An Intelligent Hybrid Manufacturing System for FDM Surface Defects Monitoring with Industry 4.0

Authors

  • Fahim Al-Rashid Chowdhury Department of Mechanical & Production Engineering, Ahsanullah University of Science & Technology, Dhaka-1208, BANGLADESH
  • Pervez Hossain Department of Mechanical & Production Engineering, Ahsanullah University of Science & Technology, Dhaka-1208, BANGLADESH
  • S M Rahaman Department of Mechanical & Production Engineering, Ahsanullah University of Science & Technology, Dhaka-1208, BANGLADESH
  • Md Shihab Shakur Department of Industrial & Production Engineering, Bangladesh University of Engineering & Technology, Dhaka-1208, BANGLADESH
  • M. Azizur Rahman McMaster Manufacturing Research Institute (MMRI), Department of Mechanical Engineering, McMaster University, Hamilton, ON L8S4L7, Canada
  • Md. Shahnawaz Bhuiyan Department of Mechanical & Production Engineering, Ahsanullah University of Science & Technology, Dhaka-1208, BANGLADESH

DOI:

https://doi.org/10.38032/scse.2025.2.10

Keywords:

Surface quality, Grinding operation, Hybrid manufacturing, Convolutional neural network, Sustainability

Abstract

Additive manufacturing (AM) and machining in a single machine colloquially known as Hybrid manufacturing help to produce customized and complex products without assembling including greater design freedom and reduced material wastage. A CNC-based grinding mechanism is introduced in the same system to overcome those defects and enhance the quality. To increase productivity and improve product surface quality, evolving additive manufacturing demand and finishing subtractive processes must be combined on the same platform. For the additive manufacturing method, Fused Deposition Modeling (FDM) has been employed, and a grinding operation can be performed for surface finishing. A camera module is used to capture surface images for defect detection such as stringing, rashing, and surface cracking after the AM process. Convolutional Neural Network (CNN) is applied to the captured image for the defect detection process. If the CNN analysis reveals any surface defects, a grinding operation will be performed on the surface for better surface quality. The architecture has provided a platform to collect data from the image captured by the camera module for evaluating and identifying surface defects using CNN. CNN model provided 89% accuracy for surface defects detection. As a result, the CNC grinding operation can be done on that particular surface of the product for smoothing the partially roughened surfaces. Therefore, this study demonstrates to improve the surface quality, reduce cycle time, set up time reduction & improve the product's sustainability. The proposed approach of a hybrid manufacturing system also provides a basic framework to increase efficiency, reduce downtime, increase efficiency, improve end part consistencies of the product as a consequence of post-processing & defect detection in the same system, and enable I4.0

Downloads

Published

08.05.2025

How to Cite

An Intelligent Hybrid Manufacturing System for FDM Surface Defects Monitoring with Industry 4.0. (2025). SciEn Conference Series: Engineering, 2, 46-51. https://doi.org/10.38032/scse.2025.2.10

Similar Articles

1-10 of 20

You may also start an advanced similarity search for this article.