Using genetic algorithms to solve optimization problems in construction

Abstract

Genetic algorithm (GA) is a model of machine learning. The algorithm can be used to find sub‐optimum, if not optimum, solution(s) to a particular problem. It explores the solution space in an intelligent manner to evolve better solutions. The algorithm does not need any specific programming efforts but requires encoding the solution as strings of parameters. The field of application of genetic algorithms has increased dramatically in the last few years. A large variety of possible GA application tools now exist for non‐computer specialists. Complicated problems in a specific optimization domain can be tackled effectively with a very modest knowledge of the theory behind genetic algorithms. This paper reviews the technique briefly and applies it to solve some of the optimization problems addressed in construction management literature. The lessons learned from the application of GA to these problems are discussed. The result of this review is an indication of how the GA can contribute in solving construction‐related optimization problems. A summary of general guidelines to develop solutions using this optimization technique concludes the paper.

Prediction and multi-objective optimization of high-strength concrete parameters via soft computing approaches

“…GAs has been applied successfully to solve discrete, nondifferentiable, combinatory, and general non-linear engineering optimization problems (Al-Tabtabai & Alex, 1999;Chang & Hao, 2001;Goldberg, 1989). A GA is a search strategy based on the rules of natural genetic evolution.…”

“…A fitness score is assigned to each individual, based on the quality of the solution it represents. The highly fit individuals are reproduced by breeding with other individuals using three major processes known as selection, crossover, and mutation (Al-Tabtabai & Alex, 1999;Goldberg, 1989). The main steps of the GA are as follows:…”

A hybrid genetic and imperialist competitive algorithm for green vendor managed inventory of multi-item multi-constraint EOQ model under shortage

“…Many researchers have successfully used meta-heuristic methods to solve complicated optimization problems in different fields of scientific and engineering disciplines. Some of these meta-heuristic algorithms are: simulating annealing [39,40], threshold accepting [41], Tabu search [42], genetic algorithm [35,43,44], particle swarm optimization [45][46][47][48][49][50], neural networks [51], ant colony optimization [28,52], evolutionary algorithm [53,54], harmony search [55,56] and gravitational search algorithm [57]. Among these algorithms, the population-based ones are usually preferred to others and in some cases show better performances.…”

A discrete shuffled frog optimization algorithm

“…More details on the mechanism of the BGA could be found in (Al-Tabtabai and Alex 1999;Goldberg 1989). Four main parameters which affect the performance of the BGA are population size, number of generations or function evaluations, crossover type and mutation rate.…”