Two-Stage Stochastic Problem Solver for INDUSTRY - MANUFACTURING

This repository hosts the solution code for the Problem 1 from the assignment of MATHEMATICAL MODELING (CO2011) - HCMUT 2023, focusing on the INDUSTRY - MANUFACTURING problem. The code implements a two-stage stochastic problem-solving algorithm that optimizes pre-order and production decisions under uncertainty.

Overview

The algorithm is designed to address a stochastic two-stage problem, which is a common challenge in the field of industrial manufacturing. The goal is to make informed decisions in the first stage (Pre-order) without complete knowledge of future events, and then adjust those decisions in the second stage (Production) once the uncertainty is resolved.

The solution utilizes the GAMS Python API (GAMSPy) to model the problem and compute the optimal strategy, taking into account the probabilistic nature of demand and other uncertain parameters.

By leveraging the power of mathematical optimization and stochastic modeling, this code helps students and professionals in the manufacturing industry to visualize robust strategies that can withstand the unpredictability of real-world scenarios.

Description

The project consists of two main files: main.py and helper.py. The main.py file defines the sets, parameters, variables, equations, and model for the optimization problem. The helper.py file contains functions to print the optimal results and scenario-specific outcomes.

Getting Started

Dependencies

• Python 3.10.11
• GAMSPy: A Python API for GAMS that allows for modeling and solving optimization problems.
• NumPy: A fundamental package for scientific computing with Python.

Installing

• Clone the repository to your local machine.
• Ensure that you have Python and the required libraries installed.

Executing program

• Run the main.py file in your Python environment.
• Input the required data when prompted by the program.

python main.py

Authors

• Nguyễn Ngọc Đình Khoa - Team Leader, Work Distribution, Problem Understanding, Documentation
• Lê Phong Hào - Assistant Leader, Documentation
• Lê Hà Nguyên Khánh - Documentation
• Phạm Trần Đăng Khoa - Code Prototype, Source Code Documentation
• Huỳnh Thanh Duy - Source Code Completion, Input Implementation

Acknowledgments

• Heartfelt thanks to all team members for their contributions and dedication to the project.
• Special appreciation to Nguyễn Ngọc Đình Khoa for leading the team and guiding the project's direction.
• Gratitude to Lê Phong Hào and Lê Hà Nguyên Khánh for their efforts in documenting and understanding the problem.
• A big thank you to Huỳnh Thanh Duy for his work on finalizing the source code.
• Kudos to everyone involved for their hard work and collaboration.