ML Algorithm to minimize cost of manufacturing process
FIRAS TLILI
Data Scientist
ML Engineer
AI Model Developer
pandas
Python
scikit-learn
ML-Algorithm-to-minimize-cost-of-manufacturing-process
This repository contains implementations of various optimization and simulation projects aimed at solving real-world problems in manufacturing and inventory management, along with a classification task for widget quality assessment.
Projects Overview
1. Particle Swarm Optimization (PSO) for Manufacturing Cost Minimization
Objective: Minimize the total manufacturing cost by optimizing the number of workers and machines.
Key Features:
Calculates labor and machine costs with penalties for underproduction.
Utilizes PSO to find optimal resource combinations.
Visualizes cost landscape with a 3D surface plot.
2. Monte Carlo Simulation for Inventory Management
Objective: Optimize inventory levels to reduce customer wait times and shelf times.
Key Features:
Simulates customer orders and inventory availability over a year.
Analyzes average wait and shelf times.
Visualizes results with histograms and summary statistics using Pandas.
3. Widget Classification Using Random Forest
Objective: Classify widgets as defective or non-defective based on their features.
Key Features:
Generates synthetic data for widget characteristics.
Trains a Random Forest classifier and evaluates its performance.
Visualizes feature importance to identify key predictors.
Getting Started
To run the projects locally, clone this repository and follow the instructions in the respective project directories.
Project Dependencies
Python 3.x
Libraries: numpy, pandas, matplotlib, scikit-learn
