2018-09-27

AndreasBrekke_RamonFernandezZaragoza_MilouvanderLinden_FinalPaper_ArtificialAutomations

Automated design
The working of the design process nowadays


Type: Paper, Final paper

Title: Automated design

Author: Andreas Brekke
Ramon Fernandez Zaragoza
Milou van der Linden


Summary: The digitalisation is taking over the entire society, also the field of architecture
is involved in this development. The design process is becoming more and
more automated and Flux is one of the platforms that pursuing the potential
of automated design.
Keywords: Automation, Digitalisation, Flux, Architecture, Database, AI

Key-thinkers: Jen Carlile, Co-founder of Flux

Tutor: Sandra Manninger
Course: Artificial automations

Faculty: RMIT University (Royal Melbourne Institute of technology)

Department: Architecture

Hand-in date: 2018-09-28

Language: English

Submitter email: s3741538@student.rmit.edu.au (Milou van der Linden)

Study number: s3741519 (Andreas Brekke)
s3749127 (Ramon Fernandez Zaragoza)
s3741538 (Milou van der Linden)
Abstract
The role of the architect is a changing principle. Automation is becoming a leading element in the field of architecture with the aim to take over the design process by digital tools. Where we first went into a design process trusting our intuitions, we do not dare to do that anymore and we stick to tools that give us concrete information on which we nowadays trust. Therefore, it is important that we know how to use the digital tools and how far we can go with the automation of the design process.  

Google is one of the main companies pursuing the potential of automated design. They came up with Flux, a service based platform used by architects, engineers and construction workers to share models, markups and more. Flux started with the knowledge that in the future 70 percent of the population will live in the city centres. The approach is mainly focussed on speeding up the design process, make buildings more efficient and design eco friendly. To reach this goal flux platform has three different divisions in using, collaboration platform, access to urban facts through the site extractor and a modelling interface.

To define the working of Flux, the importance of the role of the architect can be discussed. To make some approaches for the future there is taken a look to exciting digital tools in other field. Architecture could be more automated that Flux is developed now.

Table of contents


Introduction

At the moment we are developing a new world, one day you will see the diversity of information packages is endless and that we no longer trust on our own intuitions. Only the ones that are programmed are familiar to us. We have to face that the boundaries of a human body are limited and there could be a future whereby cyborg will replace the human. They are faster, more intelligent and controlled. But what an extremely important point is that only our body is replaceable, our ghost will always be from ourselves. If our body is no longer part of our own system, we do not dear to trust our own intuitions anymore. We could ask ourselves, till what extend are cyborgs ok?

The world entered the rapid area of change. In our modern society, science is pushed forward with leaps and bounds. It is generally accepted that our lives are affected by new developments of technology. These technologies makes life easier, faster and more liveable then before. Also architecture is taking part of the advancement of modern technology. Our design process is going to look like cyborgs, there are platforms that are going to take over this process. Where is the architect in this picture, what could the role of the architect be in the future that consists of technologies and automated tools. A computer can do our taxes, drive our cars, do our communication and book our holidays, but could it also design our buildings? Technical development in the field of architecture is rising, put some facts and requirements into an automated tool and a building design will come up. Design programs such as Flux, Revit, Archicad, Autocad and Rhino could be used for this development and is accessible for everyone. That is remarkable, because that means that everyone can use those tools to design their own buildings without any knowledge of designing. The program knows the rules a building needs and the client knows his requirements. These two elements together are enough to start a design process. But where is the creativity of the architect in this new development. Architecture is going to lose his qualities by making the process automated. Is it possible to make the design process fully automatic, how far are the developments and how far is the automatisation of the process acceptable? These questions will be discussed in this paper.

In this paper we will research the automation of the design process. First we will discuss what a design process in general is, looking to the past and the future. The topic automation and digitisation will come up in the next chapter, resulting in a research on Flux, a platform to collaborate, share and design. Flux started with the question how are we going to deal with the growth of the population in the inner city centre. Started from this question Flux introduced three different divisions, one of them is the modelling interface. A tool that can design buildings.

Glossary



1. The architectural design process

In the field of architecture, the design process is one of the most important parts for making good architecture. The design process shows all the steps, information and needs a building has to include. Nowadays the technology is taking over the design process. The way an architect works in a design process is through designing. Designing is a act of working out a form, making sketches or plans. During this process there is some research, though, modeling, coding and adjustment involved. Some of this basic tools could be automatic in the future with the thoughts that computers will have the ability to be programmed just like a human. There are already a couple of researchers that made design tools shown the faith that quantitative of a building can be done by automatisation. Beside the quantitative aspects of architecture, like constructability, the qualitative characteristics are nearly as important. Thinking of context, aesthetics and spatial experience as criteria of qualitative. This part of architecture is still done and could only be done by architects as far as we know, but the development on the automation of qualitative is rising. The rise of big data and cloud-based computing has renewed the notion that computers can tackle the complex task of designing on the quantitative aspects.

Autodesk and google are the two main companies pursuing the potential of automated design. Autodesk chief technology officer Jeff Kowalski introduced in this presentation at Autodesk University (2014) that his company is “developing a CAD system that learns the same way we do, by referencing mechanical engineering textbooks, building codes, part catalogs and even by observing real-world examples.” In his nine years as CTO, he said, “This is the biggest, most fundamental change that I’ve ever seen coming our way.” (Davis, D. June 15, 2015)

A more interesting development in the field of automated design is Google X. They came up with Flux.
Flux is a startup in San francisco, building an open platform for the architect, engineering and construction industry. They got their start as Google X and Moonshot and became an independent company in 2012.

2. The architectural design process with Flux

Flux is a popular platform used by architects, engineers and construction workers to share models, markups and more. This new approach of automated design will change the way architect works nowadays. It will shift the focus from researching, analysing and processing, so the architect has more time for the aesthetics of the building. To understand what Flux is, how it works and what it could bring to architects we have to deepen in this platform.

2.1 Flux

Flux is a ‘moonshot’ lab, dedicated to finding radical solutions to huge problems. One of the problems in this day and age is the grow of the inner cities. Flux figured out that the world’s urban areas adjust to the fact that over 6 billion people will live in cities by 2045. The Flux software was introduced to solve this problem and find a solution to tackle this important issue.

To tackle this issue, the design process must change. Co-founder of Flux, Jen Carlile (June 17, 2016) said that “The big challenge is we’re asking people to think about design a little bit differently. It’s a little bit scary to people, like, Are you going to automate our job away? That’s not what we’re trying to do. It’s true computer-aided design. We want to give designers and architects superpowers.” What she means by this statement is that Flux needs to invest in speeding up the design process by taking tasks over from the architect. They started with the idea of creating a singular design tool that would help architects more efficiently design buildings, but they realised that every architecture firm has its own way of designing things. So they shifted the focus to make a collaborative tool that makes it easy to transfer data and files between commonly-used design programs such as Revit, Excel and Sketchup. This is the way to speed up the design process as a solution for cities that quickly need to become more dense in order to accommodate an influx of new resident. Later the software started to provide data about urban facts and requirements of new buildings and became a basic tool to design. Kaufmann (October 2015) stated, that Flux is an architectural model and no technology on his own. What she means is that Flux helps the architect to improve their designs by sharing knowledge, facts and models, easy, fast and efficient through technology but everything is still architectural.

The focus is on making the design process faster and buildings more efficient, therefore Flux is using artificial intelligence (AI). AI is used to cut waste in the design process, and allowing users to easily share information. AI is an intelligence demonstrated by machines, in contrast to the natural intelligence displayed by humans. It has functions that humans associate with other human minds, such as "learning" and "problem solving”. This means that AI in Flux avoid work that could be done twice by different architects. Flux is a collaborative tool and AI ensure that all the information the architect needs for his design is provided before they start researching and analysing. All the information will be transported through this system. Now the architect knows immediately what information is missing and can add this to the Flux platform. According to Yuan (June 17, 2016), it could now theoretically save architects weeks of work on a building project. For cities dealing with rapid growth, this could be key to speeding up the building design process by taking work away that architects do every day. It seems that this development stimulate the purpose of taking all the standard components away and shift the focus to making buildings more efficient and eco-friendly.

This idea of Flux is to manage the collaborative tool and provide the tool for data about urban fact and requirements of new buildings. Harnessing shared knowledge and ideas to build eco-friendly homes and a sustainable future for the construction industry, ­improving the places we live and work. After making those tools, a last development came ahead, Flux made a version called, the modelling interface. It is an interface that includes all the rules a building needs, to grow a design by itself (Carlile, J. 2014).  

2.2 The three divisions of Flux

Flux started as a company with one investment, making a platform to share information easier, faster and efficient. Soon, Flux grew out to a platform that combines three different divisions. Firstly, Flux came with the ability to share data and models easily between architects, engineers and construction workers. They were creating a database with all the general information that is needed and everyone can append new information about problem so that ones it is solved it will always be solved and no one is doing work twice. This collaborative tool helps to speed up the design process and work more efficient on your design by shifting the focus from researching and analysing to the aesthetics of the design. Beside that platform of collaboration, Flux introduced the site extractor. The site extractor provide urban fact about the environment where you are going the build your design. It is to organise the site data and dictates what you can build on a site, like how large can a building be, how tall can it be and dictates what is already there. It knows if your building is in a historic place, if it is near a significant public building, is there a heritage tree on the site or are we in a water sheat and is there a high limit (Carlile, J. 2014). This information provided from Flux is useful for everyone and has nothing to do with the automation of the design process but the computer is doing the site analysis on the environment, the history, climate and program. The general information and knowledge that is the same for everyone. The last development in Flux is the modeling interface as earlier mentioned, where parametric buildings are automatically adapted to site constraints. By connecting dots, Flux automatically design a building for any site and by the analytics from codes, materials, and widgets that calculate emissions can be used to reimagine building design.  

Flux wants to be a software where the other softwares like CAD, Revit, Sketchup ect. can work on. It is not trying to be a plug-in or package for other software. What Flux tries to do is making a platform where all the different existing building modelling software is compatible and work horizontally with a common language which is Flux. (Flux.io, Dec 9, 2015)

2.2.1 The working of the collaboration tool

After knowing what the aim of Flux is, we come to the point that we want to know how it works in the field of architecture. Flux can be used to exchange data between Excel, Grasshopper, Dynamo, SketchUp, Revit and more via the web. Flux connects your design tools enabling smooth exchange of data between design team members and parties. They call it collaboration in real-time. Through the web browser you have instant access to your data and geometry and you can share changes immediately. (Flux.io, Dec 9, 2015) This real-time collaboration tool ensures an ideal workflow by supporting collaborations between team members. To start working with Flux you have to create a project and share it with your team members. After inviting your team members, you can download the Flux plug-in for the applications you use. By selecting the data you want to share and send it to Flux, now the data will be live and there will be a live connection between you and your team members. The tools we use nowadays in a design process are powerful, but they do not work always together. Pressures time is wasted converting files between formats and sometimes critical data get lost along the way. Flux changes all that, allowing everyone to stay and sink. To give an example of this real-time collaboration, if the architect changes his column layout in one application and the structural engineer update the column sizes in another, flux will automatically translate, save and share those changes with everyone making it easier than ever to response on changing requirements and keep the project moving forward. Flux (2015) stated that you and your team now can focus more on creative work and less on creative workarounds. Your design tools working together, this is Flux. But is Flux inventing new BIM software? BIM is a building information model that involves the generation and management of digital representations. Its purpose is that we can work easier, be more efficient, design better and instead of planning in 2D they plan in 3D. Till so far the aim of Flux and BIM looks similar. With BIM is it also possible to share all the information to the different participants in the project where each participant such as architects, engineers and services managers gets and share all their work in a connected model. When someone change or update something in the project, is it changed in all the files also from the other participants similar to Flux.

Flux and BIM are comparable if it comes to share models. Both platforms have the ability to work in the same model with different participants and the files will update automatically. The only differences is that BIM uses one software where everyone has to work in and by using Flux it doesn’t matter in which design software you work. It could be in Sketchup when someone else is working in Revit. But next the the tool of collaboration Flux invested in more divisions. Flux is building two classes of tools: the first class connects existing tools together to allow seamless execution of complex workflows, and the second class captures design intent. For instance, the rules to lay out exit stairs are fairly easy to explain. When converted to software, our system can apply those rules to generate a new design based on a project’s unique requirements. Moving design logic into software promotes reuse, knowledge sharing, and continuous improvement. This is the foundational concept needed for our industry to achieve scale. (Watkins, K. Nov. 21, 2014)

2.2.2 The working of the site extractor
Beside the collaboration tool Flux developed the site extractor. With this tool you can find all the urban fact you need about the site where you project is situated. Type in the address of the site and the site extractor will take you there and with the movement of a square you can get the provided information from a really specific place. The extractor is extracting a number of components like, building footprints, building models and height data, generate other buildings, topography, contour lines, roads, parks and water features. From this listed menu you can pull down where you are interested in. When you download the plugin for flux you can pull this down into sketchup, rhino, revit and start designing.  This plugin is live so when you refresh your fill it will automatically update new data. Flux is getting their terrain data from United States Geological Survey (USGS) Executives, NASA. (Flux.io, Mar. 24, 2017)

2.2.3 The working of the modeling interface

For the approach from flux, to speed up the design process, build eco friendly and effective, Flux stopped designing individual buildings but started building seeds (Carlile, J. 2015). Building that have characteristic forms without making them separately. In this way we can design 10.000 of 100.000 building in the time we design one. The interface knows all the rules a building needs, its structural system, facades and its eichfac. It goes so far that it know that a building needs external sheets on the south facade to block afternoon sun. The north side do not need this shading. All those codes are build in, in this software. It knows how to design or grow his floor plans and how to connect them, so that you do not have a space no one can get to. There is also a knowledge about the grow of the structural, mechanical and connecting systems. Not only the role of the architect is being reduced also the engineering and constructors are less important. It know how to design staircases, bathrooms, pipe systems. The only thing that we as architects can do is move the different buildings or grow the floor plans. The automatic designs fits the plot and will grow is size regarding to the surrounding buildings. Jen Carlile (2015), Google Flux founder compared this process with a tree. You plan a seed and a tree will grow in that spot. If you plan another seed form the same tree specie in another spot the same sort of tree will come up but fitting in that area shown through a different shape and height. The impact that Flux have on architecture in this state is that the process now is a little more like genetic engineering with seeds. They are working on it, but now it is still a kind of genetic architecture (Carlile, J. 2015) .  

2.3 The artificial intelligence application of Flux

Flux is shut down at the moment, it will be transitioning to a new, as-of-yet-unknown business model. The new platform will provide more architectural approaches and will get rid of the genetic architecture they are creating now (Carlile, J. 2015). Together with AI, complex buildings can be instantly organized and architects only have to add some of the finishing touches (Beqiri, R. May 4, 2016). In Flux, AI makes it possible for architects to simulate the environment. With the open-source plugins for design software we can do all the necessary building and environmental analysis, without ever having to leave our computers. If we take a new project in a neighbourhood we do not know, the first thing you want to do is understand the context of your site, the neighboring buildings, roads and parks. This is the start to get real knowledge of the site. For this aim we can use the AI approach of Fluxs site extractor. In this AI approach there is the ability to share new fact and rules between participants in the collaboration platform. AI filters the excess information out of the system to provide the more valuable and requireble information for the design process. Up to here Flux is becoming a really good tool with the use of AI, but how are we gonna provide the quality of the data in the AI generator? Everyone has access to this tool and it seems that AI features never have been more accessible than they are today (Ismail, N. April 23, 2018). All the facts and rules in architecture are similar for everyone so it is easy for AI to cut out the waste for the design process but the concern lays in the subjective side of architecture, the creative approach. AI works best when large amounts of big data are available. The more facets the data covers, the faster the algorithms can learn and fine tune their predictive analyses, so successful machine learning depends on large and broad data sets (Ismail, N. April 23, 2018). Because it is important to have large amounts of big data, the quality is really significant. Data quality is the foundation on which all automated, intelligent applications are built. Without quality data, there’s no basis for intelligent applications to even begin to understand what constitutes good or bad behavior (Fernandez, D. Feb. 21, 2018). As earlier stated, almost everyone has access to use this tool and therefore the people themself have a reciprocal role in data quality. If we give them the right tools, they can improve the quality of data, and in return, that data improves their decision making.

3. The future of the architectural design process

If we think about where Flux stands at the moment and take a forecast to the future, what could be interesting to look at in the field of automation in the design process? Until now Flux generated genetic architecture and it lacks architectural value. It creates building results and that is something totally different than an architectural results. How can we get from a building result in Flux, back to an architectural result of our designs?

3.1 Artificial intelligence as part of the architectural design process  

For our believe artificial intelligence could help us to make the automated design process more creative. As earlier noted AI filters the excess information in a database and can learn from a database of many models to detect automatically good and bad patterns (Hebron, P. April 26, 2017). Also by knowing what we like, post online, buy or where we go, our smartphones, tablets and computers, though complex algorithms know us emphatically better than we do ourselves (Beqiri, R. May 4, 2016). Flux is developed by Google and general known Google is also one the biggest virtual environments that put billions of dollar into the aim the discover what we as a user like and are interested in. The collaboration of AI combined with the access to people’s information, could present a great possibility in architecture. Our smartphones already knows everything from us, from our interest to friend, what we like and what we dislike. If we take this to architecture, AI can not only be used for the calculation of postulation growth, prioritize projects, categorize streets, roads and parks and knows where important building, trees and height boundaries are but now I has also the possibility to combine those facts with our personal interests. The only thing to get the best represent architecture for the client is coming from this fact what the interests, likes and dislikes are from the customer.

Everyone have an innate sense of aesthetics and design. So to think about one's interest, likes and dislikes when it comes to architecture should not be difficult. There is a current feel for what is pleasing or useful, but there is always a lack of vocabulary and experience when it comes to the design process with the client. If we were to stop a random person on the street, give him a blank piece of paper and ask him to design his ideal living room, many people would not know where to start. By giving the person access to Pinterest, for instance, and ask them the same question again, they will by picking and choosing element they like designing their own living room (Hebron, P. April 26, 2017). This could be a powerful mechanism for driving interactions with design tools. If we glance to the working of pinterest, creating boards for different categories and interests, by selecting your own elements you like, AI could use this platform to pick components for your design. A database with selected architectural elements by itself helps the architect by their decision making. In this way the architect get a better understanding of the client and by small changes a design will be finish faster, easier and more efficient, the aim of Flux. Another way of looking to this database is from the perspective of the architect. The architect could make a database that needs millions of pictures for the AI to look at, it could be a kind of dictionary of architecture varieties of styles, architectural elements or materials where architects can agree on. With this platform the client or architect itself can choose his interests and the computer could generate a design with that specific wish.

Another interesting already existing tool is the 20Q. 20Q is an electronic version of the game twenty questions. It asks the user to think of an object and then through a serie if multiply questions the game discover what the user had in mind. Coming back to the gap between and architect and a client is the knowledge and experience. But by generate a couple of genetic questions the gap could be avoid. Research shows that 20Q guesses the correct object 80% of the time after twenty questions and 98% of the time after twenty five questions (Hebron, P. April 26, 2017). If we look to the benefits of this tool according to architecture, similar to the Pinterest tool, we as an architect could get a better understanding of the interest, likes and dislikes of the client. This tool could help the architect to delivered concise statements by creating interfaces and workflow that lead the customer through a series of simple questions or decision points about the design. Not only the architects now knows more about the wishes of the client, if we use AI to remember all those facts, wishes and outcomes of the questions, this mechanism can be extended to a design interface.

There are already a lot of little digital tools for other purposes that can be used in architecture. By small changes, generic architecture could change into real architecture made by the access of people’s interests. By using AI as a software that learns from a big database like Pinterest that has the interests of the customer we can create an interface and workflow that get rid of the lack of vocabulary knowledge and experience a new way about putting creativity in a automated design process arise. But where is the human intelligence of the architect in the design process, is he still needed or could the architect become an interface like a machine?

3.2 Human intelligence as part of the architectural design process

Natural intelligence of humans works through a network of circuits generated in our brain when we learn and acquire knowledge. Once learned, when performing an action the necessary networks are activated in order to proceed. Furthermore, the human brain generates ideas that are created through the associations of tracings and patterns of these connections that exist in our brain that move in synchronicity when we concern an idea.

Machine intelligence works through cognitive functions that are associated with learning and solving problems by creating a knowledge network storing data in the machine. In this way it is able to do things like, reasoning, perceiving, natural language processing, moving objects, manipulating objects, representing knowledge, doing this by following the programmed data step by step. Machine intelligence processes have a higher speed than the human, it is always operative and is more accurate.

What differentiates the architect from the machine, is that the machine can not generate ideas, the machine will always be programmed. Therefore, human intelligence, unlike the machine, is capable of learning many skills throughout its life, but the machine is designed for certain actions. On the other hand the human being doesn’t think and solve problems through a step by step process. The human intelligence judges and uses intuition to solve problems, a thing the machines are not capable of. But imagine if we manage to let a machine think in the way we do, the machine will always be programmed and there is not why to put intuition in a programmed language. For the reason, the most important resource that an architect can give is their mind, their way of thinking, intuition and creativity. That is why they get paid, not because of the hours they spend on a project or the size of it. If two architects get the same assessment, sources, conditions and site, the projects will always be different from each other. This is what makes architecture, architecture.



4. Discussion

Patrick Hebron wrote about design tools and programming languages. He stated that these tools put a lot of power in the hands of an architect. But the power is not really ours, because AI is controlling the design tools. It could only be from us unless we know how to use it. Design tools should not help us to execute a design in a domain we already understand, they should also help us to build expertise in new design domains, something of your own and not something everyone can make. Every designer knows that the hardest thing about designing is reconciling many component decisions to one another in order to produce a cohesive whole. It is not about making individual decision, but combine all the decisions to an entire design. Hebron compares it with a rubik’s cube, we can not solve one side of the cube and then move onto the next. We must solve all sides simultaneously. And then it comes to AI and the programmed languages in the design tools. One of the most useful things about design tools is to help the designer through this process. A lot of people seem to be worried that artificial intelligence will take our jobs and render us useless. But to get there, we need to remind ourselves what tools are for. In other words, the designer has to lead the tool, and not the other way around (Hebron, P. April 26, 2017).

University of Oxford researchers Carl Frey and Michael Osborne have estimated that architects are one of the least likely professions to be automated in the next 20 years. Architects, they say, must be able to negotiate and innovate, two skills that computers struggle with (Davis, D. June 15, 2015). While the products from Autodesk and Flux have shown what the abilities are for a know automated design process, they take a narrow view of architecture that fails to address these more complicated, more human skills.

So we have to create machines that we control and work for us. An automated design process that works by itself and does not need our control is great for a process such as the production of an object repetitively or for a function like lighting a light every time that someone passes through a certain door. When we take the machine to a process of creating, designing and innovating, the architecture gets lost. If a machine with an artificial intelligence was capable of designing buildings on its own, analyzing the precedents and the site of the project, this would be a genetic production of buildings like Flux is developed for. As it is presented in Flux, the machine analyzes the precedents and site, and through plant a seed in a specific spot the computer will design the building. These designs are genetic and do not response to us. By making the design process completely automatic, qualities from an architect will disappear, only quantify can become automated. The designs will lack of innovation and creativity and will be shaped in the most basic ways.  Also with the use of AI as a database with thousands of images and series of rules, the computer will always make the same design.

Where should artificial intelligence go in the world of architecture if it is not able to generate ideas? One purpose in a distant future could be to make a combination of human intelligence with artificial intelligence. As architecture can have the precision, speed and infinite knowledge of a machine and the creativity and innovation of the human brain, a good design could be made. Another one, in the near future, could be the presented idea of Flux, where we have all the precedent knowledge integrated in the machine that it is also able to analyse and understand the site and conditions of a project. We can combine all this data with our ideas and innovation using a seed that we generate for the building design and with small innovation we can complete the design as an architectural result.


5. Conclusion

Literature research is done to define automation in the design process. To define the automation, Flux, a development of Google X, has been chosen to investigate thoroughly. Dividing Flux in three different divisions, the innovation of Flux in the design process becomes clear. The design process consists of different phases to come to a building design. Research shows that Flux is not yet ready to be used in the real world of architecture. It is still struggling with the aim to make good architecture through automation.  

The automation in an architectural design is significant, parts of the design process could be done more efficient, faster and more specific than the human is capable of. But it seems that in some ways the technological possibilities become more important than the architectural value of a design nowadays. The focus is more on the qualities of the building, rather than the influence of the design to the world.  

As stated earlier, computers can quantify walkability and glean zoning regulations, manage multidisciplinary teams and can do day to day tasks. Kaufmann (November 21, 2017) argued in an interview with A+U magazine, “we become architects because we love to design and we thrive when we are creating environments that are inspiring, beautiful and sustainable. She stated also that we end up spending most of our time doing mundane work such as data collection, code reviews, counting, emailing and repetitive work that has previously been done.” Using software like flux this activities could be avoid or taken over. Sharing general knowledge and fact will definitely improve the speed of the design process. There is less focus on researching, analysing and putting energy in things that is already done, this results in more focus on the esthetics of a building design.

Today it is important that we know how far the design process could be automated. Like cyborgs as stated in the introduction, only our body is replaceable, our ghost will always be ours. It is similar in the design process, this process can not be automated when it comes to innovation, creativity and intuition. The essence of the architect can be found in his ideas and that is what makes the architect different from a machine. The ideas come from a place of imagination, which can not be automated and manufactured in mass, it is a unique quality and that is why an architect can never be replaced.


Notes
1. Moonshot means to achieve innovative and almost impossible subject, coming from its original meaning of a rocket launch to moon (research lab).
2. AI: Artificial intelligence, an intelligence demonstrated by machines, in contrast to the natural intelligence displayed by humans. It has functions that humans associate with other human minds, such as "learning" and "problem solving”.
3. BIM: Building information Modeling, a digital representation of physical and functional characteristics of a facility. A BIM is a shared knowledge resource for information about a facility forming a reliable basis for decisions during its life-cycle; defined as existing from earliest conception to demolition.

Literature


Davis, D. (June 15, 2015). Why Architects Can't Be Automated. Consulted on August 2, 2018 from https://www.architectmagazine.com/technology/why-architects-cant-be-automated_o

Hilburg, J. (Feb 26, 2018). Flux to shut down its data-sharing software and shift direction. Consulted on August 2, 2018 from https://archpaper.com/2018/02/flux-to-shut-down/

Schwartz, A. (June 17, 2016). A startup from Google's secretive moonshot lab is changing the way we build cities. Consulted on August 5, 2018 from https://www.businessinsider.com.au/inside-flux-a-
startup-from-googles-secretive-moonshot-lab-2016-6?r=US&IR=T

Pratiksha, R. (January 17, 2017). A detailed study of Flux: the react.js application architecture. Consulted on August 5, 2018 from https://www.cabotsolutions.com/2017/01/detailed-study-\
flux-react-js-application-architecture

Carlile, J. (2014). Google Flux Video. Consulted on August 2, 2018 from http://themidnightlunch.com/

Beqiri, R. (May 4, 2016). A.I. Architecture Intelligence, Architecture and Urban Planning in the age of Artificial Intelligence. Consulted on August 20, 2018 from http://futurearchitectureplatform.org/news/

Hebron, P. (April 26, 2017). Rethinking Design Tools in the Age of Machine Learning. Consulted on august 20, 2018 from https://medium.com/artists-and-machine-intelligence/rethinking-design-

in-architecture-what-are-the-practical-applications/364#ai-archi

Ismail, N. (April 23, 2018). The success of artificial intelligence depends on data. Consulted on August 31, 2018 from https://www.information-age.com/success-artificial-intelligence-data-123471607/

Fernandez, D. (Feb. 21, 2018). Data quality is the fuel that makes AI run. Consulted on August 31, 2018 from https://www.complianceweek.com/news/news-article/data-quality-is-the-fuel-that-makes-

Flux.io (Dec. 9, 2015). Working of the collaboration tool. Consulted on August 6, 2018 from https://www.youtube.com/watch?v=kc2ykDEeO6E

Flux.io (March 24, 2017). Working of the site extractor. Consulted on August 6, 2018 from  https://www.youtube.com/watch?v=tWiTO7n9r8U

Higgins, C. (May 30, 2015). Working of the modelling interface. Consulted on August 6, 2018 from https://www.youtube.com/watch?v=807Ys5udmVU

Buckley, A. (March 13, 2017).The Google Doc-ification Of BIM Workflows. Consulted on August 24, 2018 from https://www.rtcevents.com/blog/flux-io-the-google-doc-ification-of-bim-workflows/

Watkins, K. (Nov. 21, 2014). A+U Interviews Co-Founders of Google[x] Startup, Flux. Consulted on August 12, 2018 from https://www.archdaily.com/569283/a-u-interviews-co-founders-of-google-x-

Bourne, W. How One Architect Hatched the First Startup From Google's Top Secret Lab Consulted on August 8, 2018 from https://www.inc.com/magazine/201510/will-bourne/making-architecture-

BUILTR. (Dec. 21, 2015). Flux launches - a new age of sustainable architecture a scale ha begun. Consulted on August 12, 2018 from http://www.builtr.io/flux-launches-a-new-age-of-sustainable-

Dezhic, E. (June 14, 2017). Artificial vs Natural Intelligence. Consulted on Sept. 2, 2018 from https://becominghuman.ai/artificial-vs-natural-intelligence-626b6c7addb2


Kumar, S. (Jan. 30, 2018). How Artificial Intelligence And Natural Intelligence Works. Consulted on Sept. 2, 2018 from https://www.c-sharpcorner.com/article/how-artificial-intelligence-and-natural-

Stafford, T. (Sept. 24, 2014). Cómo se forman las ideas. Consulted on Sept. 2, 2018 from https://www.bbc.com/mundo/noticias/2014/09/140903_vert_fut_ciencia_pensamiento_lenguaje_neuronas_np
Posted by at
Subscribe to: Post Comments (Atom)