Start Date
13-5-2021 2:50 PM
End Date
13-5-2021 3:10 PM
Document Type
Full Paper
Keywords
modeFRONTIER, Optimization, Design, Manufacturing, Commercial, Industry
Description
Optimization is a vital tool to advance the world around us. In order to optimize a variety of design problems, the industry implements the use of a commercial optimization software named modeFRONTIER. It was initially developed as a collaboration between universities, research companies, and private industries in Europe. This software can streamline the design process via its workflow-based environment, multi-objective optimization algorithms, and post-processing tools delivering statistical analysis and data visualization. This provides engineers and researchers with the necessary information for effective decision making in the design process. modeFRONTIER can be coupled with other design tools such as Computer Aided Design (CAD), Finite Element Analysis (FEA), Structural Analysis (SA), Computational Fluid Dynamics (CFD), Computer Aided Engineering (CAE), and other know commercial software to solve for a desired optimum. Throughout this analysis, one single-objective and one multi- objective function was tested to evaluate the performance of this software. Each case used the implementation of response surfaces for analysis fitted from a Sobol’s algorithm approach using Radial Basis Functions. The Ackley single-objective function was evaluated with two gradient based optimizers which were the Levenberg-Marquardt (LM) and Bounded-BFGS (B-BFGS) optimizers. In addition, the Poloni multi-objective function was evaluated with two non-gradient based optimizers which were the Multi-Objective Genetic Algorithm II (MOGA-II) and Non- Dominated Sorting Genetic Algorithm II (NSGA-II) optimizers. This analysis resulted in close to optimal objective values for each function tested. Additionally, to discuss the importance of this software’s usage in industrial problems and applications, two research studies were identified and discussed. When compared to additional software and open source programs, that focuses on specified and individualized algorithms and optimizers, modeFRONTIER outperforms them all. This is done through an all-inclusive incorporation of the best optimizers developed, such as the ones based on the Genetic Algorithm. This makes for a highly beneficial effective tool in making design decisions as it reduces design cycle time, improves product performance, and reduces development time overall. These benefits are exhibited in a vast number of industries such as the Aerospace, Architecture Engineering & Construction, Automotive, Biotech & Pharma, Consumer Goods, Electronics, Energy & Environment, Industrial equipment, Marine, and the Robotics industry.
DOI
https://doi.org/10.5038/GVXG7608
Optimization Techniques using modeFRONTIER - Process Review
Optimization is a vital tool to advance the world around us. In order to optimize a variety of design problems, the industry implements the use of a commercial optimization software named modeFRONTIER. It was initially developed as a collaboration between universities, research companies, and private industries in Europe. This software can streamline the design process via its workflow-based environment, multi-objective optimization algorithms, and post-processing tools delivering statistical analysis and data visualization. This provides engineers and researchers with the necessary information for effective decision making in the design process. modeFRONTIER can be coupled with other design tools such as Computer Aided Design (CAD), Finite Element Analysis (FEA), Structural Analysis (SA), Computational Fluid Dynamics (CFD), Computer Aided Engineering (CAE), and other know commercial software to solve for a desired optimum. Throughout this analysis, one single-objective and one multi- objective function was tested to evaluate the performance of this software. Each case used the implementation of response surfaces for analysis fitted from a Sobol’s algorithm approach using Radial Basis Functions. The Ackley single-objective function was evaluated with two gradient based optimizers which were the Levenberg-Marquardt (LM) and Bounded-BFGS (B-BFGS) optimizers. In addition, the Poloni multi-objective function was evaluated with two non-gradient based optimizers which were the Multi-Objective Genetic Algorithm II (MOGA-II) and Non- Dominated Sorting Genetic Algorithm II (NSGA-II) optimizers. This analysis resulted in close to optimal objective values for each function tested. Additionally, to discuss the importance of this software’s usage in industrial problems and applications, two research studies were identified and discussed. When compared to additional software and open source programs, that focuses on specified and individualized algorithms and optimizers, modeFRONTIER outperforms them all. This is done through an all-inclusive incorporation of the best optimizers developed, such as the ones based on the Genetic Algorithm. This makes for a highly beneficial effective tool in making design decisions as it reduces design cycle time, improves product performance, and reduces development time overall. These benefits are exhibited in a vast number of industries such as the Aerospace, Architecture Engineering & Construction, Automotive, Biotech & Pharma, Consumer Goods, Electronics, Energy & Environment, Industrial equipment, Marine, and the Robotics industry.