Abstract:Fused deposition modeling (FDM) is a widely used additive manufacturing (AM) process due to its simplicity, cost-effectiveness, and low waste generation. The mechanical properties of FDM products are greatly influenced by the process parameters. Thus, it is important to use optimal process parameters to achieve desired mechanical properties. One way to determine the optimal process parameters is using response surface methodology (RSM) which has been widely implemented for determining optimal manufacturing process parameters. For RSM, experimental investigations are performed in a systematic way using design of experiments (DOE). However, in some cases, experimental data is not collected in a systematic manner using DOE. Additionally, it is expensive and time consuming to conduct experimental investigations of a well-designed DOE to acquire data. This results in generating insufficient experimental data for developing response surface models using RSM. To address these challenges, this study presents a streamlined five-step framework to optimize FDM process parameters for achieving desired mechanical properties of products. In this process, once experimental data is collected, it is analyzed and preprocessed to remove or repair any missing data. Next, this data is used to develop an artificial neural network (ANN) predictive model to predict the systematic DOE data necessary for developing response surface models. Finally, these models are used to optimize the process parameters for maximizing mechanical properties of FDM products. The efficacy of this framework is demonstrated in detail for a real-world FDM application with limited experimental data. This framework determines the four optimal process parameters for simultaneously maximizing three mechanical properties. By utilizing ANN and genetic algorithm-based optimization, this approach minimizes the need for extensive experiments. Thus, unlike any other work in the literature, this framework not only determines optimal FDM process parameters which results in achieving the desired mechanical properties, but also significantly reduces the time and resources required in the process, thereby paving the way for a more efficient manufacturing process. The optimized results obtained using this framework are found to be very close to experimental data, thereby establishing that the framework is effective for determining optimal parameters in case of limited or non-systematic DOE data. In future, the generic nature of the framework can be utilized to include other FDM process parameters, material characteristics affecting the properties, and different predictive modeling and optimization techniques. Finally, such a framework can be modified as necessary for utilizing it for other AM and traditional manufacturing processes.

Neural network-based modeling of FFF process for PET-G: Evaluating MLPNN and RBFNN performance in mechanical property prediction

Melt index prediction by adaptively aggregated RBF neural networks trained with novel ACO algorithm

A comparative analysis between backpropagation and general regression neural networks to predict the tensile properties of fused filament fabricated parts

Modelling and prediction of mechanical properties of FFF-printed polycarbonate parts using ML and DA hybrid approach

Optimization of FDM 3D printing process parameters for multi-material using artificial neural network

Development of artificial intelligence‐based neural network prediction model for responses of additive manufactured polylactic acid parts

Optimizing FDM process parameters: predictive insights through taguchi, regression, and neural networks

Artificial neural network for predicting the mechanical performance of additive manufacturing thermoset carbon fiber composite materials

Prediction of Stress–Strain Behavior of PET FRP-Confined Concrete Using Machine Learning Models

A framework for optimizing process parameters in fused deposition modeling using predictive modeling coupled response surface methodology

Effects of printing parameters on the mechanical characteristics and mathematical modeling of FDM-printed PETG

Temperature Distribution Prediction in Laser Powder Bed Fusion using Transferable and Scalable Graph Neural Networks

Experimental investigation and neural network development for modeling tensile properties of polymethyl methacrylate (PMMA) filament material

Modeling of additive manufacturing processes with time‐dependent material properties using physics‐informed neural networks

Predictive modeling and optimization of Rockwell hardness of additively manufactured PEEK using RSM, ANFIS and RNN integrated with PSO.

Artificial Neural Networks for Predicting Mechanical Properties of Crystalline Polyamide12 via Molecular Dynamics Simulations

Process parametric optimization of fused deposition modeling for manufacturing of acrylonitrile butadiene styrene parts

Comparative Study of the Sensitivity of PLA, ABS, PEEK, and PETG’s Mechanical Properties to FDM Printing Process Parameters

Can a Black-Box AI Replace Costly DMA Testing?—A Case Study on Prediction and Optimization of Dynamic Mechanical Properties of 3D Printed Acrylonitrile Butadiene Styrene

Predicting Mechanical Properties of Polymer Materials Using Rate-Dependent Material Models: Finite Element Analysis of Bespoke Upper Limb Orthoses

MechProNet: Machine Learning Prediction of Mechanical Properties in Metal Additive Manufacturing