DL-Based-Permutation-Flowshop-Problem

DL Based Permutation Flowshop Problem This Repository Contains My Solution to DL Based Permutation Flowshop Problem in CSE DataHack Competition. Introduction What is this Challenge About? The Permutation Flow Shop Problem (PFSP) is a classic optimization challenge that has seen various solution approaches over the years, from heuristics to metaheuristics and hyper-heuristics. Despite extensive research, PFSP remains challenging, particularly for large-scale instances with numerous jobs and machines. Recently, deep learning (DL) has emerged as a powerful tool for handling complex, high-dimensional data spaces. This has led researchers to explore DL applications in solving PFSP, either by combining it with traditional heuristics or developing pure DL approaches. Challenge Requirements The challenge focuses on developing a deep learning solution for the Permutation Flow Shop Problem: Work with 24 problem instances Instance configurations: Jobs: 10, 20, 50, or 100 Machines: 5, 10, or 20 Two instances per configuration Resources provided: Python matrices for all instances in datahack2025flowshop.ipynb Makespan calculation function included Solution