Automated Crane-lift Path Planning Using Modified Particle Swarm Optimization for High-rise Modular Integrated Construction

Abstract

Modular integrated construction (MiC) is a most advanced off-site technology. In a MiC project, it is critical but challenging to install prefabricated volumetric modules efficiently and safely, as they are much heavier and larger than conventional construction components and materials. However, the current crane-lift executions are heavily reliant on operators' subjective judgement, which turns out to be time-consuming and error-prone, most notably when jobsites are congested.

Automatic crane-lift path planning has been regarded as an important research topic for addressing this problem. Nevertheless, most previous studies did not consider the MiC-specific features of a crane lift such as correlation between module weight and crane trolley movement.

Therefore, this paper aims to propose a modified particle swarm optimization (PSO) algorithm to automatically conduct crane-lift path planning for high-rise MiC. A new fitness function and three auxiliary engines are designed for executing the proposed PSO path planner. This novel algorithm is validated using a real-life MiC project. The findings reveal that the optimized algorithm outperforms existing metaheuristics in terms of convergence characteristics and path smoothness. A collision-free crane-lift path can be worked out with a small population size and a few iterations.

Practically, this paper should facilitate safe and efficient project delivery of high-rise MiC. Scientifically, the paper contributes to the theoretical development of smart and automated technologies and algorithms in construction.