A data-driven support system for the efficient schedule delay management of the ultra-high-voltage projects considering subjective risk preferences. (April 2023)

• Record Type: Journal Article
• Authors: Wu, Yunna; He, Jiaming; Liao, Yijia; Tao, Yao; Liu, Fangtong; Zhou, Jianli
• Is Part Of: Sustainable cities and society. Volume 91(2023)
• Publication Date: 2023-04
• Subjects: Urban energy transition -- Sustainable cities -- Schedule delay -- Decision support model -- PSO algorithm -- LSSVM model; Sustainable urban development -- Periodicals; Sustainable buildings -- Periodicals; Urban ecology (Sociology) -- Periodicals; 307.76
• DOI: 10.1016/j.scs.2023.104448
• Languages: English
• ISSNs: 2210-6707

Abstract

Highlights: Support systems for schedule management of ultra-high-voltage projects is proposed. The system estimates UHV project situations considering subjective risk preferences. Dataset of project implementation is analyzed by enhanced PSO-LSSVR algorithm. Improved K-means algorithm is used to classify severity degrees of potential risks. Delay prevention measures are suggested according to severity of the deviation. Abstract: Renewable generation technologies are thriving due to sustainable transformation of the whole society, while there exists a general gap between large-scale renewable energy sources and loads in major cities which could be addressed by Ultra-high voltage (UHV) transmission projects. This paper proposes a data-driven support system for schedule delays of UHV projects. The system is composed of three modules. Firstly, general project schedule routines are interpreted by logistic curves. Classical earned values are measured by trapezoidal fuzzy numbers which are used for subjective judgement of project managers. These measures serve as the system input. Then, combined objective schedule delay and subjective severity judgement are analyzed by least square support vector regression. Enhanced particle swarm optimization is applied to determine optimal parameter combinations according to the structure of trained data. Next, delay prediction results combining subjective and objective information are divided into three severity degrees by an enhanced K-means algorithm and finally, integrated prevention measures are raised for project managers. The proposed system is validated by a real case study and results indicate good performance and reliability. The system could be integrated into the daily workflow of project managers and constantly provides references that facilitate efficient schedule management.