MATRIC : 86022

ROLE : DOCUMENTATION AND PADLET

Matric: 84812

Role: Responsible for system developer

 Case 2: 100 Cities
- Initial Random Distance: ~8200 units
- Best Distance Found: 3154.2 units
- Convergence: ~190 generations

Plot Descriptions:
- City Plot: Blue circles represent cities.
- Optimized Route Plot: Red lines show the shortest route found by GA.

Matric: 85024

Role: Fuzzy System Research 2 & Documentation

This project is about creating a fuzzy expert system to help form student groups for assignments. Sometimes it’s hard to choose the right groups for assignments. Sometimes it’s hard to choose the right group members because people have different skills, ways of communicating, and levels of responsibility. So, in this project we use fuzzy logic to check those aspects and suggest better group matches in a smarter and more flexible way, just like how a human would decide.

The GA was designed with the following key elements:
- Encoding
- Fitness Function
- Selection
- Crossover
- Mutation
- Replacement

Algorithm Steps
1. Generate Initial Population (random permutations of cities).
2. Evaluate Fitness for all chromosomes (shorter distance = better).
3. Selection using tournament method.
4. Crossover to produce new offspring.
5. Mutation to introduce variability.
6. Repeat steps 2–5 for a set number of generations (200).
7. Return the best solution found.

MATRIC : 86348

ROLE : TEAM LEADER, GENETIC ALGORITHM OPTIMIZATION & DOCUMENTATION

Matric No: 86202

Role: Fuzzy System Researcher 1 & Documentation

No Matric : 87043

Role : System Developer

Each FIS is driven by IF-THEN rules that reflect expert logic. Example rules include:

a) Skill Level FIS

• IF Subject Knowledge is High AND Technical Ability is High THEN Skill Level is Strong.

b) Communication FIS

• IF Responsiveness is Prompt AND Collaboration is Often THEN Team Behavior is Active.

c) Reliability FIS

• IF Task Completion is Low AND Submission is Often Late THEN Reliability is Unreliable

1. FIS 1: Skill Level Evaluation

• Inputs:

  • Subject Knowledge (Low, Medium, High) 

  • Technical Ability (Low, Medium, High)

• Outputs:

  • Skill Level (Weak, Moderate, Strong)

2. FIS 2: Communication Styles Evaluation

• Inputs:

  • Responsiveness (Slow, Average, Prompt)

  • Collaboration Frequency (Rarely, Sometimes, Often)

• Outputs:

  • Team Behavior (Passive, Balanced, Active)

3. FIS 3: Reliability Assessment

• Inputs: 

  • Task Completion Rate (Low, Medium, High)

  • Submission Puntuality (Often Late, Sometimes Late, On Time)

• Ouput: 

  • Reliability Level (Unrealiable, Fair, Reliable)

On the right, the 3D surface viewer shows the non-linear correlation between the 'knowledge' and 'Ability' inputs to the output 'skill_level'. This surface serves as a visual indicator of the combined effects of a student's subject knowledge and technical skill on overall ability.

The right-hand side 3D surface viewer demonstrates the influence on overall 'GroupMatchScore' of 'Ability' (which could simply be general 'Skill Level' from FIS 1) and 'responsiveness' (from FIS 2). The surface graph is useful for observing how the general system output changes as these significant parameters change, and for obtaining an understanding of how the system determines overall student compatibility for group assignment.

Subject Knowledge: High

Technical Ability: High

Responsiveness: Prompt

Collaboration: Often

Task Completion: Low

Submission: Often Late

→ Outputs: Skill Level: Strong, Team Behavior: Active, Reliability Level: Unreliable, Group Match Score: Medium

Student A possesses good subject expertise and high technical ability, high skill overall. They can respond in a timely manner and work often together, evidencing active team behavior and willingness to work in groups. However, their high frequency of late submission and low percentage of task completion evidence unreliability. Their reliability score is thus low, which lowers their group compatibility overall despite high skills and communication. Overall, Student A’s group match score is medium because their work is inconsistent due to their unreliability in task management.

Technical Ability: Medium

Responsiveness: Average

Collaboration: Sometimes

Task Completion: High

Submission: On Time

→ Outputs: Skill Level: Moderate, Team Behavior: Balanced, Reliability Level: Reliable, Group Match Score: High

Student B possesses a medium subject knowledge and technical skill, giving him a medium skill level. His responsiveness is also medium and, at times, collaborative, which displays a balanced team behavior neither dominant nor submissive but assistive when required. With his high task completion and on-time submission, student B is a consistent and reliable performer. These traits make Student B an extremely compatible team player. Hence, Student B’s group match score is high, on the basis of a balanced and consistent team contribution.

This fuzzy expert system supports the grouping process in education by using intelligent reasoning to evaluate students beyond numerical grades. By analyzing skills, communication behavior, and reliability, it helps in creating balanced teams that are likely to collaborate efficiently. This approach is especially useful for instructors managing large classes or online collaborations.