UNIVERSITY OF WEST ATTICA
SCHOOL OF ENGINEERING
DEPARTMENT OF COMPUTER ENGINEERING AND INFORMATICS

University of West Attica · Department of Computer Engineering and Informatics

---

Artificial Intelligence

Real Genetic Algorithm Application

Vasileios Evangelos Athanasiou
Student ID: 19390005

GitHub · LinkedIn

---

Supervision

Supervisor: Paris Mastorokostas, Professor

UNIWA Profile

Co-supervisor: Panagiota Tselenti, Laboratory Teaching Staff

UNIWA Profile · LinkedIn

---

Athens, January 2023

---

---

README

Real Genetic Algorithm Application

This repository contains a Genetic Algorithm (GA) implementation in C for solving combinatorial optimization problems involving settlements in Halkidiki.

The program simulates the GA process across a specified number of generations, using:

• Population Initialization
• Selection (Roulette Wheel / Proportional)
• Crossover (Single-Point)
• Mutation (Random within bounds)
• Fitness Evaluation (Euclidean Distance to points)

Two problem cases are provided:

1. 25 settlements (Halkidiki_25.txt)
2. 12 settlements (Halkidiki_12.txt)

The results include best solutions, fitness values, and performance metrics over multiple experiments.

---

Table of Contents

Section	Folder	Description
1	assign/	Assignment material for the Artificial Intelligence course
1.1	assign/Assignment_2_AI.pdf	Assignment description in English
1.2	assign/Εργασία_2_ΤΝ.pdf	Assignment description in Greek
2	docs/	Project documentation for genetic algorithms
2.1	docs/Genetic-Algorithms.pdf	English documentation covering genetic algorithms
2.2	docs/Γενετικοί-Αλγόριθμοι.pdf	Greek documentation covering genetic algorithms
3	res/	Execution results and outputs
3.1	res/results_12.txt	Results for parameter set 12
3.2	res/results_25.txt	Results for parameter set 25
4	src/	Source code and input files
4.1	src/Halkidiki_12.txt	Input data for parameter set 12
4.2	src/Halkidiki_25.txt	Input data for parameter set 25
4.3	src/project_GA_new.c	Main C implementation of the genetic algorithm
4.4	src/simpleGA_new.c	Supporting C code for genetic algorithm
4.5	src/simpleGA_new.h	Header file for supporting functions
4.6	src/Project1.dev	Development or IDE project file
5	README.md	Repository overview
6	INSTALL.md	Usage instructions

---

1. Contents

1. The World of the Problem
   Defines coordinates of settlements and problem objectives.

2. The Genetic Algorithm in Application for 25 Settlements
   GA parameterization and initial population setup.

3. The Final State of the Problem for 25 Settlements
   Best solutions and corresponding fitness after GA convergence.

4. Commenting on the Genetic Algorithm Solutions
   Analysis of solution quality and convergence behavior.

5. Show Paradoxical Solutions?
   Discussion of unexpected or non-intuitive GA outcomes.

6. The Genetic Algorithm in Application for 12 Settlements
   Implementation for a smaller problem instance.

7. The Final State of the Problem for 12 Settlements
   GA results and performance metrics for 12 settlements.

8. Differences in the Solutions of the Two Problems
   Comparative analysis highlighting insights between problem sizes.

---

2. Technologies Used

• Programming Language: C
• Algorithm: Genetic Algorithm (SGA)
• Key Concepts:
  • Population Initialization
  • Fitness Evaluation (Euclidean distance)
  • Selection (Roulette Wheel)
  • Single-Point Crossover
  • Random Mutation
• Data Files:
  • Halkidiki_12.txt – Coordinates for 12 settlements
  • Halkidiki_25.txt – Coordinates for 25 settlements
• Platform: Cross-platform C compiler (GCC recommended)
• Output: Log file (results_12.txt or results_25.txt) for analysis