Multi-site household waste generation forecasting using a deep learning approach

Cubillos, Maximiliano

doi:10.1016/j.wasman.2020.06.046

Cited by 71 publications

(32 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Mean absolute error (MAE), mean square error (MSE), mean absolute percentage error (MAPE), correlation coefficient (R), and index of agreement (IA) were adopted to assess model performances in this study. MAE, MSE, MAPE are common model error indicators used in waste studies ( Abdoli et al 2011 , Abbasi and Hanandeh 2016 , Kannangara et al 2018 , Vu et al 2019b , Cubillos 2020 ), and they are selected here to facilitate rapid comparison with literature. R and IA measure the goodness of fit between the predicted and actual values.…”

Section: Methodsmentioning

confidence: 99%

“… Vu et al (2019b) modeled the effects of lag times on weekly yard waste time-series models and found that modelling error reduced by 50% at optimal lag times. In a Danish study, Cubillos (2020) attempted to model waste generation at household levels using a Long Short-Term Memory (LSTM) Neural Network. Wu et al (2020) , on the contrary, explored regional scale ANN models on municipal solid waste generation in China.…”

Section: Introductionmentioning

confidence: 99%

“…Inputs: temperature, precipitation, wind speed, and snow, demographic and socioeconomic variables Weather variables were important features for all prediction models Vu at al. (2019b) Austin, Taxes, US ANN time series Weekly yard waste prediction with variables of temperature, humidity, wind speed, precipitation, population, stock market index Lag times affected model performance Cubillos (2020) Herning,Denmark Long Short-Term Memory (LSTM) Neural Network Weekly household waste generation from 2011 - 2018. Inputs: temperature, humidity, wind speed, precipitation, atmospheric pressure, sun time The model can combine predictions of different households Wu et al (2020) 26 cities in China ANN Annual municipal solid waste prediction.…”

Section: Introductionmentioning

confidence: 99%

“…ANN-based models are versatile and applicable to many non-linear problems, provided that a good training data set is provided ( Xu et al 2021 ). Common inputs used in waste studies include socio-economic variables such as earnings and income, education level, employment status, dwelling and household characteristics, workplace and demographic parameters ( Younes et al, 2015 ; Kannangara et al, 2018 ; Kontokosta et al, 2018 ; Wu et al, 2020 ), or climatic variables such as temperature, humidity, wind speed, and precipitation ( Kontokosta et al, 2018 ; Cubillos, 2020 ). Some studies utilized both socio-economic and climatic variables on waste forecasting ( Kontokosta et al, 2018 ; Vu at al., 2019b ).…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

The use of a recurrent neural network model with separated time-series and lagged daily inputs for waste disposal rates modeling during COVID-19

Richter

et al. 2021

Sustainable Cities and Society

View full text Add to dashboard Cite

A new modeling framework is proposed to estimate mixed waste disposal rates in a Canadian capital city during the pandemic. Different Recurrent Neural Network models were developed using climatic, socio-economic, and COVID-19 related daily variables with different input lag times and study periods. It is hypothesized that the use of distinct time series and lagged inputs may improve modeling accuracy. Considering the entire 7.5-year period from Jan 2013 to Sept 2020, multi-variate weekday models were sensitive with lag times in the testing stage. It appears that the selection of input variables is more important than waste model complexity. Models applying COVID-19 related inputs generally had better performance, with average MAPE of 10.1%. The optimized lag times are however similar between the periods, with slightly longer average lag for the COVID-19 at 5.3 days. Simpler models with least input variables appear to better simulate waste disposal rates, and both ‘Temp-Hum’ (Temperature-Humidity) and ‘Temp-New Test’ (Temperature-COVID new test case) models capture the general disposal trend well, with MAPE of 10.3% and 9.4%, respectively. The benefits of the use of separated time series inputs are more apparent during the COVID-19 period, with noticeable decrease in modeling error.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

The use of a recurrent neural network model with separated time-series and lagged daily inputs for waste disposal rates modeling during COVID-19

Richter

et al. 2021

Sustainable Cities and Society

View full text Add to dashboard Cite

show abstract

“…The forecasting methods used for different types of waste predictions include machine learning (Kontokosta et al, 2018), small area estimation (Kontokosta et al, 2018), artificial neural networks (Akgül et al, 2020;Cubillos, 2020;Hao et al, 2019), regression analysis (Ghinea et al, 2016;Davidavičiene et al, 2012;Pavlas et al, 2020;Wei et al, 2013), ARIMA (Buhl et al, 2020;Ghinea et al, 2016;Ghomi & Marandi, 2016;Mwenda et al, 2014), artificial intelligence (Abbasi & Hanandeh, 2016), multivariate grey models (Intharathirat et al, 2015), Holt's double Exponential Smoothing (Islam & Huda, 2019), Holt-Winters Exponential Smoothing (Wąsik & Chmielowski, 2016), Winters multiplicative methods (Denafas et al, 2014).…”

Section: Methodsmentioning

confidence: 99%

Trends of Municipal Waste Flows, Composition, Treatment in Lithuania and Its Regions

Stankevičienė

Bužinskė

2021

International Scientific Conference „Contemporary Issues in Business, Management and Economics Engineering"

View full text Add to dashboard Cite

Purpose – to propose conceptual model for forecasting of waste trends and empirically implement the model based on the case of Lithuania and its regions. Research methodology – 1) scientific literature analysis on circular economy, zero waste and waste management, 2) gathering of statistical data on waste flows, composition and treatment 3) creation of conceptual model of forecasting with Exponential Smoothing for prediction of waste-related trends based on literature review. Findings – proposed conceptual model for prediction of waste-related trends is adequate for prognosis of waste flows, composition and treatment ways. The main forecasting results are that the total waste flows will increase in Lithuania, on a regional level, Alytus, Kaunas, Klaipėda, Telšiai, have a tendency of the increase in municipal waste flows. The results imply that in order to contribute to the reduction of waste, the active involvement on a regional level is necessary. Research limitations – the research can be extended with statistical data on waste of other countries to check adequacy of the conceptual model for waste-related trends prognosis. Practical implications – the findings of the research can be applied in planning and decision-making process of gov-ernment bodies on national or local level. The results are also useful for the general public in educational purposes. Originality/Value – the study provides original conceptual model for the forecasting of waste-related trends which provides robust results of predictions and can be replicated by different countries.

show abstract

Filter Method Feature Selection Techniques for Solid Waste Prediction Based on GRU Deep Learning Model

Batool,

Arbain,

Ghazali

et al. 2024

Lecture Notes in Networks and Systems

View full text Add to dashboard Cite

Multi-site household waste generation forecasting using a deep learning approach

Cited by 71 publications

References 33 publications

The use of a recurrent neural network model with separated time-series and lagged daily inputs for waste disposal rates modeling during COVID-19

The use of a recurrent neural network model with separated time-series and lagged daily inputs for waste disposal rates modeling during COVID-19

Trends of Municipal Waste Flows, Composition, Treatment in Lithuania and Its Regions

Filter Method Feature Selection Techniques for Solid Waste Prediction Based on GRU Deep Learning Model

Contact Info

Product

Resources

About