Multidisciplinary Design Optimization Framework For Multi-Phase Building Design Process - Technology Demonstration Using Design Of Office Building And Robotically 3D Printed Habitat
Open Access
- Author:
- Muthumanickam, Naveen Kumar
- Graduate Program:
- Architecture
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- August 12, 2021
- Committee Members:
- Timothy Simpson, Co-Chair & Dissertation Advisor
Jose Pinto Duarte, Co-Chair & Major Field Member
Lisa Iulo, Major Field Member
Gordon Warn, Outside Unit & Field Member
Loukas Kalisperis, Major Field Member
Somayeh Asadi, Program Head/Chair - Keywords:
- Building Design Optimization
Multidisciplinary Design Optimization
Concurrent Engineering
Model and Simulation-Based Engineering
Metamodelling
Multifidelity modelling
Surrogate Modelling
Building Information Modelling (BIM)
Energy Modelling
Daylighting Simulation
Robotic Constructability
3D Printing
Additive Concrete deposition
Machine Learning - Abstract:
- A growing body of work in the building design field acknowledges that the design of buildings is growing into a complex multidisciplinary problem that must reconcile multiple conflicting environmental, economic, social, and technical design objectives. This fact has been driving research in Architecture, Engineering and Construction (AEC) toward development of rigorous evidence-based (simulation-based) design optimization frameworks that generate and evaluate numerous building design alternatives using optimization in concert with simulation and analysis models. While such frameworks are well-known and widely employed in engineering design domains, similar efforts in the AEC field are largely compartmentalized into separate domains such as structural, energy, daylighting, and other performance factors. Most of the building design optimization efforts are either multidisciplinary optimization (MDO) confined to just a specific design phase or single disciplinary optimization (SDO) spanning across multiple phases. This dissertation is a three-part presentation on (1) identifying technological deficiencies impeding the implementation of multi-phase MDO in AEC field; (2) developing technologies to address the identified deficiencies, thereby constituting a multi-phase MDO framework for AEC field; and (3) test the developed MDO framework by implementing it in representative building design problems. Specifically, the developed MDO framework includes a generative algorithm for batch modelling of large sets of building designs, metamodels for rapid analysis of such large sets of designs for energy, daylighting (using machine learning) and constructability and computational framework to support interoperability between these tools. The developed MDO framework is used to design a sample office building (optimized for energy and daylighting) and a 3D Printed habitat (optimized for structural, spatial layout and robotic constructability). Finally, the advantages and limitations of the developed framework along with learnings and future avenues for research in AEC field are also discussed.