tested for its outputs for the given inputs. The entire system is evaluated as one unit against the original requirements.
Carried out at the late stages of testing. The users themselves are allowed to use the system. They either approve it or request further improvements.
| Order | Test Type | What is Tested | Who Tests |
|---|---|---|---|
| 1st | 🧩 Unit Testing | Individual modules/functions | Developers |
| 2nd | 🔗 Integration Testing | Combined modules together | Developers |
| 3rd | 🖥️ System Testing | Complete system end-to-end | Test team |
| 4th | ✅ Acceptance Testing | System in real-use conditions | Users |
Deployment means installing and launching the tested system for actual use. There are four different approaches.
The existing system is completely terminated and the new system is installed immediately. Old system abandoned if new system is successful.
Both the existing and new systems run simultaneously for a period. If new system succeeds, old system is terminated.
The new system is deployed in a selected area or location first. If successful, rolled out everywhere.
New system deployed in stages. Each stage replaces part of the old system, until fully replaced.
| Approach | Old System | Risk Level | Cost | Best For |
|---|---|---|---|---|
| Direct | Stopped immediately | Highest | Lowest | Small simple systems |
| Parallel | Runs alongside new | Lowest | Highest | Critical systems |
| Pilot | Still runs in other areas | Medium | Medium | Large national systems |
| Phased | Replaced part by part | Low-Medium | Medium | Complex, large systems |
Once deployed and in use, a system requires ongoing maintenance to remain useful and reliable.
Developing systems to suit new user requirements that emerge after deployment — the business grows, needs change.
Issues that surfaced with use that were not identified during development — edge cases, unusual inputs, real-world loads.
Adopting new technology into the system to improve efficiency — hardware upgrades, OS updates, security patches.
Different models organise the SDLC phases in different ways to suit different project types and team sizes.
Builds a working prototype early, gathers user feedback, and refines it until the final system meets requirements. Similar to the iterative model but prototype-focused.
Combines elements of Waterfall and Iterative. Each loop of the spiral represents a phase. Emphasises risk analysis at each cycle. Used for large, complex, high-risk projects.
12 questions covering all Chapter 2 topics. Click an answer to check immediately.
Exam-style questions with detailed answers. Try each one before revealing the solution.
Definition: A system is a collection of components that interact to achieve a specific task.
Example — Weather Forecasting System:
Unit Testing: Each individual unit (module) in the system is tested separately. Outputs are tested against expected results.
Eg: Testing the Accounts branch and Establishment branch independently.
Integration Testing: Units already tested and debugged are combined into the complete system and then tested together.
| Feature | Unit Testing | Integration Testing |
|---|---|---|
| What is tested | Single module in isolation | Modules combined together |
| Order | 1st (done first) | 2nd (after unit testing) |
| Errors found | Bugs within a module | Interface/communication errors |
Acceptance Testing (also called User Acceptance Testing) is carried out at the late stages of testing. The actual users are allowed to operate the system in real conditions. They either approve the system or request further improvements.
Why users must be involved:
Direct Deployment: The existing system is completely stopped and the new system is installed immediately.
Eg: The manual library system is completely abandoned and new library software is installed on the same day.
Advantage: Simple and fast; no cost of running two systems. Disadvantage: Highest risk — if new system fails, no backup exists.
Parallel Deployment: Both old and new systems run simultaneously for a period. Old system terminated only when new system proves reliable.
Eg: New library software and the manual system used together for one term; then manual system is dropped.
Advantage: Lowest risk — fallback to old system always available. Disadvantage: Highest cost — double the work and resources.
Pilot Deployment: The new system is first installed in one selected location. If successful, deployed everywhere.
Eg: A national vehicle revenue system installed in one province first. If successful, deployed to all provinces.
Use when: Deploying to a large number of locations — start with one to find problems before full rollout.
Phased Deployment: The new system is deployed in stages — one part at a time. Each phase replaces part of the old system.
Eg: Library system — Dictionaries module first, Course books second, Fiction last.
Use when: The system is very large and complex, or the organisation needs gradual transition.
The Waterfall Model consists of several stages of linear development. Each phase must be fully completed before the next phase begins, like water flowing down a waterfall.
Steps:
Main Limitation: The user only sees the working system at the very end of the life cycle. This creates a risk of developing a system that is different from what the user actually wants — because requirements may have changed, or may have been misunderstood early on, and the user has no opportunity to correct this until too late.
In the Iterative Incremental Model, a system is developed through repeated cycles (iterative) and in smaller portions at a time (incremental).
The process starts with a simple implementation of a sub-set of requirements and iteratively enhances the evolving versions until the complete system is implemented. At each iteration, design modifications are made and new capabilities are added.
How it addresses the Waterfall limitation:
Maintenance is important because a system installed and used over time needs to remain relevant, secure, and functional.
Three reasons a system requires maintenance: