By Topic

The application of primitive heuristics to constraint satisfaction for complex real world optimisation

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

1 Author(s)
McMullan, P. ; Queen''s Univ. of Belfast, Belfast, UK

Summary form only given. The material presented in this work will outline the advances in the development of existing and new techniques produced from research, to complex scheduling problems encountered in industrial and commercial settings. The work outlines the success of using primitive or simple heuristics to achieve fast, efficient and good quality solutions to complex constraint satisfaction problems. The techniques used highlight the robustness and generality observed, with comparisons to approaches and results from existing research in the current literature, and application to a number of varied problem domains. It is intended to demonstrate the unique approach and subsequent advantages of application of this research. These include (a) efficiency, in terms of increasing the likelihood of obtaining high-quality solutions, (b) generality, in that current techniques can be applied quickly and successfully to various problem domains, (c) scope, in which the satisfaction of more complex constraints to problem areas which up until now have been limited to specific subsets of the problem, can be achieved. The latter can be demonstrated by case studies with commercial feedback from sources of application. The work described expands on well-tried benchmarks, exploiting cutting edge research-based techniques in order to gauge effectiveness on wider-scope problems within varied problem domains. A common goal for most problem domains is achieving efficiency (cost) savings to the provider while maintaining satisfaction for those recipients influenced or affected by any solution. Almost as important as the quality of any technique, is speed in achieving solutions; it is important to allow the quick production of high-quality solutions, allowing many scenarios with multiple varied objectives to be explored. This work will be presented in two sections, i.e. the research work undertaken and benchmark results obtained, followed by the application of this work to the real-w- orld commercial environment including the challenges posed and feedback from this. Initially an introduction to the main research area will be given, covering problem definitions and current work in the literature. Each problem domain will be described, with results demonstrating the success of the approach in solving complex problems under examination, while maintaining generality. This will be expanded to highlight the scope and depth of the Real-World issues of these areas as faced in Industry, with the challenges and goals involved, advances made and feedback obtained. The work will conclude with a discussion of future plans for application to other areas with wider goals and more complex requirements.

Published in:

Data Mining and Optimization, 2009. DMO '09. 2nd Conference on

Date of Conference:

27-28 Oct. 2009