Analysis of Support Vector Regression for Approximation of Complex Engineering Analyses

2004 ◽  
Vol 127 (6) ◽  
pp. 1077-1087 ◽  
Author(s):  
Stella M. Clarke ◽  
Jan H. Griebsch ◽  
Timothy W. Simpson
Keyword(s):  
Support Vector Regression ◽  
Computer Code ◽  
Multivariate Adaptive Regression Splines ◽  
Support Vector ◽  
Test Bed ◽  
Computationally Efficient ◽  
Empirical Performance ◽  
Adaptive Regression ◽  
Complex Engineering ◽  
Function Approximations

A variety of metamodeling techniques have been developed in the past decade to reduce the computational expense of computer-based analysis and simulation codes. Metamodeling is the process of building a “model of a model” to provide a fast surrogate for a computationally expensive computer code. Common metamodeling techniques include response surface methodology, kriging, radial basis functions, and multivariate adaptive regression splines. In this paper, we investigate support vector regression (SVR) as an alternative technique for approximating complex engineering analyses. The computationally efficient theory behind SVR is reviewed, and SVR approximations are compared against the aforementioned four metamodeling techniques using a test bed of 26 engineering analysis functions. SVR achieves more accurate and more robust function approximations than the four metamodeling techniques, and shows great potential for metamodeling applications, adding to the growing body of promising empirical performance of SVR.

Analysis of Support Vector Regression for Approximation of Complex Engineering Analyses

2003 ◽  
Author(s):  
Stella M. Clarke ◽  
Jan H. Griebsch ◽  
Timothy W. Simpson
Keyword(s):  
Support Vector Regression ◽  
Computer Code ◽  
Multivariate Adaptive Regression Splines ◽  
Great Promise ◽  
Support Vector ◽  
Computationally Efficient ◽  
Adaptive Regression ◽  
Complex Engineering ◽  
Function Approximations ◽  
Adaptive Regression Splines

A variety of metamodeling techniques have been developed in the past decade to reduce the computational expense of computer-based analysis and simulation codes. Metamodeling is the process of building a “model of a model” that provides a fast surrogate for a computationally expensive computer code. Common metamodeling techniques include response surface methodology, kriging, radial basis functions, and multivariate adaptive regression splines. In this paper, we present Support Vector Regression (SVR) as an alternative technique for approximating complex engineering analyses. The computationally efficient theory behind SVR is presented, and SVR approximations are compared against the aforementioned four metamodeling techniques using a testbed of 22 engineering analysis functions. SVR achieves more accurate and more robust function approximations than these four metamodeling techniques and shows great promise for future metamodeling applications.

scholarly journals Prediksi Jumlah Pengiriman Barang Menggunakan Kombinasi Metode Support Vector Regression, Algoritma Genetika dan Multivariate Adaptive Regression Splines

2020 ◽  
Vol 7 (6) ◽  
pp. 1169
Author(s):  
Nendi Nendi ◽  
Arief Wibowo
Keyword(s):  
Support Vector Regression ◽  
Mean Absolute Percentage Error ◽  
Multivariate Adaptive Regression Splines ◽  
Percentage Error ◽  
Support Vector ◽  
Regression Splines ◽  
Absolute Percentage Error ◽  
Logistics Company ◽  
Adaptive Regression ◽  
Adaptive Regression Splines

<p>Sektor usaha logistik telah berkembang sangat pesat di Indonesia saat ini. PT. XYZ  adalah sebuah perusahaan logistik yang menyediakan jasa pengiriman barang dari satu tempat menuju ke tempat yang lain. Sebagai perusahaan logistik dengan jumlah kendaraan 2.100 unit armada truk dan akan terus bertambah seiring dengan target yang dicanangkan perusahaan, dimana pada 2020 jumlah armada truk harus mencapai 6.000 unit truk. Saat ini strategi operasional logistik dihasilkan berdasarkan pengalaman dari steakholder. Hal ini tentu tidak bisa dipertanggung jawabkan secara ilmiah. Prediksi jumlah pengiriman barang harian dapat menjadi solusi dalam membantu perusahaan dalam merencanakan, memonitoring dan mengevaluasi strategi operasional logistik. Hasil pengujian menunjukkan penggabungan metode <em>Support Vector Regression</em> (SVR), algoritma genetika dan <em>Multivariate Adaptive Regression Splines </em>(MARS) dapat menghasilkan prediksi jumlah pengiriman barang harian dengan nilai <em>Mean Absolute Percentage Error</em> (MAPE) yaitu 0.0969% dengan parameter <em>epsilon</em>(𝜀) 1.92172577675873E-20, <em>complexitas</em>(𝑐) 62 dan <em>gamma</em>(γ) 1.0.</p><p> </p><p class="Judul2"><strong><em>Abstract</em></strong></p><p class="Abstract"><em>The logistics business sector has developed very rapidly in Indonesia today. PT XYZ is a national logistics company that provides freight forwarding services from one place to another. As a national-scale logistics company, the company is supported by a fleet of 2,100 trucks. The number of fleets will continue to grow in line with the target set by the company, namely in 2020 the number of truck fleets must reach 6,000 trucks. Currently the logistics operational strategy is produced based on stakeholder experience, this certainly causes problems in the company's overall operations. Prediction of the number of daily goods shipments can be a solution in helping companies in planning, monitoring and evaluating logistical operational strategies, based on the company's ability in the availability of a fleet of vehicles for shipping. This study proposes a combination of Support Vector Regression (SVR) methods, genetic algorithms and Multivariate Adaptive Regression Splines (MARS) for problem solving in the prediction process, including in the selection of appropriate training data. The test results show that the combination of the three methods can produce predictions of the number of daily shipments with values of Mean Absolute Percentage Error (MAPE) 0.0969%, epsilon (𝜀) 1.92172577675873E- 20, complexity (𝑐) 62, and gamma (γ) 1.0.</em></p><p class="Judul2"><strong><em><br /></em></strong></p>

Integrating Least Square Support Vector Regression and Mode Pursuing Sampling Optimization for Crashworthiness Design

2011 ◽  
Vol 133 (4) ◽  
Author(s):  
Hu Wang ◽  
Songqing Shan ◽  
G. Gary Wang ◽  
Guangyao Li
Keyword(s):  
Support Vector Regression ◽  
Polynomial Regression ◽  
Computational Cost ◽  
Nonlinear Problems ◽  
Integrated Approach ◽  
Least Square ◽  
Multivariate Adaptive Regression Splines ◽  
Support Vector ◽  
Highly Nonlinear ◽  
Crashworthiness Design

Many metamodeling techniques have been developed in the past two decades to reduce the computational cost of design evaluation. With the increasing scale and complexity of engineering problems, popular metamodeling techniques including artificial neural network (ANN), Polynomial regression (PR), Kriging (KG), radial basis functions (RBF), and multivariate adaptive regression splines (MARS) face difficulties in solving highly nonlinear problems, such as the crashworthiness design. Therefore, in this work, we integrate the least support vector regression (LSSVR) with the mode pursuing sampling (MPS) optimization method and applied the integrated approach for crashworthiness design. The MPS is used for generating new samples which are concentrated near the current local minima at each iteration and yet still statistically cover the entire design space. The LSSVR is used for establishing a more robust metamodel from noisy data. Therefore, the proposed method integrates the advantages of both the LSSVR and MPS to more efficiently achieve reasonably accurate results. In order to verify the proposed method, well-known highly nonlinear functions are used for testing. Finally, the proposed method is applied to three typical crashworthiness optimization cases. The results demonstrate the potential capability of this method in the crashworthiness design of vehicles.

scholarly journals A COMPARATIVE STUDY OF DATA-DRIVEN MODELING METHODS FOR SOFT-SENSING IN UNDERGROUND COAL GASIFICATION

2019 ◽  
Vol 59 (4) ◽  
pp. 322-351
Author(s):  
Ján Kačur ◽  
Milan Durdán ◽  
Marek Laciak ◽  
Patrik Flegner
Keyword(s):  
Coal Gasification ◽  
Calorific Value ◽  
Multivariate Adaptive Regression Splines ◽  
Ex Situ ◽  
Gaussian Kernel ◽  
Support Vector ◽  
Underground Coal Gasification ◽  
Process Data ◽  
Adaptive Regression ◽  
Data Driven Modeling

Underground coal gasification (UCG) is a technological process, which converts solid coal into a gas in the underground, using injected gasification agents. In the UCG process, a lot of process variables can be measurable with common measuring devices, but there are variables that cannot be measured so easily, e.g., the temperature deep underground. It is also necessary to know the future impact of different control variables on the syngas calorific value in order to support a predictive control. This paper examines the possibility of utilizing Neural Networks, Multivariate Adaptive Regression Splines and Support Vector Regression in order to estimate the UCG process data, i.e., syngas calorific value and underground temperature. It was found that, during the training with the UCG data, the SVR and Gaussian kernel achieved the best results, but, during the prediction, the best result was obtained by the piecewise-cubic type of the MARS model. The analysis was performed on data obtained during an experimental UCG with an ex-situ reactor.

scholarly journals Hybrid Support Vector Machine to Preterm Birth Prediction

2018 ◽  
Vol 8 (2) ◽  
pp. 191
Author(s):  
Noviyanti Santoso ◽  
Sri Pingit Wulandari
Keyword(s):  
Support Vector Machine ◽  
Preterm Birth ◽  
Prediction Model ◽  
Research Area ◽  
Multivariate Adaptive Regression Splines ◽  
Human Reproduction ◽  
Support Vector ◽  
Premature Babies ◽  
Adaptive Regression ◽  
Adaptive Regression Splines

Preterm birth is one of the major contributors to perinatal and neonatal mortality. This issue became important in health research area especially human reproduction both in developed and developing country. In 2015 Indonesia rank fifth as the country with the highest number of premature babies in the world. The ability to reduce the number of preterm birth is to reduce risk factors associated with it. This research will be made the prediction model of preterm birth using hybrid multivariate adaptive regression splines (MARS) and Support Vector Machine (SVM). MARS used to select the attributes which suspected to affect premature babies. The result of this research is prediction model based on hybrid MARS-SVM obtains better performance than the other models

Sign in / Sign up

Export Citation Format

Share Document