Back Propagation Neural Network Architecture Research Articles

Texture and Flag Color Extraction in Backpropagation Neural Network Architecture

A flag is a rectangular or triangular piece of cloth or paper used as a symbol of the state, association, body, etc. or as a sign. It is often also used to symbolize a country to show its sovereignty. Along with the large number of countries, the country's flag also has many varieties and colors. The use of computers as a human aid is expected to the extent that the computer's ability can replace the limitations that humans have. Humans can recognize an object by using their eyes and brain, but if the eyes and brain cannot work properly it will hamper human work. In this research, training will be conducted on the Back Propagation Neural Network Architecture. Characteristic data for image recognition is obtained by extracting texture features and RGB color features. So that the network can recognize the flags by matching the feature data obtained from the training carried out. Characteristic data obtained from 24 data consisting of 16 training images and 8 testing images. From the results of the image network training can be identified properly, the accuracy rate of object identification is 87.50%. GUI users are able to identify flag images based on RGB text and color features.

RNN-Based Time Series Analysis for Wind Turbine Energy Forecasting

One significant source of renewable energy is wind power, which has the potential to generate sustainable energy. However, wind turbines have many challenges, such as high initial investment costs, the dynamic nature of wind speed, and the challenge of locating wind-efficient energy regions. Wind power predicting is crucial for effective planning of wind power generation, optimization of power generation, grid integration, and security of supply. Therefore, highly accurate forecasts ensure the efficient and sustainable operation of the wind energy sector and contribute to energy security, economic stability, and environmental sustainability. This study proposes a deep learning (DL) approach based on recurrent neural networks (RNNs) for long-term wind power forecasting utilizing climatic data. The input data that forms the basis of this study is obtained directly from a wind turbine system operating under real-world conditions. The proposed model in this study is based on a multilayer back-propagation neural network (RNN) architecture specifically designed to effectively handle complex data sets and time-dependent series. The architecture of the model is built on an RNN consisting of four separate layers, each with 50 hidden neurons, carefully structured to increase its capacity to capture complex features. To improve the robustness of the model and avoid overlearning, each RNN layer is followed by a dropout (regularizing) layer that randomly deactivates 20% of the neurons to enhance the generalization ability of the network. To finalize the prediction capability of the model, a linear function was chosen in the last layer to directly match the actual values. Evaluating the model performance metrics, the proposed architecture achieved a prediction accuracy of 91% R2 on the test dataset. The findings indicate that proposed method based on multilayer RNN can successfully capture the relationships between the sequential data of the wind turbine.

Dynamics of bioconvection radiative MHD flow with gyrotactic microorganisms: an intelligent computation approach

Bioconvection flows relate significantly to real-life problems, such as designing bio-conjugates, bio-microsystems, and bio-cells. This study investigates a mathematical characterization of the electrically time-dependent conducting fluid flow containing gyrotactic microorganisms with activation energy, mass, and heat transfer. It is assumed that the flow is towards the extended surface and affects thermal radiations. In model formulation, the Navier–Stokes equations are reduced to a system of ordinary differential equations (ODEs) using a similarity transformation approach. For reference solutions, the RK4 technique is used with Dorman Prince coefficients. We have designed a new backpropagation neural network architecture to formulate new solution models with Bayesian regularization. The effects of variations in two parameters, Schmidt number (sc) and Prandtl number (pr), are investigated. Our experimental outcomes are linked to the reference outputs. It was observed that there is a strong agreement between our solutions and the reference solutions. Our approach can be used for systems of fractional differential equations.

Photovoltaic Performance Using Modify Back Propagation Neural Network Based On 3 Different Solar Panels

Depletion of natural resources and increasing demand of power, sustainable energy sources such as photovoltaics (PV) are becoming increasingly more reliable and important. However, the production PV output power is unable to meet in terms of supply and demand due to irregular of solar irradiance and temperature daily. Therefore, a proper method/tool of prediction is very crucial to manage photovoltaic performance system and optimize the corresponding energy management system. In this paper, the modified Back Propagation Neural Network (BPNN) architecture is investigated. The investigating was conducted to provide more accurate prediction and error for short-term period prediction. The networkcompromises of 2 inputs parameter; the module temperature and solar irradiance and 1 output parameter; AC Power. Early research was conducted based on forecast time period on May 2021 starting from 7am to 8pm (14 hour) daily from 3 different panels; monocrystalline, polycrystalline and thin film, which are located on the rooftop of PEARL lab building. The result obtained shows that polycrystalline panel performed the best compare to monocrystalline and thin film panels.

Development of rain attenuation prediction in south west Nigeria on terrestrial link using artificial neural network

The predicted precipitation accuracy in light of the current global climate change is very important. The Back-Propagation Neural Network (BPNN) technique then utilized the Artificial Neural Network (ANN) to correctly forecast rainfall. The study is based on data from three Nigerian terrestrial microwave connections operating at frequencies of 23 and 38 GHz. At 23 GHz, a normal distribution with an average of zero is the best represented fade slope distribution, contrary to the model ITU-R earth to satellite rain fade slope. Based on the data analysis observed at 23 GHz, a novel prediction model is developed. The suggested model is validated using 38 GHz fade slope data and shown to work well with nearly all attenuation levels. A chi-square fitness test is used to further validate the model. The suggested model will be critical in the development of rainfall mitigation methods for tropical terrestrial connections. The findings of the research indicate that BPNN models may be utilized as a prediction algorithm that offers a high predictive accuracy of three types of design: 500, 1000 and 1,500.The experiment examined data on precipitation using the BPNN Architecture's two-hidden layers, from three periods of time [2-50-10-1, 500]. The average square error is used for categorization work performance evaluation. The results of the experiment indicate that the design of the 100-year-old [2-50-20-1, MSE of 0 00096341] is excellent. Moreover, the BPNN algorithm provides an efficient Southwestern Nigeria predictor model.

Artificial neural network modelling for average surface roughness in citrate stabilised electroless nickel boron coatings

In this study, electroless nickel boron coatings were developed on Al 7075 aluminium alloy from the citrate stabilised electrolyte bath to investigate the applicability of artificial neural network (ANN) in the prediction of the average surface roughness Ra of the coatings. The process parameters such as concentrations of nickel, reducing agent and stabiliser were considered as input parameters and Ra was selected as an output parameter. Two different feed forward back propagation neural network architectures 3-10-1 and 3-5-5-1, were constructed and trained using Levenberg-Marquardt (LM) algorithm to predict the Ra. The accuracy of the networks was compared with the predicted coefficient of determination (R2) and mean squared error (MSE). The R2 and MSE of 0.9998 and 6.32E-05 were obtained for 3-5-5-1 against 0.9850 and 6.29 E-04 for 3-10-1 network architecture, which proved that the multi-layer perceptron (3-5-5-1) fits well for the experimental data presented.

Modeling and optimization of A-GTAW process using back propagation neural network and heuristic algorithms

Apart from different merits of using conventional gas tungsten arc welding (C-GTAW) process, some demerits have been introduced among which shallow penetration is the most important ones. In order to cope with the mentioned disadvantage, some procedures have been proposed among which using a paste like coating of activating flux during welding process known as activated-GTAW (A-GTAW) is the most extensively used ones. In this study effect of the most important process variables (welding current (C), welding speed (S)) and percentage of activating fluxes (TiO2 and SiO2) combination (F) on the most important quality characteristics (depth of penetration (DOP), weld bead width (WBW), and consequently aspect ratio (ASR)) in welding of AISI316L austenite stainless steel parts have been considered. To gather the required data for modeling and optimization purposes, box-behnken design (BBD) in design of experiments (DOE) approach has been used. In order to establish a relation between process input variables and output characteristics, back propagation neural network (BPNN) has been employed results of which have been compared with regression modeling outputs. Particle swarm optimization (PSO) algorithm has been used for determination of BPNN architecture (number of hidden layers and neurons/nodes in each hidden layer). Dragonfly (DFA) and PSO algorithms have been employed for process optimization in such a way that desired AR, minimum WBW, and maximum DOP achieved simultaneously. Finally, confirmation experimental tests have been carried out to evaluate the performance of the proposed method. Based on the results, the proposed procedure is efficient in modeling and optimization (with less than 3% error) of A-GTAW process.

Development of rain attenuation prediction in south west Nigeria on terrestrial link using artificial neural Network

The predicted precipitation accuracy in light of the current global climate change is very important. The Back-Propagation Neural Network (BPNN) technique then utilized the Artificial Neural Network (ANN) to correctly forecast rainfall. The study is based on data from three Nigerian terrestrial microwave connections operating at frequencies of 23 and 38 GHz. At 23 GHz, a normal distribution with an average of zero is the best represented fade slope distribution, contrary to the model ITU-R earth to satellite rain fade slope. Based on the data analysis observed at 23 GHz, a novel prediction model is developed. The suggested model is validated using 38 GHz fade slope data and shown to work well with nearly all attenuation levels. A chi-square fitness test is used to further validate the model. The suggested model will be critical in the development of rainfall mitigation methods for tropical terrestrial connections. The findings of the research indicate that BPNN models may be utilized as a prediction algorithm that offers a high predictive accuracy of three types of design: 500, 1000 and 1,500. The experiment examined data on precipitation using the BPNN Architecture's two-hidden layers, from three periods of time [2-50-10-1, 500]. The average square error is used for categorization work performance evaluation. The results of the experiment indicate that the design of the 100-year-old [2-50-20-1, MSE of 0 00096341] is excellent. Moreover, the BPNN algorithm provides an efficient Southwestern Nigeria predictor model.

Comparison of RSM and ANN Optimization Techniques and Modeling of Ultrasonic Energy Assisted Trans-esterification of Salviniaceae Filiculoides Oil Blended to Biodiesel

For the transesterification of biodiesel from Azolla oil, the safe and successful use of feed stocks is a very significant prerequisite. It is of high importance to determine the optimal reaction parameters to maximize the yield of low-cost biodiesel generated from Azolla oil. Ultrasonic energy was used in this work for the development of biodiesel from Azolla oil catalyzed by the KOH catalyst under different conditions. The effect on the transesterification of Azolla Oil to biodiesel of four reaction parameters, namely the methanol/Azolla oil molar ratio (A), KOH catalyst concentration (B), reaction time (C) and reaction temperature (D) were considered. In order to optimize the effects of reaction parameters for the transesterification of Azolla oil to biodiesel, response surface methodology (RSM) based on central composite rotatable design (CCRD) is applied. To obtain a good correlation between the input reaction parameters and the output response parameter (FAME yield) from Azolla oil to biodiesel, an artificial neural network (ANN) model with two feed-forward back-propagation neural-network architecture Multilayer Perceptron Network (MLP) and Radial Basis Function Network (RBFN) was developed. With the experimental information obtained from the RSM model, the built ANN models were trained and evaluated. Absolute Average Deviation (AAD), Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE) and coefficient of determination were statistically compared with the predictive capacity of both RSM and ANN models (R2). The statistical analysis showed that the measured FAME yield from both the RSM and ANN models was able to predict the FAME yield, and the findings limited the ANN model to the much more reliable FAME yield prediction compared to the RSM model.

BPNN-ACO application on minimization of hole delamination during GFRP drilling process

Delamination defects during the drilling process on the glass fiber reinforced polymer (GFRP) have a great contribution to the component failure. Hence, it is necessary to properly choose the combination of machining variables to minimize hole entry delamination (EnDel) and hole exit delamination (ExDel) during drilling process simultaneously. This study underlines the modelling and minimizing the EnDel and hole ExDel during GFRP drilling process by combining a backpropagation neural network (BPNN) method and ant colony optimization (ACO). The varied drilling parameters were type of drill point angle, feeding speed, and cutting speed. The optimum BPNN architecture could be obtained by using 3-4-2-5-8-2 network architecture with tansig activation function. The optimum GFRP drilling parameters that can minimize EnDel and ExDel simultaneously were 116° of drill point angle, 51.3 mm min−1 of feeding speed and 4975 rpm of spindle speed.

Minimization of thrust force and surface roughness during MQL coolant drilling on tool steel using BPNN-ACO

The excessive thrust force that generated during the minimum quantity lubrication (MQL) drilling process of tool steel can lower the hole surface quality. Hence, it is necessary to properly choose the combination of machining variables to minimize thrust force (TF) and hole surface roughness (HSR) simultaneously. This study underlines the modelling and minimizing the thrust force and hole surface roughness developed during MQL drilling process by integrating a backpropagation neural network (BPNN) method and ant colony optimization (ACO). The varied drilling parameters were type of drill bit, drill point angle, feeding speed, and cutting speed. The optimum BPNN architecture could be obtained by using 4-20-2 network architecture with tansig activation function. The optimum MQL drilling parameters that can minimize TF and HSR simultaneously were HSS M2 drill bit, 107° of drill point angle, 0.045 mm/rev of feeding speed and 36 m/min of cutting speed.

Assessment of Earthquake Destructive Power to Structures Based on Machine Learning Methods

This study presents a machine learning-based method for the destructive power assessment of earthquake to structures. First, the analysis procedure of the method is presented, and the backpropagation neural network (BPNN) and convolutional neural network (CNN) are used as the machine learning algorithms. Second, the optimized BPNN architecture is obtained by discussing the influence of a different number of hidden layers and nodes. Third, the CNN architecture is proposed based on several classical deep learning networks. To build the machine learning models, 50,570 time-history analysis results of a structural system subjected to different ground motions are used as training, validation, and test samples. The results of the BPNN indicate that the features extraction method based on the short-time Fourier transform (STFT) can well reflect the frequency-/time-domain characteristics of ground motions. The results of the CNN indicate that the CNN exhibits better accuracy (R2 = 0.8737) compared with that of the BPNN (R2 = 0.6784). Furthermore, the CNN model exhibits remarkable computational efficiency, the prediction of 1000 structures based on the CNN model takes 0.762 s, while 507.81 s are required for the conventional time-history analysis (THA)-based simulation. Feature visualization of different layers of the CNN reveals that the shallow to deep layers of the CNN can extract the high to low-frequency features of ground motions. The proposed method can assist in the fast prediction of engineering demand parameters of large-number structures, which facilitates the damage or loss assessments of regional structures for timely emergency response and disaster relief after earthquake.

Assess Reliability Parameters of an Electronic Voting Machine using a Neural Network Technique

The structure of Electronic Voting Machine (EVM) is an interconnected network of discrete components that record and count the votes of voters. The EVM system consists of four main subsystems which are Mother board of computer, Voting keys, Database storage system, power supply (AC and DC) along with various conditions of functioning as well as deficiency. The deficiency or failure of system is due to its components (hardware), software and human mismanagement. It is essential to reduce complexity of interconnected components and increase system reliability. Reliability analysis helps to identify technical situations that may affect the system and to predict the life of the system in future. The aim of this research paper is to analyze the reliability parameters of an EVM system using one of the approaches of computational intelligence, the neural network (NN). The probabilistic equations of system states and other reliability parameters are established for the proposed EVM model using neural network approach. It is useful for predicting various reliability parameters and improves the accuracy and consistency of parameters. To guarantee the reliability of the system, Back Propagation Neural Network (BPNN) architecture is used to learn a mechanism that can update the weights which produce optimal parameters values. Numerical examples are considered to authenticate the results of reliability, unreliability and profit function. To minimize the error and optimize the output in the form of reliability using gradient descent method, authors iterate repeatedly till the precision of 0.0001 error using MATLAB code. These parameters are of immense help in real time applications of Electronic Voting Machine during elections.

High-accuracy Automated Diagnosis of Parkinson's Disease

Parkinson's disease (PD), which is the second most common neurodegenerative disease following Alzheimer's disease, can be diagnosed clinically when about 70% of the dopaminergic neurons are lost and symptoms are noticed. Neuroimaging methods such as single photon emission computed tomography have become useful tools in vivo to assess dopamine transporters (DATs) in the striatal region. However, inter- and intra-reader variability of construing the images might result in misdiagnosis. To overcome the challenges posed by classification of the disease, image preparation techniques and a back propagation neural network (BPNN) have been proposed. The aim of this study is to show that the proposed method can be used for the classification of PD with high accuracy. In this study, we used basic image preparation techniques and a BPNN on DAT imaging datasets from the Parkinson's Progression Markers Initiative. 1,334 PD and 212 normal control (NC) subjects were included. In the image preparation phase, adaptive histogram equalization was applied to the cropped images, followed by image binarization. Then, the mass-difference method was applied to separate the regions of interest with similar values. Finally, the binarized images were subtracted from the original images, and the average pixel per node approach was applied to the images to minimize the inputs. In the BPNN phase, 400 input neurons and 2 output neurons were used. The dataset was divided into three sets: training, validation, and test. The BPNN was trained several times in order to obtain the optimum values. The use of 40 hidden neurons, a learning rate of 0.00079, and a momentum factor of 0.90 produced superior results and were applied in the final BPNN architecture. The tolerance value used was 0.80. Uniquely, we found the sensitivity, specificity, and accuracy for PD vs. NC classification to be 99.7%, 99.2%, 99.6%, respectively. To the best of our knowledge, this is the highest accuracy value achieved in the existing literature. Our method increases computational speed together with improved performance. We have shown that effective image processing methods and the use of BPNN can successfully be applied to PD datasets to accurately determine any abnormalities in DATs. Using the shallow neural network, this procedure requires less processing time compared to other methods, and its accuracy, sensitivity, and specificity are reliable. However, further studies are needed to establish a prediction method for the preclinical and prodromal stages of the disease.

Multi-objective optimization in wire-EDM process using grey relational analysis method (GRA) and backpropagation neural network–genetic algorithm (BPNN–GA) methods

PurposeThe purpose of this paper is to investigate prediction and optimization of multiple performance characteristics in the wire electrical discharge machining (wire-EDM) process of SKD 61 (AISI H13) tool steel.Design/methodology/approachThe experimental studies were conducted under varying wire-EDM process parameters, which were arc on time, on time, open voltage, off time and servo voltage. The optimized responses were recast layer thickness (RLT), surface roughness (SR) and surface crack density (SCD). Arc on time was set at two different levels, whereas the other four parameters were set at three different levels. Based on Taguchi method, an L18 mixed-orthogonal array was selected for the experiments. Further, three methods, namely grey relational analysis (GRA), backpropagation neural network (BPNN) and genetic algorithm (GA), were applied separately. GRA was performed to obtain a rough estimation of optimum drilling parameters. The influences of drilling parameters on multiple performance characteristics were determined by using percentage contributions. BPNN architecture was determined to predict the multiple performance characteristics. GA method was then applied to determine the optimum wire-EDM parameters.FindingsThe minimum RLT, SR and SCD could be obtained by setting arc on time, on time, open voltage, off time and servo voltage at 2 ms, 3 ms, 90 volt, 10 ms and 38 volt, respectively. The experimental confirmation results showed that BPNN-based GA optimization method could accurately predict and significantly improve all of the responses.Originality/valueThere were no publications regarding multi-response optimization using a combination of GRA and BPNN-based GA methods during wire-EDM process available.

Optimized neural network using beetle antennae search for predicting the unconfined compressive strength of jet grouting coalcretes

SummaryThis investigation studied the coalcrete, a new supporting material produced by jet grouting (JG) for supporting surrounding coal seams. For support design, the unconfined compressive strength (UCS) of the coalcrete is an essential parameter to evaluate the jet grouting effect in coal mines. In this study, an intelligent technique was proposed for predicting the UCS of the coalcrete by combining back propagation neural network (BPNN) and beetle antennae search (BAS). The architecture of BPNN was first tuned by BAS, and then, the optimized BPNN‐BAS model was subsequently used for nonlinear relationship modeling. Several crucial influencing variables including water‐cement ratio, coal‐grout ratio, and curing time were selected as the inputs. By combining these variables, 360 coalcrete samples were prepared in a controlled laboratory environment and tested for establishing the dataset. The results demonstrate that BAS can tune the BPNN architecture more efficiently compared with random selection. Moreover, in comparison with multiple regression (MLR) and logistic regression (LR), and support vector machine (SVM), the optimized BPNN‐BAS model is more reliable and accurate for predicting coalcrete strength. Sensitivity analysis (SA) was used to obtain the variable importance, and the results demonstrate that curing time affects the UCS of the coalcrete most strongly, followed by water‐cement ratio and coal‐grout ratio. The success of this study provides an accurate and brief approach to coalcrete strength prediction.

Performance of back-propagation neural network in chaotic data time series forecasting and evaluation over parametric forecast: a case study for rainfall-runoff modelling over a river basin

Back-propagation neural network (BPN) is sufficiently suitable for forecasting of Chaotic data time series by two approaches i.e., parametric forecasting and time-series forecasting. However, conformation of its optimum architecture for a specific case is pre-requisite. Total rainfall-runoff (R–R) from Basantpur station over Mahanadi river basin was in under study. Initially modelling of R–R by parametric forecast approach was done by optimum architecture of BPN. It is found that BPN performs excellent for the months of July, August, and October. For the September performance was drastically unfortunate. Modelling was denied by hypothesis \(MAD\,\left( {\% \,of\,LPA} \right)\,<\,{\raise0.7ex\hbox{$1$} \!\mathord{\left/ {\vphantom {1 {2 }}}\right.\kern-0pt} \!\lower0.7ex\hbox{${2 }$}} SD (\% \,of\,LPA)\). By 20,000,000 epochs of training \(MAD\,\left( {\% \,of\,LPA} \right)\) were 23.40 and \(SD\,\left( {\% \,of\,LPA} \right)\) were 33.84. Therefore second approach of modelling i.e., time-series forecasting was applied. In which, ‘n’ years monthly past recorded R–R over the station is used to forecast of \((n + 1)\)th year monthly R–R over the station. This was significantly found appropriate and better evaluated over parametric approach. The model was highly acceptable under the hypothesis wherein \(MAD\,\left( {\% \,of\,LPA} \right)\) was 2.4516164772862348E-6 and \(SD\,\left( {\% \,of\,LPA} \right)\) was 33.45 and corelation coefficient (CC) between actual and predicted R–R is obtained 1.0 in training and 0.82 in testing independently. In this paper detail design of BPN in time-series forecasting, optimization of its architecture, training with 35 years data sets, testing with 7 years data sets, its evaluation over BPN in parametric forecast is discussed.

Vibration-Signature-Based Inter-Turn Short Circuit Identification in a Three-Phase Induction Motor Using Multiple Hidden Layer Back Propagation Neural Networks

Inter-turn short circuits in stator windings area fairly common fault in induction motors. Early detection of this type of fault will greatly assist in sustaining production processes in manufacture. This paper proposes a method to detect inter-turn short circuits in stator windings at an early stage. The proposed method consists of four steps: (1) preprocessing by decomposing the signal into detail and approximation signals using a wavelet transform, (2) converting the first detail signal into a frequency-based signal using fast Fourier transform, (3) calculating the values of statistical features for the signal in time and frequency domains and (4) identifying faults using a back propagation neural network (BPNN). Using BPNN architecture with 3 hidden layers and 75 neurons per layer, the identified recognition rate was96.67% with a mean square error of 1.39×10-5. The validity of the proposed method is excellent based on a receiver operating characteristic analysis, with a precision level of 94.66%.

Effective tourist volume forecasting supported by PCA and improved BPNN using Baidu index

The precise forecasting of tourist volume is a very challenging task. This paper aims to propose an effective model named PCA-ADE-BPNN for forecasting tourist volume based on Baidu index. The principal component analysis (PCA), a dimensional reduction, is employed to decorrelate the input data before training a back propagation neural network (BPNN) architecture, and the adaptive differential evolution algorithm (ADE) is for getting global optimization of BP network's weight values and threshold values to enhance the forecasting performance of BPNN. The PCA-ADE-BPNN model is a new combination of a dimensional reduction algorithm, an optimization algorithm, and a neural network. The validity of this model is demonstrated by conducting case studies of Beijing City and Hainan Province, China. The results indicate the proposed PCA-ADE-BPNN always outperforms other models in terms of forecasting accuracies. Therefore, the proposed PCA-ADE-BPNN is a potential candidate for the effective forecasting of tourist volume.

Multi-Objective Optimization in End Milling Process of ASSAB XW-42 Tool Steel with Cryogenic Coolant Using Grey Fuzzy Logic and Backpropagation Neural Network-Genetic Algorithm (BPNN-GA) Approaches

This study investigates the prediction and optimization of multiple performance characteristics in the end milling process of ASSAB XW-42 tool steel, i.e., surface roughness (SR), tool flank wear (VB) and material removal rate (MRR). The quality characteristics of SR and VB was the smaller the better, while MRR was the larger the better. Three methods, namely grey fuzzy logic, back propagation neural network (BPNN) and genetic algorithm (GA), were separately applied. The experimental studies were conducted by varying the end milling process parameters (cutting speed, feed rate and axial depth of cut) and liquid nitrogen cooling flow rate. Grey fuzzy logic was used to obtain a rough estimation of the optimum end milling parameters. The influences of end milling parameters on multiple performance characteristics were determined by using percent contributions. BPNN architecture was determined to predict the multiple performance characteristics. GA method was then applied to determine the optimum end milling parameters. The results of confirmation experiments showed a good agreement with the predicted responses.