Material Removal Rate Research Articles

Remaining useful lifetime prediction for milling blades using a fused data prediction model (FDPM)

AbstractIn various industry sectors, predicting the real-life availability of milling applications poses a significant challenge. This challenge arises from the need to prevent inefficient blade resource utilization and the risk of machine breakdowns due to natural wear. To ensure timely and accurate adjustments to milling processes based on the machine's cutting blade condition without disrupting ongoing production, we introduce the Fused Data Prediction Model (FDPM), a novel temporal hybrid prediction model. The FDPM combines the static and dynamic features of the machines to generate simulated outputs, including average cutting force, material removal rate, and peripheral milling machine torque. These outputs are correlated with real blade wear measurements, creating a simulation model that provides insights into predicting the wear progression in the machine when associated with real machine operational parameters. The FDPM also considers data preprocessing, reducing the dimensional space to an advanced recurrent neural network prediction algorithm for forecasting blade wear levels in milling. The validation of the physics-based simulation model indicates the highest fidelity in replicating wear progression with the average cutting force variable, demonstrating an average relative error of 2.38% when compared to the measured mean of rake wear during the milling cycle. These findings illustrate the effectiveness of the FDPM approach, showcasing an impressive prediction accuracy exceeding 93% when the model is trained with only 50% of the available data. These results highlight the potential of the FDPM model as a robust and versatile method for assessing wear levels in milling operations precisely, without disrupting ongoing production.

On Performance Enhancement of the STED Process Using Modified Tool Electrode

Abstract Shaped tube electrolytic drilling (STED) is an electrochemical machining variant that employs a tubular tool electrode to produce holes with a high aspect ratio on hard-to-machine materials. The tubular electrodes with a diameter below 1 mm produce undesired residue (center-peak) at the machining surface that hampers the machining rate. Therefore, this study attempts to improve the electrolyte flow and enhance the electrochemical dissolution through tool modification. The performance enhancement of the STED process in terms of material removal rate and average diametral overcut has been explored. The issues related to the limitations in material removal in the STED process are brought forth with the technique to resolve those difficulties. The experiments were conducted with the modified tubular tool, and the length of slits on the tool was selected based on the simulation insights and pilot study. The effects of input process parameters (applied voltage, tool feed rate, electrolyte concentration, and tool slit length) on the output responses obtained from the STED process are elaborated. Holes with diameters in the range 0.89-0.97 mm and 12 mm depth were fabricated. Further optimization of the process parameters in the design space is also presented to obtain sustainable process performance.

Multi-objective optimization for electric discharge drilling of waspaloy: A comparative analysis of NSGA-II, MOGA, MOGWO, and MOPSO

This study investigates the impact of pulse-on-time (TON), current (I), pulse-off-time (TOFF), voltage (V), and tool electrode speed (TES) on material removal rate (MRR), average surface roughness (Ra), power consumption (PC) and surface defects for electrical discharge drilling (EDD) of Waspaloy. Based on a Taguchi design with five factors and four levels, experimental trials reveal TOFF, TON, and TOFF as the most influential factors, contributing 50.45%, 34.29%, and 33.09% to MRR, Ra, and PC, respectively. To address conflicting conditions and optimize responses, non-dominated sorting genetic algorithm-II (NSGA-II), multi-objective genetic algorithm (MOGA), multi-objective grey wolf optimizer (MOGWO), and multi-objective particle swarm optimization (MOPSO) are employed. Comparative analysis of 8 optimal solutions provided by four optimization techniques, which exhibit less than 10% errors in confirmatory experiments, has been done, ensuring better prediction accuracy. MOPSO is preferable to the rest of the optimization techniques due to its lower percentage error. NSGA-II, MOGA, MOGWO, and MOPSO exhibit relative improvement in responses based on confirmatory experiments, establishing them as effective tools for exploring EDD applications. Higher current is confirmed as a primary contributor to increased surface damage through microscopic image analysis, and EDX analysis exposes carbon deposition migration on both the tool and workpiece.

Enhancing surface integrity and performance of EDM with sustainable dielectrics and electrode modifications

A common non-traditional method for precise machining of hard materials is electro-discharge machining (EDM). The efficiency and surface quality in the EDM process are decided by the design and type of the electrodes. Because of its lightweight, cost-effectiveness and easy machinability, aluminium is employed in a wide range of applications. In this work, AlSi10Mg electrodes are manufactured by different processing routes such as 3D printing, casting, and extrusion. Different characterization techniques are carried out to determine the mechanical, electrical, and thermal properties of all differently processed aluminium alloy electrodes. The processing route influence during the EDM of Ti-6Al-4V is evaluated with different levels of operating parameters using conventional EDM oil and lemon peel biodiesel as dielectrics. The experiments are performed with 3 different levels of current (Ip) and pulse on time (Ton) with a constant voltage. The output responses viz. material removal rate (MRR), Tool wear rate (TWR), average surface roughness (Ra) and white layer thickness (WLT) are considered to compare the EDM performance exhibited by differently processed aluminium alloy electrodes. The 3D printed aluminium alloy electrode produces about 23% less TWR and 18% improved surface quality than the conventional (extruded) electrode. The performance exhibited by the cast aluminium alloy electrode is sub-optimal when compared to electrodes processed by the other two routes. The microscopic examination revealed that WLT could be reduced to the extent of about 29% and 39% with 3D printed AlSi10Mg electrodes when compared with extruded and cast electrodes. The present study concluded that a 3D printed electrode with lemon peel dielectric is the most preferable combination for high surface finish operation.

A comprehensive review of parametric optimization of electrical discharge machining processes using multi-criteria decision-making techniques

Optimization of electrical discharge machining (EDM) processes is a critical issue due to complex material removal mechanism, presence of multiple input parameters and responses (outputs) and interactions among them and varying interest of different stakeholders with respect to relative importance assigned to the considered responses. Multi-criteria decision making (MCDM) techniques have become potent tools in solving parametric optimization problems of the EDM processes. In this paper, more than 130 research articles from SCOPUS database published during 2013–22 are reviewed extracting information with respect to experimental design plans employed, materials machined, dielectrics used, process parameters and responses considered and MCDM tools applied along with their integration with other mathematical techniques. A detailed analysis of those reviewed articles reveals that the past researchers have mostly preferred Taguchi’s L9 orthogonal array as the experimental design plan; EDM oil as the dielectric fluid; medium and high carbon steels as the work materials; peak current and pulse-on time as the input parameters; material removal rate, tool wear rate and surface roughness as the responses; and grey relational analysis as the MCDM tool during conducting and optimizing EDM operations. This review paper would act as a data repository to the future researchers in understanding the stochastic behaviour of EDM processes and providing guidance in setting the tentative operating levels of varying input parameters along with achievable response values. The extracted dataset can be treated as an input to any of the machine learning algorithms for subsequent development of appropriate prediction models. This review also outlines potential future research avenues, emphasizing advancements in EDM technology and the integration of innovative multi-criteria decision-making tools.

Sustainable Electrochemical Mechanical Polishing (ECMP) for 4H-SiC wafer using chemical-free polishing slurry with hydrocarbon-based solid polymer electrolyte

Large-diameter single-crystal silicon carbide (SiC) wafers with good surface quality are desirable to dramatically improve the performance of SiC-based power electronic devices. However, the preparation of such wafers using a strong oxidative polishing slurry has the limitations of low material removal rate (MRR) and high environmental impact. Electrochemical–mechanical polishing (ECMP) initiates surface removal by electrochemical oxidation without using harsh chemicals, making it an alternative with low environmental impact to conventional polishing. Herein, we propose the use of polystyrene sulfonic acid (PSS), a hydrocarbon-based polymer electrolyte without perfluoro compounds, for ECMP. The SiC surface was effectively oxidized by ECMP using PSS. The effects of the polyelectrolyte crosslinking treatment on the lifetime of polyelectrolyte-incorporated nonwoven cloths were investigated. The optimum electrolytic and mechanical conditions yielded high MRRs of 37 and 14 µm/h for 2- and 4-inch SiC wafers. Moreover, the surface morphologies of the growth face and nonfacet areas was discussed. The wafer-scale evaluation demonstrated the roughness reduction to 0.85 nm Sa (average) after 15-min ECMP of the 4-inch wafer. The polishing-induced residual stress was investigated using microscale and wafer-scale Raman mapping. Overall, the proposed ECMP process offers significant advantages in terms of speed, cost-effectiveness, and environmental impact.

Optimization of femtosecond laser processing parameters of SiC using ANN-NSGA-II

Abstract In the field of femtosecond laser machining, it is essential to select the appropriate process parameters to obtain near thermal damage-free and high efficient machining of SiC wafer. In this work, a method of process parameter optimization for femtosecond laser machining of 4H-SiC was proposed by using the predictive ability of the Artificial Neural Network (ANN) and the optimization algorithm of the non-dominated sorting genetic algorithm (NSGA-II). Firstly, the femtosecond laser was used to fabricate microgrooves on SiC wafers, and the effects of process parameters (laser average power, scanning speed and repetition rate) on groove depth, width, heat affected zone (HAZ) and material removal rate (MRR) were investigated. Secondly, The ANN model is established based on experimental data. Other experiments verify the accuracy of the model, and the average error in the model's predictions is around 5%. Thirdly, Pareto optimal solutions are obtained by global optimization of the ANN model using the NSGA-II. The experimental results show that the Pareto optimal solutions are effective and reliable. This proposed method offers dependable guidance for the selecting and optimizing process parameters of high hardness and brittle materials in the field of femtosecond laser processing, and reduces the cost of selecting the appropriate processing parameters in the production process. The method can also be extended to other machining means, such as turning and milling.

Experimental Investigation and Optimization of Coaxiality Error Analysis with CNC Turning Process on Delrin for Assembly Fit

This paper mainly deals with the machining operation like turning operation, Material Removal Rate and Surface Roughness are the important parameter which is to be considered for quality product. The material selected for the experiment is DELRIN 500. Turning is one of the important processes that is widely used to create cylindrical components and it is also used for surface finish the product to make it smooth. Nowadays, plastic materials are widely used for making variety of components. To make a component with high dimensional accuracy, turning operation is used. The main concerns of turning are tooling cost and the effect of process on machinability characteristics. It can be seen that the output responses value has a minimum roughness average and a high degree of geometrical quality precision. High degree surface finish is induced by medium speed, feed rate, and small nose radius. The coaxial error was minimized by using medium speed, feed and larger nose radius. Experimentally found that third specimen (RPM -750) (FEED -0.08 mm/Rev) and (NOSE RADIUS 0.8) obtained minimum geometrical error along with minimum Surface roughness. Delrin is a crystalline plastic that offers an excellent balance of properties that bridge the gap between metals and plastics. Delrin possesses high tensile strength, creep resistance and toughness. It also exhibits low moisture absorption

Model for designing process strategies in ultrafast laser micromachining at high average powers

The great potential to scale the productivity of laser micromachining processes offered by the development of ultrafast lasers with average powers exceeding 1 kW comes at the cost of new physical and technical challenges. The large number of adjustable parameters, the different physical process limits, and the various possible optimization strategies complicate the design of suitable laser micromachining processes that enable high quality and high throughput.In this contribution, a comprehensive model is proposed that helps to identify the optimization strategies and process parameters required for the scaling of laser micromachining to high throughput at high average laser power without exceeding the process limits to ensure high surface quality.The model was experimentally verified on the example of laser micromachining of pockets in metals and silicon using an ultrafast laser with an average power of 1.01 kW, a pulse duration of 600 fs, different pulse energies, pulse bursts, and beam diameters. As a result, high material removal rates exceeding 120 mm3/min were achieved for silicon, stainless steel, copper, and aluminum, which exceed previously achieved removal rates by one to two orders of magnitude. The machined surfaces exhibit high quality as confirmed by the low measured roughness of about 1 µm.

Near Dry Powder Mixed Electric Discharge Machining of AA7050 Hybrid Composites Utilizing Composite Tool Materials

Electric Discharge Machining (EDM) is a prominent technique for producing components with complex geometry and intricate shapes. However, productivity depends on the selection of optimal input variables. This research investigates EDM and Near-Dry EDM (NDEDM) on AA7050 hybrid composites, exploring the impact of tool materials, Al2O3 powder concentration, and dielectric fluids on Material Removal Rate (MRR) and Surface Roughness (Ra). NDEDM offers a higher MRR due to localized heating and energy concentration, while hydrocarbon oil EDM provides lower Ra due to cooling and lubrication. Tool material properties, thermal conductivity and bonding strength, affect MRR and TWR. Aluminum composite tools via stir casting display a strong bonding interface, leading to efficient energy transfer and higher MRR, while copper composite tools via friction stir processing exhibit weaker bonding, potentially reducing MRR. Hydrocarbon oil significantly lowers TWR due to effective lubrication and cooling, while near dry EDM results in slightly higher TWR. Al2O3 powder enhances MRR, with an optimal concentration of 1 g/l, while higher concentrations hinder MRR. The presence of Al2O3 powder enhanced the surface quality by reducing the size of the recast layers and producing a smoother surface. Globules, craters, microcracks, pock marks and re-casted layers are the noticeable features spotted on the surface. The results were optimized using the TOPSIS optimization technique and process parameters 32 A current, 30 µs Ton, 3 mm gap distance machined under 3 g/l Al2O3 incorporated hydrocarbon oil dielectric medium using copper composites tool proffers optimal machining performance.

Research on Multi-Objective Process Parameter Optimization Method in Hard Turning Based on an Improved NSGA-II Algorithm

To address the issue of local optima encountered during the multi-objective optimization process with the Non-dominated Sorting Genetic Algorithm II (NSGA-II) algorithm, this paper introduces an enhanced version of the NSGA-II. This improved NSGA-II incorporates polynomial and simulated binary crossover operators into the genetic algorithm’s crossover phase to refine its performance. For evaluation purposes, the classic ZDT benchmark functions are employed. The findings reveal that the enhanced NSGA-II algorithm achieves higher convergence accuracy and surpasses the performance of the original NSGA-II algorithm. When applied to the machining of the high-hardness material 20MnCrTi, four algorithms were utilized: the improved NSGA-II, the conventional NSGA-II, NSGA-III, and MOEA/D. The experimental outcomes show that the improved NSGA-II algorithm delivers a more optimal combination of process parameters, effectively enhancing the workpiece’s surface roughness and material removal rate. This leads to a significant improvement in the machining quality of the workpiece surface, demonstrating the superiority of the improved algorithm in optimizing machining processes.

Machining Performance Analysis of Ti6-Al-4V in Powder-Mixed EDM Using Green Dielectric Oil

In this investigation, the factors were oil type ([Formula: see text]), powder type ([Formula: see text]), and powder concentration ([Formula: see text]) level, and the machining parameters were the voltage (V), capacitance (pf), pulse-on time ([Formula: see text]), and pulse-off time ([Formula: see text]). The output parameters were the material removal rate (MRR), and tool wear rate (TWR). Three types of oils and powders, along with their respective particle concentrations, were selected for this experiment. Neem, Karanja, and sesame oils were mixed with aluminum oxide (Al2O3), chromium (Cr), and nickel (Ni) powder at 4, 6, and 8[Formula: see text]gm/l, respectively, using an overhead stirrer to investigate the conditions under which the green oil powder mixed mixture yielded a higher MRR and lower TWR during the micro-EDM process of titanium alloy (Ti-6Al-4V) using a 0.8[Formula: see text]mm tungsten carbide (WC) tool with fixed machining parameters (240[Formula: see text]V, 100[Formula: see text]pf, 20[Formula: see text][Formula: see text]s [Formula: see text], and 5[Formula: see text][Formula: see text]s [Formula: see text]). All the powders used in this experiment had a 32-micron (450 mesh) size. The Taguchi L[Formula: see text] orthogonal array and grey relational analysis methods were used to design and optimize the input process parameters for both responses. SEM, optical microscopic analysis, and EDS analysis were performed to examine the surface morphology, characteristics of the WC tool, and elemental composition of both the tool and the workpiece material. The optimum conditions for Karanja oil with a Cr powder mixture concentration of 4[Formula: see text]g/l resulted in higher MRR and lower TWR. However, machining with neem and sesame oils resulted in higher TWR and lower MRR.

Towards artificial intelligence empowered performance enhancement of EDM process using nano-graphene mixed bio-dielectric supporting the carbon neutrality and sustainable development

The growing population with every passing day sets an alarming situation with respect to the conservation climate protocols. The increasing needs of society also demand a significant enhancement in the manufacturing capacity to augment the situation. However, it's a stringent requirement of the hour to propose sustainable and clean manufacturing processes to realize the goal of carbon neutrality to support a healthy life on the earth. Specifically, the processes that are energy intensive like electric discharge machining (EDM) are of serious concern regarding sustainability viewpoint. The role of the said process cannot be essentially eliminated as advent of new materials of superior characteristics demand the application of EDM for accurate cutting of intricate profiles. Nevertheless, the commonly used oil-based dielectric (kerosene) in EDM releases aerosol, deposit particles, oxides of carbon (CO2 & CO), thus contributing to the environmental contamination. It is pertinent to mention that industries are compelled to tune their processes to achieve the goals of Net-Zero. Therefore, this study thoroughly investigates the potential of nano-graphene mixed rice bran oil to make the EDM process cleaner and sustainable which has never been investigated so far. Moreover, the process has been successfully modeled using artificial neural network (ANN) and optimized by non-dominated sorting genetic algorithm-II (NSGA-II) which is another novel aspect of this study as it eradicates the need of extensive experimentation. Experimentation has been performed via Taguchi's experimental strategy followed by a detailed explanation of the findings based on process physics. In comparison to the traditional dielectric an improvement of 98.8% in material removal rate (MRR) and 93.9% reduction in specific energy consumption (SEC) are realized if the said novel combination is applied without compromising the quality. CO2 emissions determined for both rice bran oil and kerosene oil have revealed that rice bran oil provides 99.96% lesser CO2 emission in comparison to its counterpart.

Chemical treatment in 3D dental model production for clear aligners via additive manufacturing: a comprehensive evaluation

Purpose Stereolithography (SLA) additive manufacturing (AM) technique has enabled the production of inconspicuous and aesthetically pleasing orthodontics that are also hygienic. However, the staircase effect poses a challenge to the application of invisible orthodontics in the dental industry. The purpose of this study is to implement chemical postprocessing technique by using isopropyl alcohol as a solvent to overcome this challenge. Design/methodology/approach Fifteen experiments were conducted using a D-optimal design to investigate the effect of different concentrations and postprocessing times on the surface roughness, material removal rate (MRR), hardness and cost of SLA dental parts required for creating a clear customized aligner, and a container was constructed for chemical treatment of these parts made from photocurable resin. Findings The study revealed that the chemical postprocessing technique can significantly improve the surface roughness of dental SLA parts, but improper selection of concentration and time can lead to poor surface roughness. The optimal surface roughness was achieved with a concentration of 90 and a time of 37.5. Moreover, the dental part with the lowest concentration and time (60% and 15 min, respectively) had the lowest MRR and the highest hardness. The part with the highest concentration and time required the greatest budget allocation. Finally, the results of the multiobjective optimization analysis aligned with the experimental data. Originality/value This paper sheds light on a previously underestimated aspect, which is the pivotal role of chemical postprocessing in mitigating the adverse impact of stair case effect. This nuanced perspective contributes to the broader discourse on AM methodologies, establishing a novel pathway for advancing the capabilities of SLA in dental application.

A study on die-sinking electrical discharge machining behaviour of friction stir processed AA5052/WS2 surface composite

A study on die-sinking electrical discharge machining behaviour of friction stir processed AA5052/WS₂ surface composite

Evaluation of plasma parameters’ impact on MRR and surface roughness in plasma polishing of fused silica: An investigation of surface characterization

Advanced fabrication methods with high efficiency are needed to meet the rising demand for precision optical components. This study proposes a plasma-based isotropic polishing method to remove fused silica's surface defects and subsurface damaged layers. Hence, a medium-pressure plasma process is developed to serve this purpose. The qualitative analysis of the plasma and atom emission spectrum explores the plasma generation and scattering of the activated fluorine (F*) radicals inside the plasma chamber. Material removal rate (MRR) and surface roughness variations with different total pressures of 5–30 mbar and substrate dimensions (i.e., varying length between 5 and 45 mm with fixed width and thickness of 5 and 2 mm) at fixed process parameters, i.e., radio-frequency power of 80 W, pressure ratio of 1:1, gas composition of 90:10, have been investigated. The highest and lowest MRR, i.e., 0.039 mm3/min and 0.0061 mm3/min, are achieved at 5 and 30 mbar total pressures on substrates having lengths 45 and 5 mm, respectively. The highest and lowest percentage change in surface roughness observed are 86.4 % and 17.1 % at 5 mbar and 30 mbar for substrates with 5 and 45 mm lengths, respectively. Surface morphology analysis shows the micro-cracks are reduced after plasma processing. The presence of elements, i.e., F and C, through EDX analysis, confirms the reactions on the plasma processed surface. Moreover, The XPS findings demonstrated that minimal radicals could be incorporated onto the fused silica surface during He/O2/SF6 plasma processing. At lower pressure, the surface roughness value increases due to the increased excitation and ionization rate caused by more interaction between the fluorine and surface atoms. The MRR increases with substrate dimension with constant pressure. Larger substrate dimensions result in increased MRR at lower pressure.

Deep ensemble learning for material removal rate prediction in chemical mechanical planarization with pad surface topography

Chemical mechanical planarization (CMP) stands as a critical process in semiconductor manufacturing, necessitating precise prediction of material removal rate (MRR) to achieve optimal global planarization of material surface step heights. A notable hurdle in MRR prediction lies in the nonlinear pad-wafer interaction stemming from variations in the polishing pad surface. Despite numerous extant studies examining the asperity of the pad surface and the wafer-pad contact, aspects such as slurry lubrication and micro-topographical structure of the pad surface remain unaddressed. This study firstly proposes a deep ensemble learning-based approach for MRR prediction, leveraging the intricate interactions between the overall features of the pad surface and the polishing pressure. We experimentally generated diverse pad surfaces by controlling the process variables governing pad surface topography. The Multi-TabNet model, constructed by independently training multiple TabNet instances with distinct initialization and hyperparameters, furnishes a framework for MRR prediction across a broad spectrum of pad surface features, encompassing pores and cores. The results demonstrate a root mean square error of 80 Å/min, with the error margin reduced by up to 70 % compared to predictions utilizing a single parameter of the pad surface. Furthermore, it exhibits an error improvement exceeding 17 % relative to the existing method solely considering asperity. This study not only imparts insights for addressing MRR variations contingent on pad condition but also facilitates dependable material removal even on variable pad surfaces due to conditioner wear.

Atomic surface of silicon wafers induced by grafted silica nanoparticles and sodium carbonate

Atomic surface of silicon (Si) wafers is highly desired for integrated circuit (IC) industry. To meet this demand, traditional chemical mechanical polishing usually employs noxious slurries with complex compositions, containing oxidant, dispersant, inhibitor, surfactant, etc., resulting in the pollution of the environment and high-cost post-treatments. To settle this challenge, a novel green chemical mechanical polishing was developed, only including grafted silica nanoparticles and sodium carbonate. After CMP, surface roughness Sa of 0.219 and 0.247 nm are achieved on Si wafers using amine and methyl modified nanoparticles, respectively, corresponding to material removal rate (MRR) of 132.53 and 84.82 nm/min. Amino-modified nanoparticles increase 18.2 % MRR, and methyl modified ones reduce 24.4 % MRR, compared with those of unmodified nanoparticles. X-ray photoelectron spectroscopy reveals that amino-grafted nanoparticles have strong improvement for chemical reactions during CMP. Density functional theory calculations garner the electrostatic potential energy of 4.884 and 6.753 eV for amine and methyl-modified nanoparticles, respectively, corresponding to high and low effects for MRR. This work provides a new avenue to develop novel green simple CMP for Si wafers, reducing the pollution and cost for post-treatments.

Real-time prediction of material removal rate for advanced process control of chemical mechanical polishing

Polishing torque holds significance in monitoring the chemical mechanical polishing (CMP) process due to its close correlation with material removal. This study introduces a new model-based technique for estimating the material removal rate (MRR) using in-process data from CMP machines. The proposed method employs either the sequential least squares method or Kalman filter for real-time state estimation. Real-time estimation of MRR enables material removal control without relying on conventional endpoint detection (EDP) technologies. The accuracy of the proposed approach is validated through oxide CMP experiments, demonstrating precise estimation of the center-slowed MRR profile towards the end of the pad life.

Enhancing machining accuracy of banana fiber-reinforced composites with ensemble machine learning

Through innovative predictive modeling, this study advances Abrasive Water Jet Machining (AWJM) for banana fiber-reinforced biocomposites. Utilizing Artificial Neural Network (ANN) and Long Short-Term Memory (LSTM) models, hyperparameters are meticulously tuned for predicting Surface Roughness (Ra), Material Removal Rate (MRR), and Kerf Angle (Ka). Optimal configurations are identified, such as 3–6-6–4-1, 3–5-5–4-1, and 3–5–4–4-1 for ANN models, and 250, 100, and 50 units for LSTM models. Beyond individual models, the study explores stacking ensemble models, merging ANN and LSTM strengths with a Linear Regression final estimator, showcasing robust performance validated with high R-squared values. Regression models from Design Expert 13 contribute to understanding process parameter-outcome relationships. Optimized input parameters offer insights into minimizing surface roughness and kerf angle while maximizing MRR. This holistic approach, integrating advanced machine learning and ensemble learning, enhances predictive accuracy for banana-reinforced biocomposites, providing a versatile framework for diverse materials processing applications.