Genetic algorithm (GA) is a technique used for solving optimization challenges which can either be constrained or unconstrained. The technique solves optimization problems basing on natural selection process. Damci, Arditi & Polat (2013) note that GM mimics biological natural selection process. The technique has been employed in numerous for instance, during scheduling, resource leveling, constrained and unconstrained optimization and resource allocation. On the other hand, Prasad & Park (2004) assert that GA has been found to be an effective technique in solving challenges relating to optimization issues, evolutionary search algorithms, and classical search challenges.
The use of GA optimization has been used in the field of engineering since the 1980s. Usually, when searching for algorithms, a population of solutions is often employed in the search while Pareto optimal solutions can be easily found in a single search. However, it is important to use diversity preserving methods since their incorporation in evolutionary search algorithms aids in the discovery of widely varied Pareto optimal solutions (Damci, Arditi & Polat 2013).
When conducting GA, the possible solutions for a delinquent are presented as a population of chromosomes. The genes within a chromosome represent the values of a variable for a particular issue in question. For practicality to be attained, binary numbers can be employed to fill in the values of a variable depending on the nature of the issue. One of the important consideration during GA operation is the selection of parent chromosomes. These chromosomes are then examined depending on their fitness; computed via the objective function specified for a specific challenge. Similarly, their offspring are examined basing on their fitness since the chromosomes depicting high levels of fitness are more likely to survive than the others (Damci, Arditi & Polat 2013).Similarly, the process can be used in arriving at a solution when faced with a problem. Some of the commonly used multi-objective GA are multi-objective optimization GA (MOGA), non-dominated sorting genetic algorithm (NSGA and vector enabled genetic algorithm (VEGA) among others (Prasad & Park 2004).
References
Damci, A., Arditi, D., & Polat, G. (2013). Resource leveling in line‐of‐balance scheduling. Computer‐Aided Civil and Infrastructure Engineering, 28(9), 679-692.
Prasad, T. D., & Park, N. S. (2004). Multiobjective genetic algorithms for design of water distribution networks. Journal of Water Resources Planning and Management, 130(1), 73-82.
0
714