Search

Many construction planning problems require optimizing multiple and conflicting project objectives such as minimizing construction time and cost while maximizing safety, quality, and sustainability. To enable the optimization of these construction problems, a number of research studies focused on developing multiobjective optimization algorithms (MOAs). The robustness of these algorithms needs further research to ensure an efficient and effective optimization of large-scale real-life construction problems. This paper presents a review of current research efforts in the field of construction multiobjective optimization and two case studies that illustrate methods for enhancing the robustness of MOAs. The first case study utilizes a multiobjective genetic algorithm (MOGA) and an analytical optimization algorithm to optimize the planning of postdisaster temporary housing projects. The second case study utilizes a MOGA and parallel computing to optimize the planning of construction resource utilization in large-scale infrastructure projects. The paper also presents practical recommendations based on the main findings of the analyzed case studies to enhance the robustness of multiobjective optimization in construction engineering and management. [ABSTRACT FROM AUTHOR]

This paper presents a resource sharing-based construction equipment allocation problem with multiple stages under a fuzzy environment. A multiobjective multistage decision-making model is established in which the equipment failure rate is regarded as a fuzzy variable. In contrast to traditional assignment problems, the principle of resource sharing modes including space and time sharing is considered to be based on the project practice. In addition, a dynamic programming-based genetic algorithm is developed to find feasible solutions and a dynamic programming-based initialization, and crossover and mutation are designed to avoid infeasible solutions. Finally, the Jinping-I Hydropower Project is used as a practical example to demonstrate the practicality and efficiency of the model. The comparisons and analysis of results are presented to highlight the performance of the optimization method, which is proved to be more effective and efficient than the standard genetic algorithm. [ABSTRACT FROM AUTHOR]

The heuristic method is one of the methods used for the scheduling of resource-constrained projects. This method is commonly used in programming the projects with high number of activities and resources such as construction investments. This paper investigates the effectiveness of three heuristic method priority rules applied in the resource scheduling of ten Turkish housing estate projects which were scheduled according to three preselected priority rules [maximum remaining path length (MRPL), latest finish time (LFT), and minimum slack time (MNSLCK)] in resource-constrained conditions. The performance of each priority rule was evaluated in relation to the duration of the project. The results revealed that MRPL priority reduced the project duration to minimum in six projects, whereas LFT priority yielded the best duration results in three projects and MNSLCK priority in only one project. [ABSTRACT FROM AUTHOR]

This paper presents the development of a parallel multiobjective genetic algorithm framework to enable an efficient and effective optimization of resource utilization in large-scale construction projects. The framework incorporates a multiobjective optimization module, a global parallel genetic algorithm module, a coarse-grained parallel genetic algorithm module, and a performance evaluation module. The framework is implemented on a cluster of 50 parallel processors and its performance was evaluated using 183 experiments that tested various combinations of construction project sizes, numbers of parallel processors and genetic algorithm setups. The results of these experiments illustrate the new and unique capabilities of the developed parallel genetic algorithm framework in: (1) Enabling an efficient and effective optimization of large-scale construction projects; (2) achieving significant computational time savings by distributing the genetic algorithm computations over a cluster of parallel processors; and (3) requiring a limited and feasible number of parallel processors/computers that can be readily available in construction engineering and management offices. [ABSTRACT FROM AUTHOR]

The radio frequency identification (RFID) technology has proven its potential in locating and tracking construction resources, a critical task in construction project control. However, the main challenge is how to achieve desired levels of locating accuracy. This paper presents an enhanced boundary condition method that incorporates the tag-reader angle and the reader geometric configuration factors to control the accuracy of a locating system that integrates RFID and real time kinematic (RTK) global positioning system (GPS). Controlled laboratory experiments were conducted to assess their effects and create quality control filters. This study demonstrated the relationship between the detecting range and the tag-reader angle and used it to separate valid/invalid boundary points. Spatial dilution of precision (SDOP) was formulated to measure the geometric configuration of readers forming the boundary constraint. Correlating SDOP to the locating error through a polynomial regression model, a mechanism was created to use SDOP to predict and control the locating accuracy. Open field experiments validated that appropriate SDOP, and tag-reader angle filters could be created to achieve varying levels of accuracy. [ABSTRACT FROM AUTHOR]