CCS516 Computational Intelligence Semester 1, 2022/2023 Assignment 01
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
CCS516 Computational Intelligence
Semester 1, 2022/2023
Assignment 01
Facility-Location Set-Covering Problem
District Wonderville has 58 small towns (towns 1 to 58). The district council is in the process of determining the locations of building fire stations to serve the community of these 58 towns. The district council has limited fund and the council wants to build the minimum number of fire stations.
You are hired by the district council to help them in making such decisions. The district council wants you to advise them on the minimum number of fire stations needed to ensure that at least one fire station is within the following driving times of each city.
a) 3 minutes
b) 5 minutes
c) 10 minutes
d) 15 minutes
e) 20 minutes
In other words, for each of the abovementioned scenarios, the district council needs a decision of the minimum number of fire stations to be built.
drivingtime.csv shows the driving times (in minutes) required to drive between the towns in District Wonderville.
You intend to develop a Genetic Algorithm (GA) to help the district council in deciding the minimum number of fire stations to be built in the district. The proposed GA must cover the following aspects:
1. Problem modeling: variables and representation scheme.
2. Fitness function design and constraints management.
3. Parents and offspring population management models.
4. Parent and survivor selection mechanism.
5. Crossover and mutation operators.
6. Initialization.
7. Parameter tuning and termination criteria: For each scenario, tune the parameter of the GA. Based on a particular setting, use a maximum of 100 fitness function evaluations for each run and report your results averaged over 10 runs.
This assignment can be implemented using any of the programming languages, but preferably Java .
Assignment due date: 11 Dec 2022 (Sunday), 11:59 p.m. (Week 08).
Assignment type: Pair assignment (a group of two members). The member grouping will be done via the eLearn@USM.
Submission procedure: A zip/rar package which consists of the following items is to be submitted via the eLearn@USM:
1. An assignment report (Times New Roman, font size 12, single spacing, not more than ten pages) which explains the algorithm and technique used as well as the experiments performed to obtain the best results. All the important aspects such as representation scheme, parent population size, offspring population size, parent selection mechanism, survival selection mechanism, reproductive operators, initialization, parameter setting, stopping criteria, and fitness function design. [Hint to create a good report: The report should be prepared in such a way that it allows one to replicate your experiments]
2. A cover page which contains your details is expected to be included in your assignment report.
3. All other supporting documents which can make the evaluator understand your work (e.g. source codes, screenshots, additional files from the relevant tools used in implementing the assignment).
The zip/rar package must be named according to the following notation: CCS516_[GroupNumber]_A01. For example, for Group03, they must name the zip/rar package as CCS516_Group03_ A01.
Reference: Kindly state any source of reference should you refer to various sources to complete this assignment.
Grading rubric:
The total mark for this project is 100 marks. The total will be scaled to 8% of your overall grade. It will consider the followings:
1. Problem modeling/formulation (30%): Illustrate and explain the problem modeling/formulation, fitness function design, and constraints management.
2. Application (40%): Describe and implement the technique/algorithm that are used to solve the facility-location set-covering problem.
3. Parameter setting & results (20%): Explain the results and show how the results are used to support decisions. Inclusion of what-if analysis based on the four scenarios is expected.
4. Problem and pitfalls (10%): Discuss the mistakes that have been done and the knowledge & experience gained throughout the project implementation.
|
Very Weak (1 – 2 points) |
Weak (3 – 4 points) |
Fair (5 – 6 points) |
Good (7 – 8 points) |
Very Good (9 – 10 points) |
Problem modeling / formulation (30%) |
Not able to illustrate and explain the problem modeling / formulation, fitness function design, and constraints management. |
Able to illustrate and explain the problem modeling / formulation, fitness function design, and constraints management with minimal clarity. |
Able to illustrate and explain the problem modeling / formulation, fitness function design, and constraints management with satisfactory clarity. |
Able to illustrate and explain the problem modeling / formulation, fitness function design, and constraints management with good clarity. |
Able to illustrate and explain the problem modeling / formulation, fitness function design, and constraints management with excellent clarity. |
|
Very Weak (1 – 2 points) |
Weak (3 – 4 points) |
Fair (5 – 6 points) |
Good (7 – 8 points) |
Very Good (9 – 10 points) |
Application (40%) |
• Not able to apply any new idea or knowledge to a given problem. • The algorithm implementation is not correct and not comprehensive. |
• Limited ability to apply new idea or knowledge. • The algorithm implementation is minimally correct. |
• Able to apply new idea or knowledge to a given problem. • The algorithm implementation is partially correct. |
• Able to apply new idea or knowledge to a given problem. • The algorithm implementation is correct and comprehensive. |
• Able to apply new idea or knowledge to a given problem and able to propose alternative applications. • The implementation based on the alternative applications is correct and comprehensive. |
Parameter setting & results (20%) |
• For each scenario, report the results based on at least 3 set of different parameter settings. • Interpret and explain the results. • The explanation and the results are presented unclearly, loosely, and disorganized. |
• For each scenario, report the results based on at least 3 set of different parameter settings. • Interpret and explain the results. • The explanation and the results are presented with minimal clarity, comprehensiveness, and organization. |
• For each scenario, report the results based on at least 3 set of different parameter settings. • Interpret and explain the results. • The explanation and the results are presented with satisfactory clarity, comprehensiveness, and organization. |
• For each scenario, report the results based on at least 3 set of different parameter settings. • Interpret and explain the results. • The explanation and the results are presented with good clarity, comprehensiveness |
2022-12-08