Mo H.,South China University of Technology |
Li Z.,FernUniversitaet in Hagen |
Du Q.,South China University of Technology
IJCCI 2015 - Proceedings of the 7th International Joint Conference on Computational Intelligence | Year: 2015
In line with the theory of schema sampling, a hypothesis could be made that sufficient supply of loworder building blocks (BBs) was one of the necessary conditions for a genetic algorithm(GA) to work. A consequential question of this hypothesis regards, when a certain fitness function is optimized with a commonly used GA, whether it is rare or common that there are plenty of low-order BBs. It is remarked that, when a base-m encoded GA is applied to a fitness function that is linearly combined of sinusoidal basis functions with integral frequencies, it is unlikely to obtain order-1 BBs with fixed positions at multiple loci, i.e., it is rare that there are plenty of order-1 BBs. However, if a considerable part of the sinusoidal basis functions are with frequencies exponential to a positive integer m, a base-m encoding can provide relatively more order-1 BBs compared with the encodings with cardinalities other than m. Copyright © 2015 by SCITEPRESS - Science and Technology Publications, Lda. All rights reserved.
Mo H.,South China University of Technology |
Li Z.,FernUniversitaet in Hagen |
Tian L.,South China University of Technology |
Tian X.,South China University of Technology
International Journal of Computational Intelligence Systems | Year: 2015
Abstract: Encoding plays a key role in determining the optimization efficiency of a genetic algorithm. In the optimization of a continuous function, binary encodings are normally used due to their low coding-alphabet cardinalities. Nevertheless, from the viewpoint of building-block supply, it is remarked that a binary encoding is not necessarily the best choice to express a fitness function which is linearly combined of sinusoidal functions with frequencies exponential to a positive integer m when m is not equal to 2. It is proved that, if the frequencies are exponential to m, an encoding of cardinality m can provide a better supply of order-1 building blocks than the encodings of other cardinalities. Taking the advantage of building-block supplies, a genetic algorithm with an encoding of cardinality m has higher chance to find fitter solutions. This assumption is verified via a number of randomly generated fitness functions, and encodings with different cardinalities are compared according to the optimization performance of corresponding genetic algorithms on these fitness functions. The simulation results support the assumption, and show in the statistical sense that the genetic algorithm with an encoding of cardinality m outperforms those of the other cardinalities when the frequencies of the sinusoidal functions are exponential to m. © 2015,the authors.
Agency: Cordis | Branch: FP7 | Program: CP | Phase: ICT-2009.4.3 | Award Amount: 10.33M | Year: 2010
The goal of SMART VORTEX is to provide a technological infrastructure consisting of a comprehensive suite of interoperable tools, services, and methods for intelligent management and analysis of massive data streams to achieve better collaboration and decision making in large-scale collaborative projects concerning industrial innovation engineering.\n\nSMART VORTEX captures the tractable product data streams in the product lifecycle of design and engineering. In each phase of this lifecycle, different streams of product data are generated. Amongst other, these product data streams contain streams from sensors (data rates of Gigabytes per second), simulation, experimental, and testing data (millions of complex data sets), design data (complex and exchanged between different domains), multi-media collaboration data (heterogeneous, and high information density), and higher level inferred events generated by analyses. These data streams are produced and consumed in all phases of the product lifecycle. The large volume of data in these streams makes the detection of pertinent information a hard problem for both technological infrastructures and humans. SMART VORTEX uses a Data Stream Management for managing, searching, annotating, analysing and performing feature extraction on these data streams.\n\nWithin the lifecycle of design and engineering projects a large number of people need to collaborate in order to achieve the individual project goals, such as bringing the next generation flat panel TV to the market before the competition does, identifying opportunities for improvements of existing products, or the maintenance of products in use. These projects are basically large distributed collaborative processes, where people from different domains of expertise and different organizations have to work together. SMART VORTEX supports these people, systems, and products with collaborative tools and decision support systems managing the constantly produced massive product data streams. SMART VORTEX ensures the efficiency and success of the collaboration by delivering the pertinent information at the right moment.
Agency: Cordis | Branch: FP7 | Program: CSA | Phase: INFRA-2007-3.3;INFRA-2007-3.0-03 | Award Amount: 1.80M | Year: 2008
The FP7 e-Infrastructures initiative has already resulted in a number of projects being set-up or in negotiation covering a range of areas of scientific endeavour. The creation of virtual research communities brings together stakeholders with diverse backgrounds and capabilities. One consequence is a large diversity of approaches and there is as yet little consensus across these communities about best practices in maintaining permanent access to the records of science in a sustainable way. The purpose of this proposal is to work with providers and users of scientific information and repositories to deliver insight into the issues of sustainable permanent access and provide cross-fertilisation of ideas and requirements between providers and users, between the various sectors of interest they represent and between the research community as a whole and national/European funding agencies. There will be six key outcomes for the EU - it will provide: - a roadmap for the support e-infrastructure for long-term accessibility and usability of scientific and other digital information in Europe. - insight into current and planned research and development relating to e-infrastructures and permanent access, national, European and global, regardless of the funding mechanism identification of gaps in the existing and planned infrastructure and the research agenda the ability to share and capitalise on best practices as well as understanding the impact that e-Science is having on the research communities that it is serving, through an insight study into the capabilities and practices within the various research communities - better-informed investment decisions and sustainable e-repositories through an impact analysis framework and tool based on these insights - an international process for evaluating the sustainability and trustworthiness of digital repositories, and identifying best practice
Agency: Cordis | Branch: FP7 | Program: CP | Phase: FoF-ICT-2011.7.1 | Award Amount: 11.74M | Year: 2012
The production and ramp-up of complex and highly customized products are exceptionally challenging for planning and control, especially in small lot sizes. Daily challenges like late requests for change, immature high technology products and processes create significant risks. The occurring risks are bigger than production of big series such as automotive. Thus, new ICT-based approaches are required. The aim is to develop mitigation strategies to respond faster to unexpected events. Therefore the knowledge base has to be enriched for real-time decision support, to detect early warning and to accelerate learning. Our approach is based on a new generation of service orientated enterprise information platforms, a service orientated bus integrating service-based architecture and knowledge-based multi-agent systems (MAS). A holonic MAS combined with a service architecture will improve performance and scalability beyond the state of the art. The solution integrates multiple layers of sensors, legacy systems and agent-based tools for beneficial services like learning, quality, risk and cost management. Additionally the ecological footprints will be reduced. The ARUM solution will run in two modes: predictive and real time simulation. The predictive mode supports the planning phase whereas the real-time operations mode supports dynamic, time-, cost- and risk-oriented re-planning of operations. The provision of information for engineering to alter in case of immaturity or late requests for changes is supported equally. ARUM is strongly end-user driven and the results will be tested on three industrial use cases with a focus on aircraft, aircraft interiors and ship manufacturing. The solution will be validated in a real industrial environment by industrial partners and benchmarked against todays ICT solutions. In collaboration with universities a test-bed will be established for design and testing of ARUM systems and tools and will be opened for dissemination and demonstration.