Human vs. Machine: Striking the Balance With AI in Architectural Design
CategoriesArchitecture

Human vs. Machine: Striking the Balance With AI in Architectural Design

Architizer’s Tech Directory is a database of tech tools for architects — from the latest generative design and AI to rendering and visualization, 3D modeling, project management and many more. Explore the complete library of categories here.

‘Will artificial intelligence (AI) replace architects?’

This question has been at the forefront of several discussions that have been vibrantly presenting many dystopic scenarios where big data, algorithms and machine learning will eventually replace human creativity and intuition. However, AI technology existed long before the AI bloom, rarely causing excessive panic within creative fields. So why is AI technology suddenly so threatening to the architectural profession?

Looking back at science fiction movies, AI was oftentimes presented as a human avatar that could accomplish assignments, make calculations and provide insightful information in unprecedented speeds. Nevertheless, it never performed extremely creative tasks. It was always depicted as a trusted artificial “advisor” capable of processing large amounts of data and automate complex procedures.

When platforms such as Midjourney, DALL-E and Stable Diffusion were introduced to the world, the creative skills and intuition necessary to visualize spaces and settings were abruptly redefined. Consequently, AI technology became a tool for generating impressive compositions and rapid design prototyping through the use of simple text prompts, eliminating the need for manual drawings or an overall architectural knowledge.

At the same time, a different set of tools was developed. AI apps such as qbiq and Laiout have the ability to process data, thus optimizing building sustainability and offer design solutions that are in accordance with local building codes. This type of AI software follows the more familiar pattern of speedy data processing, resulting in designs based purely on logical thinking. For the architectural profession, making this distinction between “creative” and “logical” AI algorithms is crucial.

Architizer’s new Tech Directory aggregates tech tools for architects, allowing you to search, compare and review AI softwares before selecting which to you in your next project:

Explore Architizer’s Tech Directory

On the one hand, even though “logical” AI tools are able to generate countless design iterations based on predetermined parameters, they lack the nuanced understanding of cultural context, emotional resonance, and human experience. There’s also a risk of homogenization, where buildings designed by AI algorithms may lack the diversity and richness that stem from individual human perspectives and cultural influences. Another challenge lies in the inherent biases embedded within AI algorithms.

Machine learning models are trained on historical data, which may reflect societal biases and perpetuate inequalities in architectural design. Without careful oversight and intervention, AI algorithms could inadvertently reinforce existing biases related to race, gender, socio-economic status, and accessibility, leading to inequitable outcomes in the built environment. Admittedly, the risks associated with data-driven AI technology and architectural design are well-known and somewhat obvious. In such a subjective discipline, logic alone is not enough to create spaces that are truly regarded as pieces of architecture. Design is partially an instinctive process that is unique for each individual architect and cannot be replicated by any intelligent machine.

“Creative” AI algorithms, however, pose a different type of challenge since they bring an ‘unpredictability factor’ to the table. Instead of relying purely on data, text-to-image generators produce designs based on subjective (human generated) prompts. Instead of spending numerous hours sketching, drawing and model making, “creative” AI apps can generate countless concepts in a matter of seconds by simply feeding them client briefs. Nevertheless, while they seem ideal for producing impressive concepts for early stage design, the images they create are fairly fictional. Ironically, they lack data.

Even during the initial brainstorming phase, architects do consider an array of objective as well as subjective information regarding their project. From climatological site conditions to potential present or future occupants and even immaterial data such as past histories, myths or cultural norms. This framework of complex interrelationships becomes the blueprint for designing for specific sites, people and cultures, adopting a holistic approach towards proposing creative and innovative solutions.

There is no doubt that both “logical” and “creative” AI algorithms have become ‘trusted advisors’ for many architects. “Creative” AI tools are ideal for showing design intention, representation and visualization. In other words, they bridge the communication gap between architects and clients. In parallel, “logical” AI tools can rapidly access and process information such as planning codes or material specifications and energy reports. During the technical and construction stages, many AI apps can automate processes and produce standardized drawings seamlessly and with minimal effort.

In short, both types of AI tools are two sides of the same coin, showing both promise and peril. Still, as long as architects are aware of AI limitations, integrating the technology into their workflow can provide valuable insights and inspiration, inform decision-making and ultimately enhance the practice of architecture. Architects have always been the mediators between logic and creativity, with AI being no exception. This paradigm shift presents an opportunity as well as a challenge, which — if navigated correctly — will enable architects to focus on what they do best: creating innovative, human-centered and culturally significant spaces.

Architizer’s Tech Directory is a database of tech tools for architects — from the latest generative design and AI to rendering and visualization, 3D modeling, project management and many more. Explore the complete library of categories here.

Reference

Managing warehouses with machine learning
CategoriesSustainable News

Managing warehouses with machine learning

Spotted: Although many people returned to in-person shopping after the COVID-19 pandemic, e-commerce has continued to grow and is expected to make up nearly 20 per cent of all retail sales in 2023. At the same time, the need for efficient logistics is growing, with the market for warehouse management systems poised to reach $12.3 billion (around €11.2 billion) by 2031. 

One new player in this field is Fulfilld, which has developed an innovative platform that uses real-time data to optimise warehouse operations by coordinating tasks between humans and robots. Fulfilld’s platform harnesses cloud-based technology, ultra-wideband connectivity, RTLS beacons, scanners, tags, and digital-twin warehouse simulations to connect systems and track real-time flow. 

The system includes both software and hardware, in the form of hand-held scanners with natural language processing capabilities, to optimise inventory and co-ordinate task instructions to both human and robotic workers. The startup’s artificial intelligence (AI) and machine learning models can also proactively recommend “on-the-fly opportunities” for better warehouse optimisation. 

The system further co-ordinates tasks and inventory locations and creates a Google Maps-like solution for warehouse workers. 

Fulfilld aims to serve mid-market customers in industries including warehousing, manufacturing, distribution, and logistics. It offers a subscription package that allows for rapid roll-out. The novel technology, the startup claims, can boost efficiency by 15-20 per cent, and reduce employee turnover – a crucial benefit amid labour shortages and supply chain disruptions. 

Many companies are calling on robots to help optimise operations across various sectors, including to build houses and support security guards.

Written By: Lisa Magloff

Reference

A rapid cleaning machine for reusable glasses 
CategoriesSustainable News

A rapid cleaning machine for reusable glasses 

Spotted: Europe’s top three coffee-consuming countries drink more than 1,000 cups per person every year. From free reusable crockery on German trains to additional fees charged to customers wanting to use single-use food packaging in the Netherlands, nations around the world are trialling ways to cut down on plastic pollution, particularly in the takeaway food and drink industry. 

Many companies encourage the use and washing of mugs and bowls in the office, and with a goal of reducing the water used to provide such a service, French company Auum has created a new method of cleaning. The Auum-S single glass dishwasher is made from a patented design that disinfects reusable cups in 10 seconds, allowing one washer to clean up to 2,000 glasses per day. The company is building a circular economy with French-made parts and assembly, along with a rental service that ensures that machines are properly maintained, and glasses recycled at the end of their life. 

Auum cups are made from double-walled borosilicate by Swiss glassmakers Bodum and the glasses can be custom branded. The glasses are lightweight and cool to the touch, even after cleaning. The countertop washer cleans a single glass using 140-degree Celsius dry steam and less than a single ounce of water per wash. The high temperature negates the need for chemicals.  

The low rate of water use makes the washer much more economical, both financially and environmentally, than using biodegradable or paper cups or running a traditional dishwasher. In a company with 150 employees, using the Auum-S washer saves up 130,000 litres of water each year and eliminates close to 100,000 disposable cups. 

Countertop appliances are getting smarter and smaller, with Springwise spotting a no-smell compost device as well as a homebrewing solution for plant-based milk.

Written By: Keely Khoury

Reference

Combining machine learning and ancestral wisdom to uncover plant-based food ingredients 
CategoriesSustainable News

Combining machine learning and ancestral wisdom to uncover plant-based food ingredients 

Spotted: According to the UN, the Earth’s population will likely reach 8.5 billion by 2030. At the same time, climate change is going to make it more difficult to grow food, requiring a rapid and collaborative approach to the global food industry. For startup, The Live Green Company, the answer can be found in plants. The company has developed a way to use biotechnology and machine learning to replace animal, synthetic, and ultra-processed foods with precise plant-based alternatives.  

Live Green’s platform, dubbed Charaka, uses machine learning to analyse data about thousands of plants and find appropriate plant substitutes for animal-based and artificial ingredients. Charaka’s algorithms analyse complex data about the phytochemical compounds, bioactive molecules, and nutritional profiles of various plants. The company claims that the platform can “uncover hidden and non-linear relationships and predict innovative functionalities and uses” of different ingredients to find a perfect plant-based substitute.  

Developing these substitutes involves creating blends of natural plant ingredients like sunflower protein, banana, and flax meal without changing the taste, texture, or mouthfeel of the finished product. In addition, Live Green’s platform also identifies more sustainable local alternatives to vegetarian ingredients like avocado. From the idea stage to putting a new all-plant product on the shelf can take as little as 90 days. 

Live Green has thus far piloted several product lines – including burger mixes, baking mixes, frozen burgers, ice-creams, and protein bars – that are plant-based and free of additives, allergens, gluten, cholesterol, and trans fats.  

Other recent food and drink innovations spotted by Springwise include fungal fermentation for natural food colourings, protein and umami extracted from cabbages, and microbial protein for people with modified diets.

Written By: Lisa Magloff

Reference

Using machine learning to map worldwide waste
CategoriesSustainable News

Using machine learning to map worldwide waste

Spotted: Every year we dump a colossal 2.12 billion tonnes of waste. And if we filled trucks with this rubbish, there would be vans to go around the globe 24 times. Although our rubbish clearly isn’t stored just in trucks, do we know where it actually goes? Edinburgh-based data analytics firm Topolytics is looking for answers with its data aggregation and analytics platform that aims to make the world’s waste visible, verifiable, and valuable.

The platform, called WasteMap, collects and analyses data about waste types, amounts, and movements to create insights for waste producers, investors, and governments across numerous sites, regions, and countries. With this, Topolytics envisions waste and management resources becoming more transparent and effective for both commercial use and the environment.

Michael Groves, Topolytics chief executive, explains that, “it helps waste producers and recyclers to drive resource and cost efficiencies, to trace the movement of waste and measure their impact, whilst validating performance and improving the quality and reliability of ESG and carbon reporting.”

The Scottish firm has reportedly raised £1.5 million (around €1.69 million) from “seasoned hight-net-worth-investors” and a UK Research and Innovation grant. With this money, the company claims it will launch its WasteMap solution and further commercial business.

Springwise has previously spotted other intelligent technologies that help manage the waste we make, including an AI-powered litter-picking vehicle and an AI-handled waste-sorting process.

Written By: Georgia King

Reference

Computer vision, automation, and machine learning boost insect farming
CategoriesSustainable News

Computer vision, automation, and machine learning boost insect farming

Spotted: Experts are becoming increasingly concerned about how the world’s growing population will be fed in an equitable and sustainable way. One solution is edible insects – both for human consumption, and as pet food in order to help free up land and resources. Insects require far less space and fewer resources to farm than other proteins like beef or chicken, but producing them en masse has proved challenging so far. Tech company Entocycle is using innovative technology to help. 

The London-based startup uses smart technology to help insect farms work efficiently and sustainably (and manage billions of insects at any one time). Its technology aims to help farms improve accuracy, efficiency, and enable less need for manual involvement, such as by measuring populations in a farm to automate food requirements and controlling the temperature to optimise insect health.

The company focuses on black soldier fly farms, an insect that grows very rapidly – and can survive on food waste. They contain all the nutrients humans need for good health, including more zinc and iron than lean meat, and more calcium than milk.  

Entocycle recently raised $5 million (around €4.7 million) in a recent Series A funding round, which the startup will use to expand the commercial roll-out of its products and services. 

Entocycle is not alone in developing technology to help make insect-growing a viable and sustainable operation. Springwise has also spotted vertical mealworm farms that produce plant and animal feed, and AI-powered insect microfarms.

Written By: Jessica Bradley

Reference

Subtractive Manufacturing: The Quiet Machine Revolution that is Reshaping Architecture
CategoriesArchitecture

Subtractive Manufacturing: The Quiet Machine Revolution that is Reshaping Architecture

Browse the Architizer Jobs Board and apply for architecture and design positions at some of the world’s best firms. Click here to sign up for our Jobs Newsletter.

Architecture is defined by materials and construction. In terms of building envelopes, structures, and systems, there are two primary manufacturing methods used. Additive manufacturing processes add material layer-by-layer, while subtractive manufacturing removes material to create building components or parts. Additive manufacturing, usually under the umbrella of 3D printing, is increasingly well-known, but subtractive methods have been used throughout history. In modern architecture, designers, fabricators, and manufacturers are reimagining the possibilities of subtractive manufacturing to create new forms and spaces for contemporary life.

Subtractive manufacturing is an umbrella term for different kinds of controlled machining and material removal processes. These are either performed manually or more commonly, driven by computer numerical control (CNC). Subtractive methods can start with solid blocks or material forms made from metals, plastics, wood, or other materials that are shaped through boring, cutting, drilling, and grinding. This machining process spans from CNC methods to electrical discharge machining in hard metals, as well as laser cutting (for thermoplastics, wood, acrylic, fabrics, metals) and water jet cutting for plastics, hard and soft metals, stone, glass and composites.

Mold Tool

A two-part mold tool from MDF being machined with a 1/16″ flat end mill. Courtesy Portland CNC

The key advantage to subtractive manufacturing is that it’s ideal for tight tolerances and geometries that are difficult to mold, cast, or produce with traditional machining. Larger, less complex objects also tend to lend themselves more to subtractive manufacturing. This machining process is also most often the choice for fabricating finished parts. Subtractive manufacturing processes can create parts for prototyping, manufacturing tooling, and end-use parts. The following projects showcase how subtractive manufacturing processes are leading innovations in design and construction, and what the future of these methods might hold.


By Bernard Tschumi Architects, Tianjin, China

 

The Binhai Science Museum was designed to showcase artifacts from Tianjin’s industrial past through large-scale contemporary technology, including spectacular rockets for space research. The project is part of the city’s Binhai Cultural Center and contains facilities for cultural events and exhibitions as well as galleries, offices, and restaurant and retail spaces. The project was made to relate to the rich industrial history of the area, the site of high-volume manufacturing and research. A series of large-scale cones create major rooms throughout the museum. The central cone, lit from above, connects all three levels of the building.

The exterior of the museum is covered with aluminum rain screen over a sealed aluminum surface, giving the building a unified presence despite its large size and the disparate elements of the structure. Approximately 3,600 copper-colored panels in two sizes (4×7 ft and 4×11.5 ft) make up the flat portions of the building’s façade. The perforated metal panels also help reduce heat gain. The design team developed 52 different sizes of panels with each row of the cones corresponding to a unique width. Each panel is backed by two aluminum U-channels located between the perforations.


By Acton Ostry Architects, Vancouver, Canada

Completed in 2017, Brock Commons Tallwood House is an 18-story, LEED Gold certified, 404-bed student residence building located at the University of British Columbia in Vancouver, BC. It was the world’s tallest mass wood tower at the time of its completion. The timber structure and prefabricated facade also went up in only 66 days.

Key to receiving approvals and achieving economic viability was a “keep it simple” design approach that makes the building appear ordinary through encapsulation of the wood structure with gypsum board. The prefabricated facade, arranged in a pattern of vertical striations, features pre-installed windows and cladding consisting of 70% wood fibers.

A CLT canopy runs the length of a curtain wall base, which reveals the warm wood finishes of amenity spaces within. Elevator lobbies are clad with the same material as the exterior. Hallway finishes include natural wood doors and a palette of rich umber and ochre accent finishes. Living unit interiors are bright white, spare and simple.

CNC machines cut all the mass-timber components, including the penetrations in the CLT panels and the connection holes in both ends of the columns. A unique identifier was assigned to each mass-timber component for quality-assurance and quality-control tracking and on-site measurement of the structural system assembly heights.


By Morphosis Architects, New York, NY, United States

The Emma and Georgina Bloomberg Center is the academic hub of the new Cornell Tech campus on Roosevelt Island. With the goal of becoming a net zero building, The Bloomberg Center forms the heart of the campus, bridging academia and industry while pioneering new standards in environmental sustainability through state-of-the-art design. T

he Bloomberg Center is a four-story building set beneath a photovoltaic canopy, with a low and narrow profile framing stunning views across the island. One of the building’s most distinctive features is its façade, optimized to balance transparency —maximizing daylighting and exterior views, and opacity — while ensuring sufficient insulation and reducing thermal bridging.

Designed as a rain screen system, the outermost layer of the façade is composed of aluminum panels surfaced in an iridescent, PPG polymer coating. Designed in collaboration with Zahner, an architectural metal fabricator, the façade utilizes Zahner’s Louvered ZIRA™ system to create the image patterning.

Each pixel of the image is translated into the specific turn-and-tilt of a two-inch circular tab punched into the aluminum paneling; the depth and rotation of each tab determines the amount of light reflected. This pixel map was fed into a repurposed welding robot, which processed the digital information into the mechanical turning-and-tilting of the façade’s 337,500 tabs.


By Partisans, Georgian Bay, Canada

Perched on an island’s edge in Georgian Bay, Ontario, the Grotto Sauna is a feat of old-world craftsmanship and new world sustainability made possible by cutting-edge software and fabrication technology. The selected concept for the Grotto prescribed a solid, simple presence on the exterior, while the interior followed dynamic air movements in curvature forms; requiring design solutions.

As a result, the team proceeded to experiment further with the materials, and selected wood as the primary medium. Importantly, the Grotto established a successful methodology for addressing the challenges of building ambitious architecture in remote and environmentally sensitive regions.

Partisans collaborated directly with their fabrication partner, MCM Inc., to develop new prototyping methods and with engineers and develop novel software patches for the toolpaths. The latter enabled the fabricators to override the automated limitations of the CNC machinery and ultimately use it as a sculpting tool to achieve the aesthetic vision, all the while maximizing the available wood and milling along the grain so that the pieces would match one another. The successful production of the panels also had to anticipate the method by which they would be sequentially assembled. This required the team to develop a sophisticated installation plan in tandem with the fabrication process.


By Bjarke Ingels Group and JDS Architects, København, Denmark

When Bjarke Ingels Group and JDS Architects set out to create The Mountain residential project, it was the 2nd generation of the VM Houses: same client, same size and same street. The program, however, is 2/3 parking and 1/3 living. What if the parking area became the base upon which to place terraced housing, like a concrete hillside covered by a thin layer of housing cascading from the 11th floor to the street edge? Rather than doing two separate buildings next to each other — a parking and a housing block — the team decided to merge the two functions into a symbiotic relationship.

The parking floors are covered with a continuous perforated aluminum surface, folding into four parts from the southern to the eastern facades. This controls sunlight and air circulation. The folded surface makes reference to the name of the project, displaying a realistic image of Mount Everest.

For its aluminum panel facade, each perforation plays a specific part in the composition of the image. In some parts, perforations are so large and dense that there is almost no visual restriction between inside and outside. During the day, the perforation in the aluminum plates appears black on the bright aluminum, and the mountain image resembles a roughly rasterized photo. At night time, the facade is lit from the inside and appears as a photo negative in different colors as each floor in the parking area has different colors.

Courtesy Portland CNC


Today, subtractive manufacturing is being radically reimagined. We are seeing projects where 4D self-transforming materials respond to changes in heat, sound, or moisture levels to change shape. Subtractive manufacturing also offers a variety of material and processing methods for diverse design applications. As the process of building evolves, the convergence of technologies opens new possibilities for architecture, design and machining.

Browse the Architizer Jobs Board and apply for architecture and design positions at some of the world’s best firms. Click here to sign up for our Jobs Newsletter.

Reference