Automated Bottle Filling and Capping Plant

Abstract

Background: Manual bottle-filling and capping in industries is often labor-intensive and error-prone, leading to inconsistent fill levels, wasted product, and reduced quality. Advances in Industry 4.0 and vision systems suggest the need for automated solutions. Objective: This study aims to develop an industry-level automated plant that integrates computer vision to precisely fill and cap bottles of carbonated liquid, detecting and correcting under-filled or uncapped bottles. Methods: The prototype uses a Raspberry Pi for real-time image processing (Canny edge detection) and an Arduino UNO to control conveyor, filling pump, and capping motors. Infrared sensors and limit switches synchronize the bottle's position. Results: In testing, the system successfully transports bottles through filling and capping stations while machine vision reliably classifies bottle status. Level detection achieved 97% accuracy, and automation reduced total production time by 30%. Under-filled bottles were automatically refilled, and uncapped bottles were flagged for correction. Conclusions: The integrated system met its objectives, significantly improving throughput and quality control. It realizes an automated bottling process that can cut costs and waste. The results demonstrate the feasibility of replacing manual quality checks with smart machine vision in bottling operations.