File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Estimating construction waste generation in the Greater Bay Area, China using machine learning

TitleEstimating construction waste generation in the Greater Bay Area, China using machine learning
Authors
KeywordsConstruction waste
Waste quantification
Greater Bay Area, China
Machine learning
Issue Date2021
PublisherElsevier Ltd. The Journal's web site is located at https://www.journals.elsevier.com/waste-management/
Citation
Waste Management, 2021, v. 134, p. 78-88 How to Cite?
AbstractReliable construction waste generation data is a prerequisite for any evidence-based waste management effort, but such data remains scarce in many developing economies owing to their rudimentary recording systems. By referring to several models proposed for estimating waste generation, this study aims to develop a reliable and accessible method for estimating construction waste generation based on limited publicly available data. The study has two objectives. Firstly, it aims to estimate construction waste generation by focusing on the Greater Bay Area (GBA) in China, one of the world’s most thriving regions in terms of construction activities. Secondly, it aims to compare the strengths and weaknesses of various waste quantification models. 43 sets of annual socio-economic, construction-related and C&D waste generation data ranging from 2005 to 2019 were collected from the local government authorities. By analyzing the data using four types of machine learning models, namely multiple linear regression, decision tree, grey models, and artificial neural network, it is found that all calibrated models, with their respective strengths and weaknesses, can produce acceptable results with the testing R2 ranging from 0.756 to 0.977. This study also reveals that the 11 cities in the GBA produced a total of about 364 million m3 of construction waste in 2018. The result can be used for monitoring the urban metabolism, quantifying carbon emission, developing a circular economy, valorizing recycled materials, and strategic planning of waste management facilities in the GBA. The research findings also contribute to the methodologies for estimating waste generation using limited data.
Persistent Identifierhttp://hdl.handle.net/10722/306655
ISSN
2023 Impact Factor: 7.1
2023 SCImago Journal Rankings: 1.734
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorLu, W-
dc.contributor.authorLOU, J-
dc.contributor.authorWebster, C-
dc.contributor.authorXue, F-
dc.contributor.authorBAO, Z-
dc.contributor.authorCHI, B-
dc.date.accessioned2021-10-22T07:37:44Z-
dc.date.available2021-10-22T07:37:44Z-
dc.date.issued2021-
dc.identifier.citationWaste Management, 2021, v. 134, p. 78-88-
dc.identifier.issn0956-053X-
dc.identifier.urihttp://hdl.handle.net/10722/306655-
dc.description.abstractReliable construction waste generation data is a prerequisite for any evidence-based waste management effort, but such data remains scarce in many developing economies owing to their rudimentary recording systems. By referring to several models proposed for estimating waste generation, this study aims to develop a reliable and accessible method for estimating construction waste generation based on limited publicly available data. The study has two objectives. Firstly, it aims to estimate construction waste generation by focusing on the Greater Bay Area (GBA) in China, one of the world’s most thriving regions in terms of construction activities. Secondly, it aims to compare the strengths and weaknesses of various waste quantification models. 43 sets of annual socio-economic, construction-related and C&D waste generation data ranging from 2005 to 2019 were collected from the local government authorities. By analyzing the data using four types of machine learning models, namely multiple linear regression, decision tree, grey models, and artificial neural network, it is found that all calibrated models, with their respective strengths and weaknesses, can produce acceptable results with the testing R2 ranging from 0.756 to 0.977. This study also reveals that the 11 cities in the GBA produced a total of about 364 million m3 of construction waste in 2018. The result can be used for monitoring the urban metabolism, quantifying carbon emission, developing a circular economy, valorizing recycled materials, and strategic planning of waste management facilities in the GBA. The research findings also contribute to the methodologies for estimating waste generation using limited data.-
dc.languageeng-
dc.publisherElsevier Ltd. The Journal's web site is located at https://www.journals.elsevier.com/waste-management/-
dc.relation.ispartofWaste Management-
dc.subjectConstruction waste-
dc.subjectWaste quantification-
dc.subjectGreater Bay Area, China-
dc.subjectMachine learning-
dc.titleEstimating construction waste generation in the Greater Bay Area, China using machine learning-
dc.typeArticle-
dc.identifier.emailLu, W: wilsonlu@hku.hk-
dc.identifier.emailWebster, C: cwebster@hku.hk-
dc.identifier.emailXue, F: xuef@hku.hk-
dc.identifier.authorityLu, W=rp01362-
dc.identifier.authorityWebster, C=rp01747-
dc.identifier.authorityXue, F=rp02189-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1016/j.wasman.2021.08.012-
dc.identifier.pmid34416673-
dc.identifier.scopuseid_2-s2.0-85112838316-
dc.identifier.hkuros328477-
dc.identifier.volume134-
dc.identifier.spage78-
dc.identifier.epage88-
dc.identifier.isiWOS:000692322200009-
dc.publisher.placeUnited Kingdom-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats