China
Africa
Explore chapters and articles related to this topic
Adaptation in Genetic Algorithms
Published in Sankar K. Pal, Paul P. Wang, Genetic Algorithms for Pattern Recognition, 2017
Lalit M. Patnaik, Srinivas Mandavilli
Typically a GA is characterized by the following components: A genetic representation (or an encoding) for the feasible solutions to the optimization problemA population of encoded solutionsA fitness function that evaluates the optimality of each solutionGenetic operators that generate a new population from the existing populationControl parameters
Constrained Multivariable Optimization
Published in Nita H. Shah, Poonam Prakash Mishra, Non-Linear Programming, 2020
Nita H. Shah, Poonam Prakash Mishra
Initially, a population is created randomly with a group of individuals. These individuals are being evaluated on the basis of fitness function (objective function). A fitness function is defined by programmer over the genetic representation and measures the quality of the represented solution. The programmer provides score to individuals on the basis of their performance. Two individuals are selected on the basis of their fitness. Higher fitness score increases the chances of selection. Further, these best individuals reproduce offspring that are further muted on a random basis. This process continues until a feasible solution is attained. Let us discuss all the stages of GA step-by-step.
Optimization in Product Design – Synthesis
Published in S. Ratnajeevan H. Hoole, Yovahn Yesuraiyan R. Hoole, Finite Elements-based Optimization, 2019
S. Ratnajeevan H. Hoole, Yovahn Yesuraiyan R. Hoole
The representation of solutions, genetic representation (Rothlauf, 2006), is highly problem specific. It is essentially the process of transforming our data into a form that maximizes its expressiveness, that is, how well it describes its features and inheritability or how well its characteristics can be passed on to successive generations. In genetic representation a solution is often referred to as an individual or a chromosome and individual features are referred to as genes (see Figure 3.9).
Genetic algorithms for planning and scheduling engineer-to-order production: a systematic review
Published in International Journal of Production Research, 2023
Anas Neumann, Adnene Hajji, Monia Rekik, Robert Pellerin
As one can see in Figure 8, GAs also rely on three main mechanisms: (i) a genetic representation: an encoding format and decoding procedure with fitness evaluation, (ii) the generation of one or several initial populations, and (iii) one or several stop criteria. In the reviewed papers, the only two criteria used were a fixed number of generations or a number of iterations without improvement (convergence). Finally, several architectural features can help the GA to better adapt to the characteristics of the studied problem or reach better solutions (by increasing either its exploration or exploitation capacities). Some approaches also intend to reduce the computation time and memory needed by the algorithm. Among those approaches are parallel multi-population, hybrid, (self-)adaptive, and layered methods.
Applying hybrid genetic algorithm to multi-mode resource constrained multi-project scheduling problems
Published in Journal of the Chinese Institute of Engineers, 2022
James C. Chen, Hung-Yu Lee, Wen-Haiung Hsieh, Tzu-Li Chen
The fitness evaluation of every chromosome in the population aims to find better chromosomes and retain them into the next generation through the evolutionary process. Thus, many chromosomes in the next population are randomly selected from the current population according to the fitness values of all chromosomes. During this process, parents are selected to generate new offsprings through the reproduction and recombination operations, and a new population including the chromosomes with better fitness values is formed from current parent and offsprings. Fitness function is defined over genetic representation; it measures the quality measurement of the represented solution. Overall, the fitness is problem-dependent and described in the following function:
An adaptive hybrid genetic algorithm for pavement management
Published in International Journal of Pavement Engineering, 2019
João Santos, Adelino Ferreira, Gerardo Flintsch
After identifying the parameters that characterize the problem at hand and defining the objective function and constraints (problem formulation), the problem solutions are encoded into genetic representation. The use of an appropriate encoding representation is of great importance for the efficiency of a GA when applied to real-world problems. In the developed AHGA an integer coding is adopted to represent the M&R alternatives. Each individual represents a potential solution (M&R strategy) and consists of a sequence of S × T genes, where S is the number of pavement sections considered for analysis, T represents the PAP defined by the decision-maker, and the allele values for each of these genes represent a possible M&R activity.