Multiobjective Evolutionary Model of the Construction Industry Based on Network Planning
Dehu Yu,Qing Lv,Gautam Srivastava,Chun-Hao Chen,Jerry Chun-Wei Lin
DOI: https://doi.org/10.1109/tii.2022.3190566
IF: 12.3
2022-12-17
IEEE Transactions on Industrial Informatics
Abstract:The study of multiobjective optimization problems, such as resources, time limits, and cost based on network planning, can lead to an intelligent construction plan, which is very important for improving the comprehensive benefit of the project. In this article, a multiobjective static network planning optimization (i.e., sNP-RTC) model is proposed to integrate the heuristic local search algorithm and adaptive operation to improve the multiobjective particle swarm optimization (MOPSO) algorithm to solve the comprehensive "resource–duration–cost" optimization problem. The results show that this algorithm can achieve a unified and comprehensive Pareto frontier compared with the genetic MOPSO, providing a new method for solving resource–duration–cost optimization problems. Moreover, a dynamic "resource–time–cost" problem model for dynamic network planning (i.e., dNP-RTC) is presented and applied to improve the multiobjective subgroup algorithm to solve the model in real time after updating the network planning parameters. Compared with the static problem, the model increases the flexibility of problem solving, is more in-line with engineering practice, saves more cost in work scheduling, and has more application value for actual project scheduling.
automation & control systems,computer science, interdisciplinary applications,engineering, industrial