Multi-objective Construction Site Layout Planning Using Genetic Algorithms

Efficient layout planning of a construction site is fundamental for successful project undertaking as it enhances both productivity and safety in construction sites. This task usually consists of identifying the temporary facilities needed to support construction operations, determining their size and shape, and optimally positioning them in the unoccupied areas within the site boundaries. The site layout planning problem is a complex combinatorial optimization problem involving multiple objectives and it grows significantly in size as the number of facilities and constraints increases. The existing literature includes a variety of analytical, heuristic, and meta-heuristic techniques for solving the problem but existing studies usually examine a small number of facilities and focus on travel distance minimization, ignoring generally cost related or other decision parameters. The objective of this study is to develop feasible and efficient site layout solutions in a realistic representation scheme taking into consideration not only the total distance traveled but also cost and safety parameters as well. A multi-objective optimization model is developed aiming at minimizing a generalized cost function which results from the construction cost of a facility placed at alternative locations, the transportation cost among locations, and any safety concern in the form of preferred proximity or remoteness of particular facilities to other facilities or work areas. The development integrates the required robust search objective with the optimization capabilities of the genetic algorithms (GAs). The model has been tested on several test cases and the results of a comparative study with existing methods from the literature are presented. The evaluation indicates that the proposed model provides effective and rational solutions, in response to decision parameters and problem constraints, and that it results in more robust layout planning than previous methods both qualitatively and quantitatively.