An effectiveness assessment ensemble for transmission corridor mechanized construction schemes based on weighted itemset mining and factor criticality analysis

• Published in: Scientific reports Vol. 15; no. 1; pp. 12724 - 18
• Main Authors: Yang, Xin, Hu, Junyao, Sun, Chenhao, Huang, Yafei, Tang, Ci, Huang, Jie
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 13.04.2025; Nature Publishing Group; Nature Portfolio
• Subjects: 639/166/987; 639/166/988; Accuracy; Algorithms; Construction; Correlation coefficient; Efficiency; Entropy; Entropy impact method; Evaluation of mechanized construction schemes; Humanities and Social Sciences; Item mining algorithm; multidisciplinary; Optimization; Pearson coefficient; Qualitative analysis; Research methodology; Risk assessment; Risk factors; Science; Science (multidisciplinary); Weight optimization; Weight optimization; Entropy impact method; Pearson coefficient; Item mining algorithm; Evaluation of mechanized construction schemes
• ISSN: 2045-2322; 2045-2322
• DOI: 10.1038/s41598-025-91253-z

A reasonable assessment of overall effective levels, incorporating dimensions such as risks, durations, and costs, will be salutary for the rational planning and better selection of optional transmission corridor mechanized construction schemes. With this motivation, this paper establishes an ensemble to address the issue of effective level diagnoses, and thus the hidden patterns and regularities between scheme features and effective levels can be explored. Based on the complex characteristics of input data, the Pearson correlation coefficient is deployed to handle the multidimensional data from multiple sources, and K-means clustering is then employed to classify scheme indicators into classes. The Weighted Itemset Mining (W-IM) model is proposed to identify the underlying key factors, to cope with the frequent omission of High Impact Low Probability (HILP) factors during the qualitative analysis stage. Next, the Factor Criticality Analysis (FCA) model is built to quantify the specific impact levels of these distinguished key elements. Finally, the significance proportion of individual features to the overall effectiveness can be determined and optimized according to the Entropy Impact Model (EIM) model. An empirical case study indicates that the proposed ensemble in this paper exhibits higher predictive accuracy along with better flexibility and comprehensiveness.