Abstract
In design situations where a single solution must be selected, it is often desirable to present the designer with a smart Pareto set of solutions—a minimal set of nondominated solutions that sufficiently represents the tradeoff characteristics of the design space. These sets are generally created by finding many well-distributed solutions and then either filtering out the excess ones or searching more closely in those regions that appear to have significant tradeoff. Such methods suffer from the inherent inefficiency of creating numerous solutions that will never be presented to the designer. This paper introduces the Smart Normal Constraint (SNC) method—a Pareto set generation method capable of directly generating a smart Pareto set. Direct generation is achieved by iteratively updating an approximation of the design space geometry and searching only in those regions capable of yielding new smart Pareto solutions. This process is made possible through the use of a new, computationally benign calculation for identifying regions of high tradeoff in a design space. Examples are provided that show the SNC method performing significantly more efficiently than the predominant existing method for generating smart Pareto sets.
Similar content being viewed by others
References
Aittokoski T, Ayramo S, Miettinen K (2009) Clustering aided approach for decision making in computationally expensive multiobjective optimization. Optim Method Softw 24:157–174
Barber CB, Dobkin DP, Huhdanpaa HT (1996) The quickhull algorithm for convex hulls. ACM Trans Math Softw 22:469–483
Bechikh S, Said LB, Ghedira K (2010) Searching for knee regions in multi-objective optimization using mobile reference points. In: Proceedings of the 2010 ACM symposium on applied computing
Boyce NO, Mattson CA (2008) Reducing computational time of the normal constraint method by eliminating redundant optimization runs. In: 12th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference
Deb K, Tiwari S (2006) Reference point based multi-objective optimization using evolutionary algorithms. Int J Comput Intell Res 2:273–286
Haddock ND, Mattson CA, Knight DC (2008) Exploring direct generation of smart Pareto sets. In: 12th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference
Handl J, Knowles J (2007) An evolutionary approach to multiobjective clustering. IEEE Trans Evol Comput 11:56–76
Huang HZ, Gu YK, Du X (2006) An interactive fuzzy multi-objective optimization method for engineering design. Eng Appl Artif Intell 19:451–460
Ismail-Yahaya A, Messac A (2002) Effective generation of the Pareto frontier using the normal constraint method. In: 40th Aerospace Sciences Meeting and Exhibit
Marler RT, Arora JS (2004) Survey of multi-objective optimization methods for engineering. Struct Multidiscip Optim 26:369–395
Martinez M, Sanchis J, Blasco X (2007) Global and well-distributed Pareto frontier by modified normalized normal constraint methods for bicriterion problems. Struct Multidiscip Optim 34:197–209
Martinez M, Garcia-Nieto S, Sanchis J, Blasco X (2009) Genetic algorithms optimization for normalized normal constraint method under Pareto construction. Adv Eng Softw 40:260–267
Mattson CA, Mullur AA, Messac A (2004) Smart Pareto filter: Obtaining a minimal representation of multiobjective design space. Eng Optim 36:721–740
Messac A, Mattson CA (2004) Normal constraint method with guarantee of even representation of complete Pareto frontier. AIAA J 42:2101–2111
Messac A, Ismail-Yahaya A, Mattson CA (2003) The normalized normal constraint method for generating the Pareto frontier. Struct Multidiscip Optim 25:86–98
Motta RS, Afonso SMB, Lyra PRM (2012) A modified nbi and nc method for the solution of n-multiobjective optimization problems. Struct Multidiscip Optim 46:239–259
Pareto V (1964) Cour deconomie politique. Librarie Droz-Geneve (the first edition in 1896)
Rachmawati L, Srinivasan D (2009) Multiobjective evolutionary algorithm with controllable focus on the knees of the Pareto front. IEEE Trans Evol Comput 13:810–824
Ray T, Tai K, Seow C (2001) An evolutionaryalgorithm for multiobjective optimization. Eng Optim 33:399–424
Ruzika S, Wiecek MM (2005) Approximation methods in multiobjective programming. J Optim Theory Appl 126(3):473–501
Rynne B (2007) Linear functional analysis. Springer, New York
Sanchis J, Martinez M, Blasco X, Salcedo JV (2008) A new perspective on multiobjective optimization by enhanced normalized normal constraint method. Struct Multidiscip Optim 36:537–546
Schutze O, Laumanns M (2008) Approximating the knee of an MOP with stochastic search algorithms. Springer-Verlag, New York
Tanaka M, Watanabe H, Furukawa Y, Tanino T (1995) Ga-based decision support system for multicriteria optimization. In: Proceedings IEEE international conference systems
Zitzler E, Thiele L (1998) Multiobjective optimization using evolutionary algorithms–a comparitive case study. In: Parallel Problem Solving From Nature
Acknowledgments
We would like to recognize the National Science Foundation (Grant CMMI-0954580) for funding this research.
Author information
Authors and Affiliations
Corresponding author
Appendix: Flowchart comparison of NC* and SNC methods
Appendix: Flowchart comparison of NC* and SNC methods
Rights and permissions
About this article
Cite this article
Hancock, B.J., Mattson, C.A. The smart normal constraint method for directly generating a smart Pareto set. Struct Multidisc Optim 48, 763–775 (2013). https://doi.org/10.1007/s00158-013-0925-6
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00158-013-0925-6