Abstract
During the last 10 years, many-objective optimization problems, i.e. optimization problems with more than three objectives, are getting more and more important in the area of multi-objective optimization. Many real-world optimization problems consist of more than three mutually dependent subproblems, that have to be considered in parallel. Furthermore, the objectives have different levels of importance. For this, priorities have to be assigned to the objectives. In this paper we present a new model for many-objective optimization called Prio-ε-Preferred, where the objectives can have different levels of priorities or user preferences. This relation is used for ranking a set of solutions such that an ordering of the solutions is determined. Prio-ε-Preferred is controlled by a parameter ε, that is problem specific and has to be adjusted experimentally by the developer. Therefore we also present an extension called Adapted-ε-Preferred (AEP), that determines the ε values automatically without any user interaction. To demonstrate the efficiency of our approach, experiments are performed. The method based on Prio-ε-Preferred is used to guide the search of an Evolutionary Algorithm. As optimization problem a very complex scheduling problem, i.e. a utilization planning in a hospital is used. The considered benchmarks consist of 2 up to 90 optimization objectives. First, Prio-ε-Preferred where ε is set “by hand”, is compared to the basic method NSGA-II. It is shown that Prio-ε-Preferred clearly outperforms NSGA-II. Furthermore, it turns out that the results obtained by AEP are as good as if ε is adjusted manually.
Similar content being viewed by others
Notes
But indeed, the modeling of user preferences as presented here in combination with the hypervolume indicator is an interesting task for future work.
If \(p \ge \#days\), the algorithm stops. But if the border of the next planning period is known, the operator can easily be extended such that it can mutate the first days of the next planning period.
In all Tables Millar-2Shift is an abbreviation for benchmark Millar-2Shift-DATA1.
Abbreviations
- AEP:
-
Adapted-ε-preferred
- MOO:
-
Multi-objective optimization
- EA:
-
Evolutionary algorithm
- NRP:
-
Nurse rostering problem
- SCC:
-
Strongly-connected components
- DFS:
-
Depth-first-search
- SC:
-
Satisfiability class
- VM:
-
k-Vertical mutation
- HM:
-
k-Horizontal mutation
- SSM:
-
k-Set shift mutation
- SPM:
-
Set pattern mutation
- AVG:
-
Average value
- OF:
-
Objective function
References
Auger A, Bader J, Brockhoff D, Zitzler E (2009) Articulating user preferences in many-objective problems by sampling the weighted hypervolume. In: Genetic and evolutionary computation conference, pp 555–562
Bader J, Zitzler E (2011) HypE: an algorithm for fast hypervolume-based many-objective optimization. Evol Comput 19(1):45–76
Benchmarks (2012) Employee scheduling benchmark data set. Technical report, ASAP, School of Computer Science, The University of Nottingham, UK. http://www.cs.nott.ac.uk/tec/NRP/
Brockhoff D, Zitzler E (2009) Objective reduction in evolutionary multiobjective optimization: theorie and applications. Evol Comput 17(2):135–166
Burke E, Curtois T, Qu R, Vanden-Berghe G (2012) Problem model for nurse rostering benchmark instances. Technical report, ASAP, School of Computer Science, University of Nottingham, UK. http://www.cs.nott.ac.uk/tec/NRP/papers/ANROM.pdf
Burke EK, De Causmaecker P, Berghe GV, Landeghem HV (2004) The state of the art of nurse rostering. J Schedu 7:441–499
Cormen TH, Leierson CE, Rivest RC (1990) Introduction to algorithms. MIT Press, Cambridge
Corne D, Knowles J (2007) Techniques for highly multiobjective optimization: theorie and applications. In: Genetic and evolutionary computation conference, pp 773–780
Deb K (2001) Multi-objective optimization using evolutionary algorithms. Wiley, New York
Deb K, Saxena K (2006) Searching for pareto-optimal solutions through dimensionality reduction for certain large-dimensional multi-objective optimization problems. In: IEEE congress on evolutionary computation, pp 3353–3360
Deb K, Pratap A, Agarwal S, Meyarivan T (2000) A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE Trans Evol Comput 6:182–197
Deb K, Thiele L, Laumanns M, Zitzler E (2005) Scalable test problems for evolutionary multi-objective optimization. In: In evolutionary multiobjective optimization, theoretical advances and applications, pp 105–145
di Pierro F, Khu ST, Savic DA (2007) An investigation on preference order ranking scheme for multi-objective optimization. IEEE Trans Evol Comput 11(1):17–45
Drechsler N, Drechsler R, Becker B (2001) Multi-objective optimisation based on relation favour. In: International conference on evolutionary multi-criterion optimization, pp 154–166
Farina M, Amato P (2002) On the optimal solution definition for many-criteria optimization problems. In: Proceedings of the NAFIPS-FLINT international conference 2002. IEEE Press, Piscataway, NJ, pp 233–238
Fleming PJ, Purshouse RC, Lygoe RJ (2005) Many-objective optimization: an engineering design perspective. In: International conference on evolutionary multi-criterion optimization, pp 14–32
Fonseca CM, Fleming PJ (1995) An overview of evolutionary algorithms in multiobjective optimization. Evol Comput 3(1):1–16
Geiger MJ (2009) Multi-criteria curriculum-based course timetabling—a comparison of a weighted sum and a reference point based approach. In: International conference on evolutionary multi-criterion optimization, pp 290–304
Goldberg DE (1989) Genetic algorithms in search, optimization & machine learning. Addision-Wesley Publisher Company, Inc., Boston
Hughes EJ (2007) Radar waveform optimization as a many-objective application benchmark. In: International conference on evolutionary multi-criterion, optimization, pp 700–714
Ishibuchi H, Tsukamoto N, Nojima Y (2008) Evolutionary many-objective optimization: a short review. In: IEEE congress on evolutionary computation, pp 2424–2431
Khare VR, Yao X, Deb K (2003) Performance scaling of multi-objective evolutionary algorithms. In: EMO 2003. Lecture Notes in Computer Science, vol 2632, pp 376–390
Koza J (1992) Genetic programming—on the programming of computers by means of natural selection. MIT Press, Cambridge
Li X, Wong H (2009) Logic optimality for multi-objective optimization. Appl Math Comput 215:3045–3056
Onety R, Moreira G, Neto O, Takahashi R (2011) Variable neighborhood multiobjective genetic algorithm for the optimization of routes in IP networks. In: International conference on evolutionary multi-criterion optimization, pp 443–447
Pizzuti C (2012) A multiobjective genetic algorithm to find communities in complex networks. IEEE Trans Evol Comput 16(3):418–430
Purshouse RC, Fleming PJ (2003) Evolutionary multi-objective optimisation: an exploratory analysis. In: Proceedings of the (2003) IEEE congress on evolutionary computation (CEC’2003), Canberra, Australia, pp 2066–2073
Sato H, Aguirre HE, Tanaka K (2007) Controlling dominance area solutions and its impact on the performance of MOEAs. In: International conference on evolutionary multi-criterion optimization, pp 5–20
Schmiedle F, Drechsler N, Große D, Drechsler R (2001) Priorities in multi-objective optimization for genetic programming. In: Genetic and evolutionary computation conference, pp 129–136
Sülflow A, Drechsler N, Drechsler R (2007) Robust multi-objective optimization in high-dimensional spaces. In: International conference on evolutionary multi-criterion optimization, pp 715–726
Wagner T, Trautmann H (2010) Integration of preferences in hypervolume-based multiobjective evolutionary algorithms by means of desirability functions. IEEE Trans Evol Comput 14(5):688–701
Wagner T, Beume N, Naujoks B (2007) Pareto-, aggregation- and indicator-based methods in many-objective optimization. In: International conference on evolutionary multi-criterion optimization, pp 742–756
Wickramasinghe U, Li X (2009) A distance metric for evolutionary many-objective optimization algorithms using user-preferences. In: 22nd Australasian joint conference on advances in artificial intelligence (AI’09), pp 443–453
Zhang Q, Li H (2007) Moea/d: a multiobjective evolutionary algorithm based on decomposition. IEEE Trans Evol Comput 11(6):712–731
Zitzler E, Thiele L (1999) Multiobjective evolutionary algorithms: a comparative case study and the strength pareto approach. IEEE Trans Evol Comput 3(4):257–271
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Drechsler, N., Sülflow, A. & Drechsler, R. Incorporating user preferences in many-objective optimization using relation ε-preferred. Nat Comput 14, 469–483 (2015). https://doi.org/10.1007/s11047-014-9422-0
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11047-014-9422-0