Impact of crossover and decision variable coding forms on the performance of genetic algorithm optimization a case study on pump and treat remediation design

Date

2008-05-23

Author

GÜNGÖR DEMİRCİ, Gamze
Aksoy, Ayşegül

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

139
views

0
downloads

Genetic algorithms (GAs) are robust methods applied especially for complex optimization problems. However, selection of some GA parameters may play a vital role on the efficacy of the method. This study investigates the impacts of crossover type and form of coding on the performance of optimization by genetic algorithms. For this purpose, two pump-and-treat remediation design problems were considered with different levels of complexity determined by the number of decision variables. The efficiencies of GA for these problems were compared for two different crossover operators (two-point and uniform) and decision variable coding forms (binary and gray). First problem seeks the best pump and treat system design by optimizing the location of the pumping wells and pumping rate at active wells while minimizing the system cost. The second problem is simpler such that it aims to minimize the remediation time while achieving the cleanup goals for a fixed remediation policy. Results show that uniform crossover operator outperforms two-point crossover and gray coding is superior to binary coding for the complex problem with higher number of decision variables. On the other hand, when a simpler problem was solved, the efficiency of GA was independent of the crossover and coding types.

Subject Keywords

Optimization, Genetic algorithms, Uniform crossover, Two-point crossover, Binary coding, Gray coding

URI

https://hdl.handle.net/11511/55743

Collections

Department of Environmental Engineering, Conference / Seminar

Suggestions

OpenMETU
Core

Evaluation of crossover techniques in genetic algorithm based optimum structural design Hasançebi, Oğuzhan (2000-11-01) Crossover is one of the three basic operators in any genetic algorithm (GA). Several crossover techniques have been proposed and their relative merits are currently under investigation. This paper starts with a brief discussion of the working scheme of the GAs and the crossover techniques commonly used in previous GA applications. Next, these techniques are tested on two truss size optimization problems, and are evaluated with respect to exploration and exploitation aspects of the search process. Finally, t...
A method for chromosome handling of r-permutations of n-element set in genetic algorithms Üçoluk, Göktürk (1997-04-16) Combinatorial optimisation problems are in the domain of Genetic Algorithms (GA) interest. Unfortunately ordinary crossover and mutation operators cause problems for chromosome representations of permutations and some types of combinations. This is so because offsprings generated by means of the ordinary operators are of a great possibility no more valid chromosomes. A variety of methods and new operators that handle that sort of obscenities are introduced throughout the literature. A new method for represe...
Impact of physical and chemical heterogeneities on optimal remediation design and costs Aksoy, Ayşegül (null; 2001-01-24) The impact of physical and chemical aquifer heterogeneities on optimal remediation design and costs is investigated by linking a genetic algorithm optimization library with a contaminant transport simulation model. Various levels of physical and chemical (sorption) aquifer heterogeneities are examined. In the first level, heterogeneity is limited to the hydraulic conductivity (K) field. Then systems with heterogeneity in both K and the distribution coefficient (Kd) are considered. The final level of heterog...
Genetic algorithm for constrained optimization models and its application in groundwater resources management Guan, Jiabao; Kentel Erdoğan, Elçin; Aral, Mustafa M. (American Society of Civil Engineers (ASCE), 2008-01-01) Genetic algorithms (GAs) have been shown to be an efficient tool for the solution of unconstrained optimization problems. In their standard form, GA formulations are "blind" to the constraints of an optimization model when the model involves these constraints. Thus, in GA applications alternative procedures are used to satisfy the constraints of the optimization model. In this study, the method that is utilized in the Complex Algorithm to solve constrained optimization problems is abstracted to develop a re...
Impact of Number of Interactions, Different Interaction Patterns, and Human Inconsistencies on Some Hybrid Evolutionary Multiobjective Optimization Algorithms Marquis, Jon; Gel, Esma S.; Fowler, John W.; Koeksalan, Murat; Korhonen, Pekka; Wallenius, Jyrki (2015-10-01) We investigate the impact of the number of human-computer interactions, different interaction patterns, and human inconsistencies in decision maker responses on the convergence of an interactive, evolutionary multiobjective algorithm recently developed by the authors. In our context an interaction means choosing the best and worst solutions among a sample of six solutions. By interaction patterns we refer to whether preference questioning is more front-, center-, rear-, or edge-loaded. As test problems we u...

Citation Formats

G. GÜNGÖR DEMİRCİ and A. Aksoy, “Impact of crossover and decision variable coding forms on the performance of genetic algorithm optimization a case study on pump and treat remediation design,” 2008, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/55743.