An analytic hierarchy process combined with artificial neural network model to evaluate sustainable sludge treatment scenarios.

Journal: Waste management (New York, N.Y.)
PMID:

Abstract

Sludge management in China faces critical environmental, economic, and technical challenges, necessitating urgent optimal management strategy selection. Given the limited number of comprehensive studies on sludge management, quantitative decision-making tools are urgently required. To address this gap, this study developed an integrated Analytic Hierarchy Process (AHP)-artificial neural network (ANN) model to evaluate sludge treatment scenarios. Four representative scenarios were evaluated based on carbon emissions, environmental impact, and economic costs. A hierarchical evaluation model based on the AHP was established for sludge treatment processes. Weight indicators were derived through expert questionnaire surveys and combined with empirical data to determine the comprehensive weights. The bootstrap method was applied to expand the sample size and ensure robust training of the ANN model. The ANN framework establishes mapping relationships between evaluation indicators and expected values. The AHP-ANN evaluation model demonstrated high predictive accuracy, achieving a maximum mean squared error (MSE) of 0.00052 in the test dataset. This model enabled the rapid assessment of parameter adjustments on evaluation outcomes and provided a quantitative basis for engineering optimization. Among the evaluated scenarios, the anaerobic digestion scenario (S1) demonstrated the best overall performance, characterized by low environmental impact and operational costs. Conversely, the incineration scenario (S3) exhibited the poorest overall performance, with high resource consumption, resulting in significant environmental impact and elevated operational costs.

Authors

  • Yuhan Wu
    State Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, China.
  • Diannan Huang
    School of Municipal and Environmental Engineering, Shenyang Jianzhu University, Shenyang 110168, China. Electronic address: dnhuang@sjzu.edu.cn.
  • Li Zhang
    Department of Animal Nutrition and Feed Science, College of Animal Science and Technology, Huazhong Agricultural University, Wuhan 430070, China.
  • Rongxin Zhang
    School of Life Sciences and Biopharmaceutics, Guangdong Pharmaceutical University, Guangzhou, 51006, PR China. Electronic address: rxzhang@gdpu.edu.cn.
  • Pengfei Yu
    Xijing Hospital, Fourth Military Medical University, Xi'an, China.
  • Yunan Gao
    School of Environmental and Chemical Engineering, Foshan University, Foshan 528251, China.
  • Dongbin Wu
    School of Agricultural and Animal Husbandry Engineering, Heilongjiang Polytechnic, Harbin 150100, China.
  • Yu Gao
    Department of Radiology Center, The First Affiliated Hospital of Xinxiang Medical University, Xin Xiang, China.