ruziniuuuuu / ME5414_Project

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

ME5414 Project

Author: Cao Chenyu

Introduction

This project undertakes a comparative study of two Linear Programming (LP) solvers: the classic Simplex method and the modern Barrier method. Our analysis is grounded on computational efficiency, tolerance sensitivity, and adaptability to diverse problem structures, utilizing a suite of standard LP test problems for a comprehensive evaluation.

Project Structure

ME5414_Project
├─ assets
│  ├─ results.png
│  ├─ results_constraints.png
│  ├─ results_tolerance.png
│  └─ results_variables.png
├─ data
│  ├─ adlittle.mps
│  ├─ afiro.mps
│  ├─ sc50a.mps
│  └─ share1b.mps
├─ matlabPaths.txt
├─ README.md
├─ report_typ
│  ├─ main.pdf
│  ├─ main.typ
│  └─ refs.bib
├─ src
│  ├─ barrier_solver.m
│  ├─ generate_problems.m
│  ├─ load_problem_data.m
│  ├─ main.m
│  ├─ run_experiments.m
│  ├─ run_experiments_struct.m
│  ├─ run_tolerance_experiments.m
│  ├─ simplex_solver.m
│  ├─ visualize_results.m
│  ├─ visualize_results_constraints.m
│  ├─ visualize_results_variables.m
│  └─ visualize_tolerance_results.m
└─ startup.m

Usage

To run the experiments:

  1. Ensure MATLAB is installed and the paths are configured correctly in matlabPaths.txt.
  2. Navigate to the src/ directory and run main.m to execute all experiments.
  3. View results in the assets/ directory or compile the report_typ/ directory for the full report.

Contributions

This project is part of coursework for ME5414. Contributions are limited to academic collaborations only.

License

This project is proprietary and confidential. Unauthorized copying, distribution, or use is strictly prohibited.

About

License:MIT License


Languages

Language:JetBrains MPS 69.3%Language:MATLAB 19.5%Language:Typst 10.1%Language:TeX 1.1%