Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Machine learning based construction simulation and optimization
Harbin Institute of Technology Resources Engineering, Harbin, CHINA.
Harbin Institute of Technology Resources Engineering, Harbin, CHINA.
Luleå University of Technology, Department of Civil, Environmental and Natural Resources Engineering, Industrilized and sustainable construction.ORCID iD: 0000-0002-4695-5369
2018 (English)In: Proceedings of the 2018 Winter Simulation Conference / [ed] M. Rabe; A.A. Juan; N. Mustafee; A. Skoogh; S. Jain; B. Johansson, IEEE, 2018, p. 2025-2036, article id 8632290Conference paper, Published paper (Refereed)
Abstract [en]

Building construction comprises interaction and interdependence among processes. Discrete-event simulation (DES) is widely applied to model these processes interaction. To find optimal construction plans, optimization technique is usually integrated with DES. However, present simulation-optimization integrated method directly invokes simulation model within optimization algorithms, which is found significantly computationally expensive. This study proposes a machine learning based construction simulation and optimization integrated method. After trained by DES, the machine learning model accelerates simulation-optimization integration by nearly real-time providing fitness evaluation within optimization. This method was implemented into a real construction project for construction time-cost-environment optimization. Results show that proposed machine learning based method significantly reduce computing time compared with original simulation-optimization integration. Less than 1% of construction cost and time improvement were miss, while greenhouse gas emissions obtained same performance. The new method could be a more effective DES and optimization integration approach for practical engineering application.

Place, publisher, year, edition, pages
IEEE, 2018. p. 2025-2036, article id 8632290
National Category
Construction Management
Research subject
Construction Management and Building Technology
Identifiers
URN: urn:nbn:se:ltu:diva-73236DOI: 10.1109/WSC.2018.8632290Scopus ID: 2-s2.0-85062601673ISBN: 9781538665725 (print)OAI: oai:DiVA.org:ltu-73236DiVA, id: diva2:1297010
Conference
2018 Winter Simulation Conference, WSC 2018; The Swedish Exhibition and Congress CentreGothenburg; Sweden; 9 - 12 December 2018
Available from: 2019-03-18 Created: 2019-03-18 Last updated: 2020-09-15Bibliographically approved
In thesis
1. Environmentally Friendly Construction Processes Under Uncertainty: Assessment, Optimisation and Robust Decision-Making
Open this publication in new window or tab >>Environmentally Friendly Construction Processes Under Uncertainty: Assessment, Optimisation and Robust Decision-Making
2020 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

The construction processes of building and civil infrastructure are broadly recognised as large contributors to the environmental degradation, which cause environmental impacts directly and indirectly by massive energy use, intensive greenhouse emissions, and significant resources consumption. The report from Royal Swedish Academy of Engineering Sciences (IVA) and the Swedish Construction Federation (Byggföretagen) shows that the total carbon dioxide equivalents emitted per year from construction processes are the same size as emissions from all of the cars in Sweden, and more than that is generated by all lorries and busses. Under the current scenario of practices and technologies, 35-60% of the remaining carbon budget of Paris Agreement in next 30 years would be taken by construction processes. To follow the fossil free Sweden initiative, Swedish construction and civil engineering sector has set the roadmap to reduce the greenhouse gas emissions of construction processes by 50% from 2015 to 2030 and reach net-zero emissions by 2045.

Construction processes are usually carried out in contexts full of uncertainty, which causes significant challenges to achieve the target of environmentally friendly construction. The reasons can be summarised as three aspects. Firstly, most of related environmental impact assessment methods still base on static data, which lack the ability to capture the influence of uncertainty in an environmental assessment. Secondly, the uncertainty in construction processes leads to high level of computational loads, which results in that current studies present limitations on effectively providing real-time environmental optimisation. Thirdly, the robust decision-making has been proposed as an enabling method to yield decisions with robust performances with relatively less influences by uncertainty. However, current robust decision-making methods sometimes are not applicable for construction environmentally friendly decisions because the knowledge of uncertainty such as prior probability distributions are partly unknown due to characteristics of construction.

The overall purpose of this thesis, therefore, was to formulate a holistic approach to assess, optimise and provide robust decision-making for environmentally friendly construction under the uncertain contexts. The developed environmental impact assessment method integrates discrete-event simulation (DES) and process-based life-cycle assessment (pLCA). It takes advantage of discrete-event simulation to reinforce the uncertainty analysis ability of conventional environmental assessment methods. The optimisation method achieves real-time environmental optimisation by introducing machine learning (ML) technology into simulation-based optimisation. It significantly reduces the computational loads of optimisation by the ML’s real-time feedback ability during uncertainty quantification. The developed robust decision-making method combines discrete-event simulation (DES) and data mining (DM) technologies to address the uncertain contexts of construction. It utilities construction performance dataset, i.e. a data-driven method, to quantify the robustness and identify the vulnerability of environmentally friendly decisions in the situation of partly unknown probability distributions.

To achieve research purpose, the research is designed as an explorative procedure in a loop of (1) problem identification, (2) method development, and (3) method examination. In the first step, the requirements of environmentally friendly construction in real practices are identified, the current limitations and knowledge gaps that block the environmentally friendly construction are revealed. In the second step, to solve the identified problems, the holistic approach is designed. The theoretical methods that base on relevant theories are established, and prototypes that base on relevant technologies are developed. In the last step, the holistic approach is implemented into real construction cases. The procedure will be loop to solve identified problems until the research purpose has been fully achieved.

The research provides a systematic tool to handle uncertainty and to support environmentally friendly construction practices for project decision-makers. Firstly, developed approach enables to assess the environmental impacts of construction processes involving uncertainty, which help to better understand the influences of uncertainty and develop construction planning that can improve the environmental performance. And the optimisation section enables the real-time decision support of an environmental optimisation by considering multi-objective and in a great deal of construction alternatives, which helps to efficiently narrow down numerous construction alternatives and provide practical environmentally optimal planning. Finally, it provides the decision-makers with robust environmentally friendly decisions on the construction planning that are least affected by uncertainty, and provides vulnerable scenarios of uncertain factors for an informed uncertainty management in the progress of construction.

Place, publisher, year, edition, pages
Luleå: Luleå University of Technology, 2020
Series
Doctoral thesis / Luleå University of Technology 1 jan 1997 → …, ISSN 1402-1544
Keywords
Construction processes, environmental impacts, uncertain factors, decision-making
National Category
Construction Management
Research subject
Construction Management and Building Technology
Identifiers
urn:nbn:se:ltu:diva-80793 (URN)978-91-7790-651-3 (ISBN)978-91-7790-652-0 (ISBN)
Public defence
2020-11-20, T2109 and Zoom, Campus of Luleå University of Technology, 97187, Luleå, 09:30 (English)
Opponent
Supervisors
Available from: 2020-09-16 Created: 2020-09-15 Last updated: 2020-11-02Bibliographically approved

Open Access in DiVA

No full text in DiVA

Other links

Publisher's full textScopus

Authority records

Lu, Weizhuo

Search in DiVA

By author/editor
Lu, Weizhuo
By organisation
Industrilized and sustainable construction
Construction Management

Search outside of DiVA

GoogleGoogle Scholar

doi
isbn
urn-nbn

Altmetric score

doi
isbn
urn-nbn
Total: 153 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf