Review of Research on the Application of Genetic Algorithms for Multi-objective Optimal Management of Building Construction
Description
With the construction of large public buildings, the application of optimization methods for building construction management has started to move from initial inefficiencies to artificial intelligence. This paper explores the application of genetic algorithms to solve multi-objective optimization in building construction management from the aspect of artificial intelligence. Firstly, a comprehensive description of the four types of parameter coding methods and genetic operations (crossover and variation) is given, and then the comprehensive performance of the four types of coding methods is evaluated in terms of their basic property characteristics (chromosome robustness, chromosome storage space, decoding complexity, coding completeness and genetic operation diversity). Then, the characteristics of various types of crossover and variation operators are elaborated and their applicability to the four types of parameter coding is analyzed. The duration-cost line graphs of the four types of parameter coding methods applied to a construction management example are plotted using Matlab software and the genetic operation efficiency of the four types of coding methods is evaluated comprehensively. The results show that the floating-point coding method can improve the efficiency of dealing with multi-objective optimal management problems in building construction.