Efficient Genetic Algorithm sets for optimizing constrained building design problem

The main aim of this paper is to find the appropriate set of Genetic Algorithm (GA), control parameters that attain the optimum, or near optimum solutions, in a reasonable computational time for constrained building optimization problem. Eight different combinations of control parameters of binary coded GA were tested in a hypothetical building problem by changing 80 variables. The results showed that GA performance was insensitive to some GA control parameter values such as crossover probability and mutation rate. However, population size was the most influential control parameter on the GA performance. In particular, the population sizes (15 individuals) require less computational time to reach the optimum solution. In particular, a binary encoded GA with relatively small population sizes can be used to solve constrained building optimization problems within 750 building simulation calls.