Workshops in english

0
Seats available

Event: November 7th-8th, 2022

The workshops will take place in November 7th and 8th. SIGraDi 2022 workshops will occur online via Blackboard Collaborate or similar video conferencing platform. The workshops at SIGraDi are an opportunity for students, teachers and professionals from Latin America to explore the possibilities of digital technologies as allies of the design process, sharing their experience with their peers from different parts of the planet. As in every year, we are sure that it will be an enriching and knowledge space that promotes appropriations

W1. Biomimetic Design and Prototyping: A digital toolkit for designers

Requirements
Rhinoceros3D, Grasshopper, Ultimaker Cura. Basics knowledge of Rhino and Grasshopper, basic understanding of 3D printing
Day 1
Introduction to biomimicry; Presentation of biological structures, biomimicry tools and databases; Exploration of modeling tools in Grasshopper; Participant work in groups
Day 2
Introduction to 3D printing; 3D printing and biomimicry; Participant work in groups; Feedback and guidance; Final presentations
Group
12 participants
Previous
Next

Workshop leaders: Alexandros Efstathiadis  & Ioanna SymeonidouUniversity of Thessaly, Volos, Greece

Βiomimicry is  the interdisciplinary science that studies nature’s models and draws inspiration from them to solve contemporary design challenges. However, processing, evaluating and imitating biological systems can be a daunting task for designers and architects. For this reason, specialized methodological tools have been developed to assist the biomimetic design process and bridge the existing knowledge gap. Furthermore, advancements in parametric, generative CAD software enable the extraction and emulation of complex biological structures. Algorithmic design allows for the generation of unlimited design permutations through a series of interactive parameters that can be altered by the user on-demand and in real-time, according to design requirements. Intricate biomimetic structures surpass the technical capabilities of traditional fabrication technologies. However, progress in additive manufacturing (AM) technologies like fused filament fabrication (FFF), enables the production of complex topologies. In nature, the information that guides the building process is stored in the DNA of organisms. Similarly the data that dictates the 3D printing of a model is stored in the form of a .gcode created by specialized slicing software.  Biomimetic design combined with 3D printing is driving a paradigm shift in sustainable design. The workshop will provide an analytical strategy that will enable designers to extract, emulate and prototype biomimetic structures. Initially, the concept of biomimicry will be introduced. A variety of biological structures will also be presented along with specific biomimetic tools and databases. Afterwards, a series of algorithmic and parametric modeling tools and plug-ins will be explored in Rhinoceros 3D and Grasshopper 3D. In the end, the models will be transferred to the slicing program Cura where the .gcode for the AM process will be generated. The intricacies of 3D printing complex biomimetic geometries will be analyzed.

W3. urbanGraph()

Topics
Programming Cultures, Data Analytics, Big Data, Location Based Social Network Data, Foursquare
Requirements
Python (latest version), Python (Advanced), PC or Laptop, one Mapbox account for each team, one Foursquare’s Places API account for each team
Day 1
Setting goal for each team, creating an algorithm to convert the data into the targeted information, coding the algorithm in Python, computing the process on Gephi, visualizing the results on Mapbox
Day 2
4) Continuação do desenvolvimento das torres; 5) Preparação dos arquivos para fabricação (orientado pelos ministrantes)/ 6) Apresentação e discussão dos resultados
Group
9 participants
Previous
Next

Workshop leader: Yasin Kutay Yüncüler, Keio University, Tokyo, Japan

urbanGraph() is a digital tool developed to collect data from Foursquare and designed for architects and urban designers, including those with no programming skills. Thus, the workshop’s primary goal is to discuss the potential of urbanGraph() for urban design and management on different scales and in various scenarios. With this motivation, participants will be asked to convert the data about a city they choose into knowledge while recording their process to be represented as an algorithm later. When all teams complete this phase, the accuracy of these algorithms will be tested by applying them to the cities other teams picked.  Considering the dataset collected by urbanGraph() as a directed graph, applying various algorithms proposed by graph theory is broadly possible. Also, some analysis functions are already designed by the lecturer and participants of previous events of this workshop series. So, the participants of this session will be allowed to utilize both the graph theory algorithms and mentioned Python functions. Moreover, teams will be encouraged to develop their original functions to analyze data. Eventually, all functions, algorithms, and maps produced during the workshop will be exhibited digitally.

W4. Optimizing into the probable future: predictive models of sustainable building performance

Topics
Climate Change; Design Space; Machine Learning; Data Visualization; Sensitivity Analysis
Requirements
Previous install of Rhino, Grasshopper, JMP, Excel, EnergyPlus, Open Studio, Radiance, Honeybee & Ladybug, TT_Toolbox, Lunchbox, Excel, and Weka for ML. Personal computer software compatible and basic knowledge of Rhino and Grasshopper
Day 1
Climate Model Scenarios; Generative Design; Parametric Analysis: Data Visualization
Day 2
Sampling the Design Space; Sensitivity Analysis; Predictions-based Machine Learning; Decision Making
Group
24 participants
Previous
Next


Workshop leaders: Marcelo Bernal, Ph.D., Perkins&Will Design Process Lab; Victor Okhoya, Ph.D., Perkins&Will Design Process Lab; Tyrone Marshall, AIA, NOMA, LEED AP BD+C, Perkins&Will Energy Lab; Cheney Chen, Ph.D., Perkins&Will Energy Lab: Mohamed Imam, Ph.D., Perkins&Will Vancouver

We aim to lead a hands-on discussion that asks how we might consider and anticipate the impact of climate change on buildings and the environment to speculate, imagine, and innovate new models of design intervention that can modulate better performance outcomes in an uncertain future. Although parametric analysis is an emergent data-driven approach to building performance analysis, few practitioners understand effective methods of evaluating parametric analysis data. Furthermore, even fewer understand predictive models and how to estimate uncertainties. In these two days’ workshops, the participants will learn how to simulate performance models based on weather files that represent different scenarios for climate changes (current, 2050, and 2080) and then generate design alternatives, analyze the performance of large design spaces, sample large design spaces, interpolate simulated data using machine learning predictive methods, visualize data for qualitative data exploration, and perform sensitivity analysis for quantitative assessment and estimation of the contribution of individual parameters in the overall performance.

W5. AR + Design: augmented reality immersive design method

Topics
Interactions and Collaborations, Mixed Realities (AR), Immersive Design
Requirements
Rhinoceros 6 or greater, Grasshopper, Fologram (free), Basic Knowledge of Rhino and Grasshopper, WIFI (network), Printer (optional)
Day 1
AR introduction; Basic AR interactive functions for design inputs; Design algorithm development
Day 2
AR immersive design workflow demonstration; Structural stability simulation; Participant design algorithm development and final outcome
Group
9 participants
Previous
Next

Workshop leader: Yang Song, University of Liverpool, UK

W5 aims to introduce the knowledge of exploring immersive parametric design through the Augmented Reality (AR) environment. As the quintessential 3D-4D design field, architectural design has been limited throughout its history by 2D or cumbersome 3D representation. Even though computer-aided architectural design and modelling software is widely used to produce digital 3D models, the conventional screen-based visualization methods for design and analysis are restrictive to how well the user understands the space on a computer screen, as the design is done outside the building site, hence there might be disparities between the design and final. This limitation may be eliminated by AR technology, which has become readily available, together with tools facilitating the easy creation of 3D-4D models as holograms onsite. Furthermore, with its interactive input features, AR can increase the potential for interaction between humans and data. As we immerse ourselves into rapidly developing AR, this technology can also radically change how one interacts with and experience the built environment, enhancing or altering or adding a new layer of information to the surrounding environment. This workshop explores how AR technology can change the ways of architectural design. Ideas like immersive design, as well as real-time modification experience and interaction with the built environment and the metaverse, will therefore actuate as the central core for the research streams.

W6. Get your Ph.D. done!

Topics
Life Long Learning, Ph.D., Education, Decision Making
Requirements
Software (N.A.). Pursuing a Ph.D./doctorate degree, consent to share your experience, actively participate, complete worksheets
Day 1
Why haven’t you finished?, Time management, Decide in advance
Day 2
Owning (your topic), Writing (your dissertation), Becoming (a Ph.D.)
Group
16 participants
Previous
Next

Workshop leader: Dr. Paula Gomez Z., Georgia Tech, Atlanta, USA

Finish your Ph.D. in a defined period of time. This is a participatory, non-specific topic workshop. It will cover techniques to get your Ph.D. done by understanding the drivers of your progress. It requires active participation, teamwork and sharing your academic experiences with the instructors and other participants. The outcomes are two. First, a deep understanding of a set of techniques to be able to finish the dissertation, and second, an implementation plan for the milestones: Qualifying paper and exam, topic proposal, dissertation writing, Ph.D. defense, and graduation!

W7. Architectural Intelligence: Multimodal Machine Learning Applications in Design

Topics
Programming Cultures, Machine Learning, Generative Design, Neural Networks, Metaverse
Requirements
Houdini Apprentice (free version), Rhino & Grasshopper and Python will be provided; Google Colab (1 month subscription $9.99), Intermediate knowledge of Rhino; a basic understanding of scripting languages and Houdini is welcome but not a requirement.
Day 1
Students will develop three sequential exercises related to 1) envelope making; 2) plan making; 3) a quasi-architectural or metaverse proposals to challenge the first two assignments.
Day 2
Creating constructible and discretized forms that hybridize the 2D image outputs. The two primary workflows that we shall delve into are a NURBs based creation of forms in Rhino and several Grasshopper plugins (Monolith, Pufferfish, MeshEdit, Human, Python etc) and vozelized space-based methods using Houdini.
Group
20 participants
Previous
Next

Workshop leaders: George Guida & Indrajeet Halder, Harvard University, GSD, Cambridge, USA

We have now reached a historical moment of convergence between text and image processing, bringing forward new multimodal creative processes. Dalle2, StableDiffusion, and Midjourney are challenging current cultural practices of architectural production where a single text input can generate thousands of novel images. This workshop will reposition the role of language and semantics within architecture and introduce students to the opportunities of text-image models and 3D form generation for early design stages.  Generated through a combination of the latest machine learning (ML) models, students will be taught how to generate images from text inputs, and use these to inform a computational process of 2D to 3D reconstructions. The parameters behind each model and semantics specificity of each text input establish a new creative exchange between human and machine, where human agency finds new meanings. Following an introduction to ML applications architectural design, students will develop three sequential exercises related to 1) envelope making; 2) plan making; 3) a quasi-architectural or metaverse proposals to challenge the first two assignments. The course thus engages ways to reposition language within the design process through new methods and design metrics and considered these practices as challenging current cultural practices of architectural production.

W12. ML Plugin : develop a Machine Learning plugin for Grasshopper

Topics
Programming Cultures, Generative Design, Machine Learning, Artificial Intelligence, Develop a Digital Tool
Requirements
Rhinoceros 6 or greater, Grasshopper, Visual Studio 2019 or greater. Basic knowledge of RH, GH and C# programming, PC or Laptop (1.8 GHz or faster processor. Quad-core or better recommended RAM: 2-8GB (2.5 GB minimum if running on a virtual machine) HD space: 800MB-210 GB of available space, typical installations: 20-50 GB of free space.)
Day 1
Basics of programming with C#, Basics of machine learning, Examples of using ML for architectural goals, Generative design using ML, Design evaluation with ML
Day 2
How to develop a plugin, Using ML Libraries, Building the customized plugin
Group
10 participants
Previous
Next

Workshop leader: Mohammad Pourfooladi, Art University of Isfahan, Iran

In 1950, Alan Turing asked, “Does a machine think?” A simple question that was the beginning of research on machine learning. Today, “Artificial Intelligence” (AI) is used in most fields of everyday life, and learning it can greatly help in the development of new technologies. “Machine Learning” (ML) is one of the parts of artificial intelligence that must be learned to work in this field. The learning process in machine learning begins with data as input until the machine uses them to find the patterns in that data set and make better decisions based on the discovery of their patterns and the insights gained. Machine learning and how to use it for scientific research can be a powerful tool for architects. This workshop aims to provide a basic learning platform for entering the field of machine learning and how to write code for architecture, as well as building the tools desired by the architect for research and design with the help of artificial intelligence tools. In this workshop, students learn how to use the machine learning libraries in the Rhino software to create a plugin that is customized according to their needs.