A comparative study of black-box and white-box data-driven methods to predict landfill leachate permeability

Mahdi Ghasemi; Mehrshad Samadi; Elham Soleimanian; Kwok-Wing Chau

doi:10.1007/s10661-023-11462-9

A comparative study of black-box and white-box data-driven methods to predict landfill leachate permeability

Environ Monit Assess. 2023 Jun 19;195(7):862. doi: 10.1007/s10661-023-11462-9.

Authors

Mahdi Ghasemi¹, Mehrshad Samadi¹, Elham Soleimanian², Kwok-Wing Chau³

Affiliations

¹ Department of Civil Engineering, Iran University of Science and Technology (IUST), Narmak, Tehran, Iran.
² Department of Building, Civil, and Environmental Engineering, Concordia University, Montreal, Canada.
³ Department of Civil and Environmental Engineering, Hong Kong Polytechnic University, Kowloon, Hong Kong, China. cekwchau@connect.polyu.hk.

PMID: 37335361
DOI: 10.1007/s10661-023-11462-9

Abstract

Due to the dynamic and complexity of leachate percolation within municipal solid waste (MSW), planning and operation of solid waste management systems are challenging for decision-makers. In this regard, data-driven methods can be considered robust approaches to modeling this problem. In this paper, three black-box data-driven models, including artificial neural networks (ANN), adaptive neuro-fuzzy inference systems (ANFIS), and support vector regression (SVR), and also three white-box data-driven models, including the M5 model tree (M5MT), classification and regression trees (CART), and group method of data handling (GMDH), were developed for modeling and predicting landfill leachate permeability ([Formula: see text]). Based on a previous study conducted by Ghasemi et al. (2021), [Formula: see text] can be formulated as a function of impermeable sheets ([Formula: see text]) and copper pipes ([Formula: see text]). Hence, in the present study, [Formula: see text] and [Formula: see text] were adopted as input variables for the prediction of [Formula: see text] and evaluated for the performance of the suggested black-box and white-box data-driven models. Scatter plots and statistical indices such as coefficient of determination (R²), root mean square error (RMSE), and mean absolute error (MAE) were used for qualitative and quantitative evaluations of the effectiveness of the suggested methods. The outcomes indicated all of the provided models successfully predicted [Formula: see text]. However, ANN and GMDH had higher accuracy between the proposed black-box and white-box data-driven models. ANN with R² = 0.939, RMSE = 0.056, and MAE = 0.017 was marginally better than GMDH with R² = 0.857, RMSE = 0.064, and MAE = 0.026 in the testing stage. Nevertheless, an explicit mathematical expression provided by GMDH to predict k was easier and more understandable than ANN.

Keywords: ANN; Data-driven models; GMDH; Leachate percolation; Permeability; Simulated landfill.

MeSH terms

Environmental Monitoring / methods
Neural Networks, Computer
Permeability
Waste Management* / methods
Water Pollutants, Chemical* / analysis

Substances

Water Pollutants, Chemical