Algorithmically Designed Facades for Climatic Adaptability

Ar. Sushant Verma, rat[LAB]

Ar. Sushant Verma, rat[LAB]

“A building skin is designed using a number of parameters such as environmental conditions, structural feasibility and materiality, creating a range of designs. This allows choosing an optimum design for the façade based on the set criteria.”

Façades and building envelopes determine visual identity, character and expression of architecture. Design and aesthetics of building skins impart a character to a place, externally, and is a major parameter for building performance, internally. Building façades lie at the intersections of exterior and interior environment. They form an integral part of building aesthetics as well as building performance and an integrated practice of rational and intuitive approach is followed as a design & engineering process. This is where disciplines merge and science meets art.

A building skin consists of vertical (façade) and horizontal (roof) components which protect the building from direct external environment and helps in maintaining comfortable interiors along with providing structure and stability to the building. Building skins are a vital component to resolve issues of responsive architecture as they are a medium through which intelligence can be imparted to a building system to respond to an environmental stimulus. Thus key characteristic of an effective intelligent building skin is its ability to modify energy flows through the building envelope by regulation, enhancement, attenuation, rejection or entrapment.

A building skin is designed using a number of parameters such as environmental conditions, structural feasibility, and materiality, among others; all of which can be quantified as data. Since there is a large amount of data involved, computation becomes an essential part of dealing with the complex dynamics of design. Computation can be carried out through non-digital & digital processes aligned together and algorithms can be used of complex (and simple) problem solving. These problems include structural load calculations, material behaviour calculations, fabrication data extraction, etc. An algorithmic method of designing a building skin or a façade system can aid in controlling all the parameters and objective data that is embedded in design and a flexibility to use this data for creating a smart(er) system.

In mathematics and computer science, an algorithm is a self-contained step-by-step set of operations to be performed. Algorithms that exist perform calculation, data processing and automated reasoning. In façade design, an algorithmic approach is helpful to have a rational control on design, fabrication data, assembly process, material usage & cost. Design can be optimized for various criteria and one type of algorithms that can be used for multi-parametric problem solving is ‘Genetic Algorithm’. A parametric model is controlled by multiple parameters which can be used as ‘genes’. The set of parameters form the ‘genotype’ and output system forms the ‘phenotype’.Genetic algorithms are used after understanding the complexity of the multi-parametric problem to build an iterative system. Genetic algorithms are used to create a strong feedback network in the algorithm where real-time evaluations inform the initial stages of the algorithm.

In an on-going project by rat[LAB], a prominent part of the façade is designed to create an aesthetically dramatic effect while optimizing building performance and fabrication cost. A differentiation among members of façade is being parametrized through an algorithm that calculates the running length & cost of various building members. This allows us to change the base parameters (such as angular variation, dimensional shifts, etc.) and study its changing effect on fabrication cost & environmental performance. Multiple platforms are integrated to evaluate the results in real-time, as we change the influencing parameters and create a range of designs. This allows choosing an optimum design for the façade based on the set criteria. The algorithm is also developed to extract fabrication data from the parametric model, saving on design & production time, hence cutting down on project costs.

In another research done at the Architectural Association, London and The MAK Center, Los Angeles, the studio developed a dynamic façade system with the vision of challenging the static built environment against the dynamic natural environment. This façade system has been digitally retrofitted on the façade system of Makey Apartments originally designed by Rudolph Schindler.

The project questions the static nature of architectural spaces, encouraging dynamism and motion in architecture via movable building skins. In an exhibition at MAK Center Los Angeles in September 2014, the research lab presented a prototypical interactive installation made of tensegrity components embedded with sensors that responded to human movement. This was tested as a potentially suitable building type for Southern California and other semi-arid locations.

An optimization technique is necessary to evaluate the efficiency of a system which can involve multiple simulation parameters and multiple evaluation parameters.

A computational method based on abstracted logics of biological evolutionary processes is found to be appropriate as the optimization is an iterative process for seeking solution which was described by John Holland in the 1960’s which recognizes the fact that software already has a genotype and phenotype embedded in the code.

Parametric Model
Parametric model/system is controlled by several simulation parameters which are also called genes. These parameters control the form and function of the system which in-turn affect the efficiency. The parametric model, in this case, is developed using a software called Grasshopper that runs in Rhinoceros 3D Platform (Robert McNeel & Associates). The Dynamism of the system is controlled by a plug-in for Grasshopper, Kangaroo which is a Physics Engine and can simulate physical forces in a digital medium.

Real time Analysis
Based on the Evaluation Parameters, also termed as Fitness Criteria, in the language of genetic algorithms, Solar Exposure and Solar Radiation, two parallel scripts are developed using Python Script in Grasshopper. These are scripted using Solar Exposure Tool (in-built in Grasshopper) integrated with a plug-in called Geco which links Autodesk Ecotect and Grasshopper for Real-Time analysis. To find the efficiency of the system a real time analysis is required which evaluates, in parallel, with the change in genes of the system as outputs which are fed back to the input parameters for continuous iterations in an evolutionary process.

An Evolutionary Solver
Galapagos (plug-in for Grashopper) is integrated in the algorithm due to multiple permutations of output which can be obtained from the simulations. This evolutionary solver is required to collect, store, compare and analyse the large amount of data produced and finally give an optimized result according to the required fitness criteria.

Complex algorithms are coded while designing the system to study the emergent dynamic behaviour of the system with respect to angular changes, axial shifts and force transfers that take place when the system actuates with heat. This type of a building skin can potentially be used as free standing roof structures, building façades, window panels or installations that can be pre-programmed and configured in a way that it responds to the environmental conditions of any place, while using the potentials of shape memory allow and structural principles of tensegrity, and without needing any electrical source to operate. Computational technology is used extensively throughout the development of this project in a way that challenges the conventions of computation in architecture.

About the Author
Sushant Verma (M.Arch. Em.Tech.– AA London, B.Arch. SSAA New Delhi, MCoA India) is an architect and computational designer, currently leading research organization rat[LAB]- Research in Architecture & Technology. Former architect at Zaha Hadid Architects, London and a Sr. Editor at Arch2O, he is involved in education for computational and parametric design through rat[LAB] EDUCATION, which he founded  as an initiative to spread the idea of computation in design profession and academia. Recipient of MAK Schindler Award from Vienna/Los Angeles and a finalist for AIA Emerging Leaders Fellowship from Chicago, his work is widely published and exhibited in London, Los Angeles, New York, San Francisco, Taipei & New Delhi, among other places. He was recently shortlisted in New Delhi among 5 architects for ’20 under 35’ Exhibition at Alliance Française de Delhi.

You may also like...

Leave a Reply

Your email address will not be published. Required fields are marked *


Directory Wizard powered by