Abstract
Diagnosis aims to identify faults that explain symptoms by a sequence of actions. Computing minimum cost of diagnosis is a NP-Hard problem if diagnosis actions are dependent. Some algorithms about dependent actions are proposed but their problem descriptions are not accurate enough and the computation cost is high. A precise problem model and an algorithm named NGAOAS (Niche Genetic Algorithm for the Optimal Action Sequence) are proposed in this paper. NGAOAS can avoid running out of memory by forecasting computing space, and obtain better implicit parallelism than normal genetic algorithm, which makes the algorithm be easily computed under grid environments. When NGAOAS is executed, population is able to hold diversity and avoid premature problem. At the same time, NGAOAS is aware of adaptive degree of each niche that can reduce the computation load. Compared with updated P/C algorithm, NGAOAS can achieve lower diagnosis cost and obtain better action sequence.
This work is supported by the National Grand Fundamental Research 973 Program of China under Grant No. 2003CB314802.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Huard, J.-F., Lazar, A.A.: Fault isolation based on decision-theoretic troubleshooting. Technical Report 442-96-08, Center for Telecommunications Research, Columbia University, New York, NY (1996)
Baras, J., Li, H., Mykoniatis, G.: Integrated distributed fault management for communication networks. Technical Report 98-10, Center for Satellite and Hybrid Communication Networks, University of Maryland, College Park, MD (1998)
Breese, J.S., Heckerman, D.: Decision-Theoretic Troubleshooting: A Framework for Repair and Experiment. In: Proc. 12th Conf. on Uncertainty in Artificial Intelligence, Morgan Kaufmann Publishers, San Francisco, CA, pp. 124–132. Morgan Kaufmann, San Francisco (1996)
Skaanning, C., Jensen, F.V., Kjærulff, U.: Printer Troubleshooting Using Bayesian Networks. In: Industrial and Engineering Applications of Artificial Intelligence and Expert Systems (IEA/AIE) 2000, New Orleans, USA (June 2000)
VomlelovÁ, M.: Decision Theoretic Troubleshooting. PhD Thesis, Faculty of Informatics and Statistics, University of Economics, Prague, Czech Republic (2001)
Zhi, S.Z.: Knowledge Discover. Tsinghua university press, Beijing (2002) (in Chinese) ISBN 7-302-05061-9:2652̃93
Qiang, L.M., Song, K.J., Dan, L., Quan, L.S.: Basic theory and application of Genetic Algorithm. Science press, Beijing (2002); ISDN7-03-009960-5:233-244
Skaanning, C., Jensen, F.V., Kjarulff, U., Madsen, A.L.: Acquisition and transformation of likelihoods to conditional probabilities for Bayesian networks. In: AAAI Spring Symposium, Stanford, USA (1999)
Skaanning, C.: A knowledge acquisition tool for Bayesian network troubleshooters. In: Proceedings of the Sixteenth Conference on Uncertainty in Artificial Intelligence, Stanford, USA (2000)
Steinder, M., Sethi, A.S.: Increasing robustness of fault localization through analysis of lost, spurious, and positive symptoms. IEEE (2002)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2005 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Lin, C., Jie, H., Zheng-Hu, G. (2005). Computing the Optimal Action Sequence by Niche Genetic Algorithm. In: Furbach, U. (eds) KI 2005: Advances in Artificial Intelligence. KI 2005. Lecture Notes in Computer Science(), vol 3698. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11551263_13
Download citation
DOI: https://doi.org/10.1007/11551263_13
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-28761-2
Online ISBN: 978-3-540-31818-7
eBook Packages: Computer ScienceComputer Science (R0)