Une bonne référence

Abstract

An optimal layout problem for temporary construction facilities (TFLP) is very important in large-scale construction projects. This problem involves the planning of temporary construction facilities within the boundaries of the restricted sites so that materials transportation and rearrangement costs are minimized and distances between the various departments are optimized. As construction continues, however, the temporary facilities may need to be dynamically relocated several times to accommodate the various operational demands. Thus, this study proposes a new method for dynamic TFLP with unequal-area departments. To begin with, a multi-objective dynamic optimization layout model based on facilities coordinates is presented, in which the transportation costs between the facilities are described as fuzzy random variables and the temporary facilities, represented as rectangles or squares, are restricted to two-dimensional geometric constraints. Subsequently, the multi-objective position-based adaptive particle swarm optimization (p-based MOPSO) is developed to obtain feasible optimization solutions for the proposed problem. In order to evaluate the performance of the proposed method, the Jinping-I Hydropower construction project is used as a practical example. The results and further analyses of the model, algorithm evaluation and effectiveness proved that satisfactory solutions were able to be obtained.