Scholarlink

<p>Performance assessment of the sugarcane industry has become a significant task in terms of sustainability due to its substantial economic, environmental, and social implications. Evaluating the industry’s efficiency helps in determining areas for improvement and ensuring the best possible use of resources. This study uses a two-stage Network Data Envelopment Analysis (NDEA) model to examine the efficiency of the sugarcane industry in selected Indian states between 2001 and 2024. The study focuses on critical factors influencing the industry such as areas under cultivation, production, number of factories, cane crushed, and sugar production. The NDEA model assesses the efficiency of decision-making units (DMUs) in both cultivation and production stages. Further, time-series analysis identifies historical patterns, while models—ARIMA, LSTM, ANN, and SVR—predict outputs for the most efficient DMUs. The results indicate that ANN performs best, with an <InlineEquation ID="IEq1"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="13198_2025_2946_Article_IEq1.gif" Format="GIF" Height="16" Rendition="HTML" Resolution="72" Type="Linedraw" Width="21" /> </InlineMediaObject> <EquationSource Format="TEX">\({R}^{2}\)</EquationSource> </InlineEquation> value of 0.99, RMSE of 24.56, and MAE of 15.32. The efficiency scores and forecasting results offer valuable insights for decision-makers.</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

Leveraging the potential of machine learning and NDEA for efficiency assessment and prediction in sugarcane industry

  • 期刊论文

  • 发表日期：

    2025年9月11日

查看全文
  • Atul Kumar,
  • Millie Pant

摘要

Performance assessment of the sugarcane industry has become a significant task in terms of sustainability due to its substantial economic, environmental, and social implications. Evaluating the industry’s efficiency helps in determining areas for improvement and ensuring the best possible use of resources. This study uses a two-stage Network Data Envelopment Analysis (NDEA) model to examine the efficiency of the sugarcane industry in selected Indian states between 2001 and 2024. The study focuses on critical factors influencing the industry such as areas under cultivation, production, number of factories, cane crushed, and sugar production. The NDEA model assesses the efficiency of decision-making units (DMUs) in both cultivation and production stages. Further, time-series analysis identifies historical patterns, while models—ARIMA, LSTM, ANN, and SVR—predict outputs for the most efficient DMUs. The results indicate that ANN performs best, with an \({R}^{2}\) value of 0.99, RMSE of 24.56, and MAE of 15.32. The efficiency scores and forecasting results offer valuable insights for decision-makers.