Open AccessBalancing data requirement and modelling quality in neighbourhood life cycle assessments
Author(s) -
O. Zara,
Gustavo Moura. Guimarães,
Mariana Zibetti,
Karine Regina Pompermayer,
Iúri da Costa Leite,
Vanessa Gomes da Silva
Publication year - 2020
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 - 10.1088/1755-1315/588/4/042030
Subject(s) - neighbourhood (mathematics) , computer science , archetype , sampling (signal processing) , data collection , data quality , key (lock) , compromise , data mining , operations research , data science , risk analysis (engineering) , operations management , engineering , mathematics , statistics , medicine , art , mathematical analysis , metric (unit) , literature , computer security , filter (signal processing) , computer vision , social science , sociology
When modelling complex systems such as cities, a quality-complexity compromise is to subdivide them into smaller cells. Life Cycle Assessment can help to comprehensively handle urban intricacies but is a data-intensive technique. Balancing data requirement and collection feasibility while acknowledging uncertainty become key. Methods: This research explored top-down and bottom-up approaches to generate information input for environmental modelling at neighbourhood scale and to identify strategies to improve modelling while balancing data collection needs. SimaPro v.9 supported the assessments. Results: Influence of elements like interior finishings and wall partitions is not captured by the top-down approach, but should not be neglected, for their impacts are substantial. Modelling can be improved by application of cut-off rules to limit data requirements and cluster sampling techniques to derive a minimum range of archetypes to adequately describe the studied area. Finally, an evolutive hybrid approach is suggested to gradually improve both background archetypes and foreground bottom-up objects.