Scholarlink

<p>To address the challenge of computing-intensive large-scale data envelopment analysis (DEA), our study introduces the enhanced build hull (EBH) algorithm for swift assessment of China’s coal-fired electricity supply chain (CFESC) enterprises. Compared to the build hull (BH) algorithm, a benchmark algorithm in the literature, the proposed algorithm is more efficient in evaluating the environmental efficiency under large-scale data context because it solves fewer linear programs and is compatible with parallel processing. Numerical implementation demonstrated that the EBH algorithm outperforms the BH algorithm, achieving a nearly fivefold greater computational speed on average across various scenarios. Applied to China’s CFESC enterprises, our computational analysis on a firm-level dataset demonstrates the remarkable improvement of the EBH algorithm by completing tasks two orders of magnitude faster than the traditional DEA method while maintaining identical results. The evaluation of environmental efficiencies of China’s CFESC enterprises indicates that only 6.75% of the enterprises were found to be environmentally efficient; furthermore, a mere 10.35% of the enterprises achieved an environmental efficiency greater than 0.50, with the majority falling between 0.05 and 0.50. The kernel density estimation highlights a bimodal pattern, illustrating a polarized environmental efficiency phenomenon in CFESC enterprises.</p>

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

Environmental efficiency evaluation of China’s coal-fired electricity supply chain enterprises with large-scale datasets: an enhanced build hull algorithm

  • 期刊论文

  • 发表日期：

    2025年3月22日

查看全文
  • Jie Wu,
  • Chao-Chao Zhang,
  • Jun-Fei Chu,
  • Guang-Cheng Xu,
  • Ying-Hao Pan

摘要

To address the challenge of computing-intensive large-scale data envelopment analysis (DEA), our study introduces the enhanced build hull (EBH) algorithm for swift assessment of China’s coal-fired electricity supply chain (CFESC) enterprises. Compared to the build hull (BH) algorithm, a benchmark algorithm in the literature, the proposed algorithm is more efficient in evaluating the environmental efficiency under large-scale data context because it solves fewer linear programs and is compatible with parallel processing. Numerical implementation demonstrated that the EBH algorithm outperforms the BH algorithm, achieving a nearly fivefold greater computational speed on average across various scenarios. Applied to China’s CFESC enterprises, our computational analysis on a firm-level dataset demonstrates the remarkable improvement of the EBH algorithm by completing tasks two orders of magnitude faster than the traditional DEA method while maintaining identical results. The evaluation of environmental efficiencies of China’s CFESC enterprises indicates that only 6.75% of the enterprises were found to be environmentally efficient; furthermore, a mere 10.35% of the enterprises achieved an environmental efficiency greater than 0.50, with the majority falling between 0.05 and 0.50. The kernel density estimation highlights a bimodal pattern, illustrating a polarized environmental efficiency phenomenon in CFESC enterprises.