Abstract
The construction industry is compelled to evolve toward industrialized and sustainable means and processes. Indeed, the Portland cement production alone is responsible for 4 to 8% of global greenhouse gas emissions. In this context, concrete 3D printing demonstrates a significant potential for the reduction of material use (formwork-free, structural optimization), but the majority of 3D printing materials still hold a high carbon intensity mainly due to the rheological constraints associated with pumpability, extrudability and buildability. The present study proposes to integrate a parametric life cycle assessment model in the multi-objective optimization of a 3D printing mortar in order to minimize the environmental impact with a significantly reduced mixture design workload. Applied to a limestone calcined clay cement-based mortar with high cement substitution, a non-dominated sorting genetic algorithm is used to decrease the climate change score while maintaining a set of rheological properties which are predicted by artificial neural networks and define the printability. In this way, the non-linear behavior of cementitious materials is represented in the design and the multiplicity of independent and dependent variables is handled adequately. As a result, starting from a reference mixture with already low cement content (6.8 wt%) and a dataset of 20 to 30 mixtures, this methodology allows the identification of more sustainable printable mixtures in as low as 7 additional formulations. Besides, as the process advances, its efficiency is enhanced. This methodology is reproducible with locally sourced materials and for the majority of 3D printing materials, which are usually designed through empirical trial and error. Therefore, it could also be applied with different objective functions to bicomponent materials, which offer added printing flexibility. This study introduces a systematic optimization process which establishes the sustainability at the core of its objectives and includes new tools for the formulation of cementitious materials.
| Original language | English |
|---|---|
| Title of host publication | Advances in Information Technology in Civil and Building Engineering - Proceedings of ICCCBE 2024 |
| Editors | Adel Francis, Edmond Miresco, Silvio Melhado |
| Publisher | Springer Science and Business Media Deutschland GmbH |
| Pages | 421-428 |
| Number of pages | 8 |
| ISBN (Print) | 9783031873638 |
| DOIs | |
| Publication status | Published - 1 Jan 2025 |
| Externally published | Yes |
| Event | 20th International Conference on Computing in Civil and Building Engineering, ICCCBE 2024 - Montreal, Canada Duration: 25 Aug 2024 → 28 Aug 2024 |
Publication series
| Name | Lecture Notes in Civil Engineering |
|---|---|
| Volume | 629 LNCE |
| ISSN (Print) | 2366-2557 |
| ISSN (Electronic) | 2366-2565 |
Conference
| Conference | 20th International Conference on Computing in Civil and Building Engineering, ICCCBE 2024 |
|---|---|
| Country/Territory | Canada |
| City | Montreal |
| Period | 25/08/24 → 28/08/24 |
UN SDGs
This output contributes to the following UN Sustainable Development Goals (SDGs)
-
SDG 7 Affordable and Clean Energy
-
SDG 12 Responsible Consumption and Production
-
SDG 13 Climate Action
Keywords
- 3D printing
- LC3
- Machine learning
- Optimization
- Sustainability
Fingerprint
Dive into the research topics of 'Multi-Objective Optimization of a Sustainable LC3 Mortar for 3D Printing'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver