Scholarlink

In this study, machine learning models are used to group together job characteristics with similar requirements in order to facilitate resource management and planning. Indeed, segmentation into clusters means that scheduling strategies can be tailored to the specific characteristics of each group, enabling more efficient management of instances according to their different profiles. The obtained results of different scheduling algorithms like EDD, LPT, \(Slack\_min\) , \(Cout\_min\) , WMDD, and HGAKANG are considered. Principal Component Analysis as a dimensionality reduction method is used with three classification models: Random Forest, Gradient Boosting, and Support Vector Machines and K-means as a clustering method. Several metrics are used to compare model performance, such as Accuracy, Precision, Recall, and F1-score on 1800 generated instances. The obtained results show the three clusters identified are uniform groups of the scheduling instance, each with regard to job characteristics.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Machine Learning for Characterizing the Tardiness Minimization Problem on a Single Machine

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Youcef Abdelsadek,
  • Bochra Djahel,
  • Allaoua Hemmak,
  • Imed Kacem,
  • Giorgio Lucarelli

摘要

In this study, machine learning models are used to group together job characteristics with similar requirements in order to facilitate resource management and planning. Indeed, segmentation into clusters means that scheduling strategies can be tailored to the specific characteristics of each group, enabling more efficient management of instances according to their different profiles. The obtained results of different scheduling algorithms like EDD, LPT, \(Slack\_min\) , \(Cout\_min\) , WMDD, and HGAKANG are considered. Principal Component Analysis as a dimensionality reduction method is used with three classification models: Random Forest, Gradient Boosting, and Support Vector Machines and K-means as a clustering method. Several metrics are used to compare model performance, such as Accuracy, Precision, Recall, and F1-score on 1800 generated instances. The obtained results show the three clusters identified are uniform groups of the scheduling instance, each with regard to job characteristics.