Open AccessPrediction of Industrial Hazardous Waste Production Based on Different Models
Author(s) -
Shuangliu Liu,
Liang Cheng,
Peng Chen,
Shunqing Xu,
Jun Gao,
Jing Lü
Publication year - 2019
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.179
H-Index - 26
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/384/1/012026
Subject(s) - range (aeronautics) , production (economics) , hazardous waste , standard deviation , value (mathematics) , absolute deviation , predictive modelling , econometrics , statistics , mathematics , engineering , economics , macroeconomics , aerospace engineering , waste management
Based on national hazardous waste production data from 2008 to 2011, GM (1,1) grey model, unit industrial output value model and average annual growth rate model are used for prediction studies of which the prediction results was compared with the actual values of hazardous waste from 2012 to 2016. Under the selected background data, all the three widely-used models have large deviations. The model with the smallest deviation range is the annual growth rate model of which the prediction result is 1.25-2.05 times the actual value. The model with the second small deviation range is the unit output value model of which the prediction result is 1.36-3.63 times the actual value. GM (1,1) has the largest deviation range, of which the prediction result is 1.46-7.58 times the actual value. The causes of deviation are analyzed, and the results show that, due to the influence of different factors, the background data significantly rocketed in 2011 and 2016, resulting in the instability of the model prediction. Based on the application of the models, the applicability of different models is analyzed. It is suggested that the grey model should be preferred when there are a few data; and the easy-to-operate unit output value model and annual growth rate model shall be preferred when there are sufficient historical data. Meanwhile, the influence of industry and policy factors should be comprehensively considered to improve and optimize the prediction methods and parameters of the models, so as to provide reference for quantitative prediction of hazardous wastes.