Γενετικοί αλγόριθμοι και το πρόβλημα του περιοδεύοντα πωλητή (TSP)
Genetic algorithms and the Travelling Salesman Problem (TSP)
Bachelor Dissertation
Author
Ιμπραχίμι, Έλτον
Ibrahimi, Elton
Date
2024-09View/ Open
Keywords
TSP ; Genetic algorithms ; Dynamic programming ; Brute force ; Traveling Salesman Problem ; Γενετικοί Αλγόριθμοι ; Το πρόβλημα του περιοδεύοντα πωλητή ; Δυναμικός προγραμματισμόςAbstract
The document focuses on solving the Traveling Salesman Problem (TSP) using Genetic Algorithms (GA). It outlines the structure, processes, and methodologies applied in the implementation. Key sections cover the initialization of the genetic population, fitness evaluation, selection methods (such as Tournament and Roulette Wheel), crossover techniques (like Cycle and Order-based Crossover), and mutation operations (e.g., Inversion, Scramble). The document also emphasizes the role of fitness scaling methods, such as Sigma and Boltzmann scaling, to enhance selection pressure and avoid premature convergence. The content provides both theoretical and practical guidance for implementing GAs to address TSP efficiently.