🧬 LCVRP Genetic Algorithm

💡 Overview

This project goal was to find the best soultion for the Limited Capacitated Vehicle Routing Problem (LcVRP). To achieve it we had to used Genetic Algorithm (GA) written in lower level C++, with the code optimized down to the last detail.

⚙️ Commands

Note: In case of path error, modify the ProblemLoader::BASE_PATH, ProblemLoader::BASE_EXTENSION, folderName or instanceName parameters.

# Initialize build directory
cmake -S . -B build

# Builds project
cmake --build build --target main

# Run project
./build/main

Parameters

int populationSize;     // Controls the size of the population
double mutProb;         // Controls probability for mutation occurances
double crossoverProb;   // It's the probability for solution crossing, when it did not occur we simply copy indivduals
int numTurns;           // Number of turns, at which tournament selection will be looking for mate
int numEpochs;          // How many times Optimizer::runIteration will be invoked
int numGroups;          // Number of groups of trucks to optimize

📚 Documentation

Generate Documentation Template

# Generates documentation
doxygen -g

# Updates documentation
doxygen

# Run documentation
./docs/html/index.html

Custom Operators: Overloads < and > to facilitate integration with the HeapSort algorithm.