UTILIZING ARTIFICIAL INTELLIGENCE TO SOLVING PLANNING CONSTRUCTION SITE LAYOUT PROBLEMS

Document Type : Regular Article

Authors

1 Department of Construction Engineering and Management, Ho Chi Minh City University of Technology (HCMUT), Vietnam National University (VNU-HMC), Ho Chi Minh City, Vietnam

2 Department of Construction Engineering and Management, Ho Chi Minh City University of Technology (HCMUT), Vietnam National University (VNU-HMC), Ho Chi Minh City, Vietnam.

10.22115/scce.2024.395964.1638

Abstract

Construction site layout planning (CSLP) is an important problem in construction management. The facilities have conflicting purposes in the available space on the construction site, increasing the transportation flow of facilities, which is the cause of the increase in project construction costs. The research was adopted to solve Construction site layout planning (CSLP) depending on the model of quadratic assignment (QAP). The QAP is meant to determine the best way of allocating facilities to locations so that each facility is mapped to only one location. In contrast, every location is mapped to one exact facility in order to minimize the cost of product flows and distances. In addition, the number of facilities is equal. In the past, there were many ways to solve the problem using meta-heuristic methods such as (GA), (ACO). However, each method has its own disadvantages, which are long computation times and poor optimization results. Therefore, this study proposes a new hybrid algorithm (M-ALOB), to solve the problem (QAP), The goal of this problem is to minimize the flow of work between facilities on site, which depends mainly on two factors: the index of the density of work flows between facilities and the distance. distance between their positions. The results found from (M-ALOB) are compared with the previously studied algorithm (GA) and the original algorithm (ALO). The results showed that (M-ALOB) found the optimal value smaller than the algorithms (GA), (ALO).

Keywords

Main Subjects

Files

History

  • Receive Date: 05 May 2023
  • Revise Date: 05 August 2023
  • Accept Date: 17 February 2024

Share

How to cite

Statistics