Short answer

When post-processing additively manufactured metal parts with WEDM, carefully control pulse-on time for material removal and dimensional accuracy, and pulse-off time for surface finish and geometric stability. Utilize ANFIS-GRA modelling to predict and optimize these parameters for desired outcomes.

Field: Modelling
Source: Scientific Reports (2026)
Method: Hybrid modelling and experimental investigation
Sample: 27 experimental runs
Evidence: Strong effect

An Adaptive Neuro-Fuzzy Inference System (ANFIS) combined with Grey Relational Analysis (GRA) can accurately predict the outcomes of Wire Electrical Discharge Machining (WEDM) on Wire Arc Additive Manufacturing (WAAM) fabricated stainless steel, optimizing for multiple performance characteristics. This modelling research insight is drawn from a 2026 study published in Scientific Reports. Using Hybrid modelling and experimental investigation with 27 experimental runs, researchers explored how this design variable affects real-world outcomes. The key design takeaway: When post-processing additively manufactured metal parts with WEDM, carefully control pulse-on time for material removal and dimensional accuracy, and pulse-off time for surface finish and geometric stability. Utilize ANFIS-GRA modelling to predict and optimize these parameters for desired outcomes.

Study

ModellingNew This WeekStrong effect

ANFIS-GRA Hybrid Model Predicts WEDM Performance for WAAM Components with 99.85% Accuracy

Scientific Reports · 2026

Key Findings

01Pulse-on time (Ton) is the dominant factor influencing material removal rate (MRR), dimensional deviation (DD), and GD&T errors.
02Pulse-off time (Toff) significantly influences surface roughness (SR) and geometric stability.
03The ANFIS-GRA hybrid model achieved high predictive accuracy, with an R² value of 0.9985.

Application

Design takeaway

How to apply

Implement ANFIS-GRA or similar machine learning-based predictive models to optimize finishing processes for additively manufactured components, especially when multiple, potentially conflicting, performance objectives exist.

Project actions

01When choosing parameters for a manufacturing process, consider how they interact and affect multiple outcomes.
02Explore using computational modelling techniques (like ANFIS or other machine learning approaches) to predict and optimize your design or manufacturing process.

Method & Evidence

AimTo develop and validate a hybrid ANFIS-GRA model capable of predicting and optimizing the multi-response performance of WEDM on WAAM-fabricated stainless steel components.

MethodHybrid modelling and experimental investigation

ProcedureA Taguchi experimental design was used to conduct 27 WEDM trials on WAAM-fabricated SS316L components, varying pulse-on time, pulse-off time, and current. Grey Relational Analysis (GRA) was employed to determine a composite performance index. This index was then used to train an Adaptive Neuro-Fuzzy Inference System (ANFIS) model, which was subsequently validated against experimental results.

Sample27 experimental runs

ContextAdditive and subtractive manufacturing of stainless steel components

Variables

IV["Pulse-on time (Ton)","Pulse-off time (Toff)","Current"]

DV["Material Removal Rate (MRR)","Dimensional Deviation (DD)","GD&T errors","Surface Roughness (SR)","Geometric stability"]

CV["Material (SS316L)","WAAM fabrication process","WEDM machine type","Wire electrode material and diameter","Dielectric fluid type and flow rate"]

Strengths & Limitations

Strengths

+Integration of experimental data with advanced predictive modelling.
+Multi-response optimization approach.
+High predictive accuracy validated by low error metrics.

Limitations

The complexity of setting up and training ANFIS models can be a barrier. The accuracy of the model is highly dependent on the thoroughness and relevance of the experimental data used for training.

Reliability & validity

The study demonstrates strong validity through high correlation (R²) and low error metrics (MAPE, RMSE, MAE) between predicted and experimental results. Reliability is supported by the systematic experimental design (Taguchi) and the robust ANFIS modelling approach.

Think critically

How might the accuracy of the ANFIS-GRA model be affected if the initial WAAM process produced components with significant variations in microstructure or surface quality?

Design Principles

"Predictive modelling can significantly enhance the control and optimization of complex manufacturing processes by accurately forecasting performance based on input parameters."

This research demonstrates a powerful predictive modelling approach for complex hybrid manufacturing processes. By accurately forecasting the results of post-processing steps like WEDM, designers and manufacturers can reduce trial-and-error, optimize material removal rates, dimensional accuracy, and surface finish, leading to more efficient and higher-quality production of additively manufactured parts.

What This Means for Your Design

This research shows how a smart computer model (ANFIS-GRA) can accurately guess what will happen when you use a special cutting method (WEDM) on metal parts made by 3D printing (WAAM). It helps figure out the best settings to get a good finish and accurate shape.

How to use in your project

1.Reference this study when discussing the optimization of manufacturing processes, particularly for hybrid or additive manufacturing techniques.
2.Use the findings on parameter influence (Ton, Toff) to inform your own experimental design or process selection.

Add to My Project

Quick Cite

Paragraph starter

This research by Thejasree et al. (2026) highlights the effectiveness of hybrid modelling approaches, specifically ANFIS-GRA, in predicting and optimizing the Wire Electrical Discharge Machining (WEDM) of Wire Arc Additive Manufacturing (WAAM) fabricated stainless steel components. Their findings indicate that pulse-on time is critical for material removal and dimensional accuracy, while pulse-off time controls surface finish and geometric stability. The study achieved a high predictive accuracy (R² = 0.9985), demonstrating the potential for intelligent, data-driven optimization in advanced manufacturing.

Source

Scientific Reports

Experimental investigations on hybrid manufacturing: WEDM of WAAM-fabricated stainless-steel components using ANFIS modelling

journal · 2026

View source

Questions About This Research

What does the research say about anfis-gra hybrid model predicts wedm performance for waam components with 99.85% accuracy?: When post-processing additively manufactured metal parts with WEDM, carefully control pulse-on time for material removal and dimensional accuracy, and pulse-off time for surface finish and geometric stability. Utilize ANFIS-GRA modelling to predict and optimize these parameters for desired outcomes. Evidence: Scientific Reports (2026).
Why does "ANFIS-GRA Hybrid Model Predicts WEDM Performance for WAAM Components with 99.85% Accuracy" matter for design?: This research demonstrates a powerful predictive modelling approach for complex hybrid manufacturing processes. By accurately forecasting the results of post-processing steps like WEDM, designers and manufacturers can reduce trial-and-error, optimize material removal rates, dimensional accuracy, and surface finish, leading to more efficient and higher-quality production of additively manufactured parts.
How can designers apply this research?: When post-processing additively manufactured metal parts with WEDM, carefully control pulse-on time for material removal and dimensional accuracy, and pulse-off time for surface finish and geometric stability. Utilize ANFIS-GRA modelling to predict and optimize these parameters for desired outcomes.
What were the main findings?: Pulse-on time (Ton) is the dominant factor influencing material removal rate (MRR), dimensional deviation (DD), and GD&T errors.. Pulse-off time (Toff) significantly influences surface roughness (SR) and geometric stability.. The ANFIS-GRA hybrid model achieved high predictive accuracy, with an R² value of 0.9985.
What research method was used?: Hybrid modelling and experimental investigation with 27 experimental runs.
How strong is the evidence?: Evidence strength is rated Strong effect, based on a 2026 journal from Scientific Reports.
What should I do differently in my next project?: Implement ANFIS-GRA or similar machine learning-based predictive models to optimize finishing processes for additively manufactured components, especially when multiple, potentially conflicting, performance objectives exist.
What are the limitations?: The study focused on SS316L; results may vary for other materials. The ANFIS model's performance is dependent on the quality and range of the training data derived from the experimental runs.