Predicting the presence of hazardous materials in buildings using machine learning
(2022) In Building and Environment 213.- Abstract
- Identifying in situ hazardous materials can improve demolition waste recyclability and reduce project uncertainties concerning cost overrun and delay. With the attempt to characterize their detection patterns in buildings, the study investigates the prediction potential of machine learning techniques with hazardous waste inventories and building registers as input data. By matching, validating, and assuring the quality of empirical data, a hazardous material dataset for training, testing, and validation was created. The objectives of the explorative study are to highlight the challenges in machine learning pipeline development and verify two prediction hypotheses. Our findings show an average of 74% and 83% accuracy rates in predicting... (More)
- Identifying in situ hazardous materials can improve demolition waste recyclability and reduce project uncertainties concerning cost overrun and delay. With the attempt to characterize their detection patterns in buildings, the study investigates the prediction potential of machine learning techniques with hazardous waste inventories and building registers as input data. By matching, validating, and assuring the quality of empirical data, a hazardous material dataset for training, testing, and validation was created. The objectives of the explorative study are to highlight the challenges in machine learning pipeline development and verify two prediction hypotheses. Our findings show an average of 74% and 83% accuracy rates in predicting asbestos pipe insulation in multifamily houses and PCB joints or sealants in school buildings in two major Swedish cities Gothenburg and Stockholm. Similarly, 78% and 83% of recall rates were obtained for imbalanced classification. By correlating the training sample size and cross-validation accuracy, the bias and variance issues were assessed in learning curves. In general, the models perform well on the limited dataset, yet collecting more training data can improve the model's generalizability to other building stocks, meanwhile decreasing the chance of overfitting. Furthermore, the average impact on the model output magnitude of each feature was illustrated. The proposed applied machine learning approach is promising for in situ hazardous material management and could support decision-making regarding risk evaluation in selective demolition work. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/3329aaf6-9226-4b72-b4a6-67a8bfc754fe
- author
- Wu, Pei-Yu LU ; Sandels, Claes ; Mjörnell, Kristina LU ; Mangold, Mikael and Johansson, Tim
- organization
- publishing date
- 2022-02-16
- type
- Contribution to journal
- publication status
- published
- subject
- in
- Building and Environment
- volume
- 213
- article number
- 108894
- pages
- 17 pages
- publisher
- Elsevier
- external identifiers
-
- scopus:85124704384
- ISSN
- 0360-1323
- DOI
- 10.1016/j.buildenv.2022.108894
- language
- English
- LU publication?
- yes
- id
- 3329aaf6-9226-4b72-b4a6-67a8bfc754fe
- date added to LUP
- 2022-02-22 16:29:09
- date last changed
- 2023-10-10 16:13:07
@article{3329aaf6-9226-4b72-b4a6-67a8bfc754fe,
abstract = {{Identifying in situ hazardous materials can improve demolition waste recyclability and reduce project uncertainties concerning cost overrun and delay. With the attempt to characterize their detection patterns in buildings, the study investigates the prediction potential of machine learning techniques with hazardous waste inventories and building registers as input data. By matching, validating, and assuring the quality of empirical data, a hazardous material dataset for training, testing, and validation was created. The objectives of the explorative study are to highlight the challenges in machine learning pipeline development and verify two prediction hypotheses. Our findings show an average of 74% and 83% accuracy rates in predicting asbestos pipe insulation in multifamily houses and PCB joints or sealants in school buildings in two major Swedish cities Gothenburg and Stockholm. Similarly, 78% and 83% of recall rates were obtained for imbalanced classification. By correlating the training sample size and cross-validation accuracy, the bias and variance issues were assessed in learning curves. In general, the models perform well on the limited dataset, yet collecting more training data can improve the model's generalizability to other building stocks, meanwhile decreasing the chance of overfitting. Furthermore, the average impact on the model output magnitude of each feature was illustrated. The proposed applied machine learning approach is promising for in situ hazardous material management and could support decision-making regarding risk evaluation in selective demolition work.}},
author = {{Wu, Pei-Yu and Sandels, Claes and Mjörnell, Kristina and Mangold, Mikael and Johansson, Tim}},
issn = {{0360-1323}},
language = {{eng}},
month = {{02}},
publisher = {{Elsevier}},
series = {{Building and Environment}},
title = {{Predicting the presence of hazardous materials in buildings using machine learning}},
url = {{http://dx.doi.org/10.1016/j.buildenv.2022.108894}},
doi = {{10.1016/j.buildenv.2022.108894}},
volume = {{213}},
year = {{2022}},
}