Tag: kuenstliche intelligenz

  • More AI does not solve data problems

    More AI does not solve data problems

    The mistake begins with the investment
    It’s a familiar pattern: a company recognizes the potential of artificial intelligence, looks at solutions, chooses a tool – and gets started. The expectation is that the new technology will somehow solve existing data problems. The reality: It doesn’t. It makes them more visible.

    This is no coincidence. It is the consequence of a wrong sequence.

    Data is collected – but not made usable
    Data is available in most real estate companies. Property data, tenant data, operating figures, maintenance histories – they exist. The problem is not its absence, but its condition. They are scattered across systems, inconsistently maintained, inconsistently defined or simply cannot be linked to one another. There are sometimes three different versions of the same key figure – in three different systems.

    Anyone who sets up an AI model under these conditions will not get any answers. What you get is output that reinforces existing uncertainties – automatically and at high speed. AI recognizes patterns in data. If the data is inconsistent, the model learns from the inconsistency. If it is incomplete, it operates on an incomplete basis.

    A new layer of complexity
    What is created in practice is not a gain in efficiency. It is a new layer of complexity: AI outputs that nobody trusts. Departments that manually check results. Projects that come to a standstill. A lot of effort, little effect, growing frustration.

    The fatal thing is that many companies react to this with the next tool upgrade. The cycle starts all over again.

    A data hub is not a tool – it is a structure
    The solution does not lie in better models. It lies in a structural decision: the creation of a common, harmonized database. A data hub is not another system that is added to the existing IT landscape. It is the opposite – it replaces fragmentation with central availability. It integrates distributed data sources, breaks down silos and inconsistencies and creates the basis for scalable AI applications and automated reporting.

    The decisive factor is not where the data is stored. What matters is how it can be used: uniformly defined, quality-assured, accessible for different use cases. Only on this basis can AI deliver what it promises.

    Data quality is not preliminary work – it is an ongoing task
    Even with a data hub, a central challenge remains: Data quality is not a one-off cleansing project before go-live. It is a continuous process. Anyone who sees data quality as a preliminary project will realize after the launch that the real problem is only just beginning.

    The database is supplemented by a data catalog: It transparently documents which data exists, where it comes from and how reliable it is. It creates a common language that connects specialist departments and technology – and gives control back to the organization.

    In the webinar: From the database to scalable AI
    In our free webinar “The optimal AI architecture: How data hub, data quality and data catalog make the difference”, we show how real estate companies can tackle this transformation in concrete terms – from data architecture and quality assurance to the productive use of AI. With practical insights, concrete solutions and time for your questions.

    Register now for free

  • Zurich tests AI in the building permit process

    Zurich tests AI in the building permit process

    From April 2027, the canton of Zurich will require all municipalities to use the eBaugesucheZH platform. This lays the foundation for digitization, but only the foundation. The content of the applications will continue to be processed in different systems, depending on the municipality or canton. This historically evolved system landscape leads to media disruptions, manual coordination rounds and data inconsistencies.

    What the FHNW study shows
    The Building Directorate commissioned the FHNW Institute of Digital Construction to conduct a potential study along the entire process chain. 15 fields of action were identified, from initial digital information to building acceptance. The greatest short-term potential lies at the very beginning. Chatbots for the initial consultation, structured submission support and automated preliminary checks could immediately improve the quality of submitted applications and significantly reduce queries. Many improvements can already be achieved with rule-based systems, without generative AI.

    Prototype with the city of Kloten
    The Innovation Sandbox for AI of the Office of Economic Affairs tested an AI-based preliminary check for the notification procedure together with practice and technology partners, including the city of Kloten. For simple projects such as solar installations or heat pumps, a rule-based system automatically clarifies the admissibility and choice of procedure, and an AI then checks the completeness and quality of the entries. 3336 tests were evaluated. The results are encouraging, even if the reliable interpretation of complex plan representations remains an open challenge.

    Humans remain responsible
    Both studies agree that complete automation is currently not realistic. Where decision-making logic is clearly defined, rule-based systems are preferable to generative AI. The authority to make decisions remains with humans. Legal issues relating to data protection, liability, transparency and copyrighted blueprints as AI training material must be examined in depth before any implementation.

    The results are now being incorporated into the further development of eBaugesucheZH. Individual applications are to be tested in pilot municipalities. Zurich is thus demonstrating how the careful, step-by-step use of AI can work in a complex administrative domain.

  • From data to AI in the real estate world

    From data to AI in the real estate world

    This is precisely why it is worth looking back. Because the way in which real estate is planned, operated and managed has changed fundamentally over the last 30 years.

    Thirty years ago, many processes were still surprisingly analog. Data was stored in folders and paper documents, decisions were based heavily on experience and less on systematic analysis. A phase soon began in which the industry developed step by step: processes became more digital, data more important, buildings and companies increasingly networked.

    It was in this environment that pom was founded in the mid-1990s as a spin-off from ETH Zurich – with the idea of integrating tasks, data and processes in the construction and real estate sector more closely. Thirty years later, pom is celebrating its anniversary and the basic question is still very topical: How can real estate, organization and technology be meaningfully combined?

    In terms of technology, we are now at a new turning point. The digitalization of real estate continues to advance: cloud technologies, IoT and digital models are enabling ever more precise mapping of buildings. The so-called digital twin is increasingly becoming a reality and creating new opportunities for automating processes.

    At the same time, the way companies work is changing. Artificial intelligence will change many processes in the coming years – especially where large amounts of information have to be processed and decisions still have to be made manually. Different data can be analyzed more easily, finished results can be generated automatically and decisions can be massively accelerated, even with the involvement of humans. Assistance systems, known as agents, are becoming part of everyday working life.

    At the same time, a look at the industry reveals an interesting area of tension: technological development is progressing rapidly, while implementation in companies is much slower.

    Every year since 2016, pom Consulting AG has measured the digital maturity of the construction and real estate industry as part of the Digital Real Estate & Construction Study. The Digital Real Estate Index currently stands at 4.3 out of 10 points – a slight recovery compared to the previous year, but definitely not a quantum leap.

    Unsurprisingly, artificial intelligence is increasingly coming into focus. According to the latest study, Artificial Intelligence & Machine Learning is once again one of the most frequently used technologies, alongside Platforms & Portals and Data Analytics. However, the assessment of AI is much more differentiated than in previous years: Around two thirds of respondents see a high benefit in it. In last year’s survey, the figure was 75%. With more frequent use of AI, the possibilities of the technology, but also its limitations, are becoming much more visible, making expectations more realistic.

    Technology alone therefore does not determine success. The decisive factor remains the organization: data quality, implementation strength, clear responsibilities – and the willingness to question existing ways of working.

    Perhaps this is the real parallel to the last 30 years.

    Back then, too, it wasn’t just about new technologies, but about new ways of thinking. Artificial intelligence could therefore become the next big development step in the industry – not because it changes everything, but because it helps to better manage the growing complexity of real estate and organizations.

  • Startup accelerates engineering simulations with AI

    Startup accelerates engineering simulations with AI

    Hardware development and material testing today rely heavily on physics-based simulations for design, validation and production. These calculations often take hours or days and incur high costs, which delays projects and pushes back production launches. Engineers therefore often reduce model complexity to shorten calculation times, at the expense of accuracy and proximity to real operating conditions.

    Physics-aware AI for faster workflows
    Fainite is developing a physics-aware AI platform that speeds up and simplifies existing simulation workflows. The engine learns from physics-based simulations and can derive accurate predictions without relying on large historical data sets. Engineers set up new workflows in minutes, run simulations much faster and can intelligently reuse previous results, even with limited amounts of data. An integrated AI agent guides them through complex steps, suggests settings and makes advanced analyses usable for broader teams.

    cHF 150,000 for scaling and market entry
    The CHF 150,000 from the Venture Kick programme will be used to expand the technology to additional engineering disciplines and use cases and to build a scalable platform with next-generation functionalities. At the same time, the funds will strengthen the team structure and go-to-market activities in order to accelerate deployment at industrial companies. The company is thus addressing around 9 million hardware engineers worldwide whose work is currently slowed down by slow, complex simulation processes.

    Founding team with physics and AI expertise
    The start-up was founded by researchers and engineers from Caltech, ETH Zurich, the University of Cambridge and Google, including CEO Alex Donzelli, Chief Scientist Prof Burigede Liu and ML Lead Matthias Bonvin. The team is complemented by former executives from established simulation software manufacturers, bringing together in-depth expertise in deep learning, computational physics and industrial simulation platforms. According to Alex Donzelli, Venture Kick’s funding, feedback and network have been instrumental in moving quickly from technical validation to the first industrial applications.

  • Why the real estate industry is tackling digitalisation

    Why the real estate industry is tackling digitalisation

    Mr Caspar, pom is regarded as one of the leading consulting companies in the areas of digitalisation, transformation and sustainable property development. How would you describe your role within this ecosystem?
    At pom, we see ourselves as an intermediary between research, development and practice in the property industry. Our role is to recognise new topics at an early stage, classify trends and develop an understanding of what will move the industry in the future. We translate this knowledge into concrete use cases, recommendations and a basis for decision-making for our clients. In doing so, we help companies to separate the important from the unimportant and to focus specifically on those methods, technologies and data that actually create added value for their role in the property industry.

    Which topics are your customers currently most concerned with: data, processes, organisation or technology?
    There is no simple answer to this question. In recent years, the focus has been very much on technology. Many companies have introduced new systems and launched numerous digitalisation projects. This has certainly brought progress, but has also led to a certain amount of disillusionment. Projects were more time-consuming, more expensive and more complex than expected. This was often due to the fact that data and processes were underestimated. We are currently observing a clear shift away from purely technology-driven projects towards more data- and process-orientated approaches. The current AI hype is further reinforcing this development.

    pom emphasises that data is the foundation of modern property management. Where do Swiss companies stand today in terms of data readiness?
    Basically, Swiss property companies are not in a bad position. Most of them have sufficient data readiness to operate their core processes reliably and answer relevant stakeholder questions, but we see a need for development in automation in particular. This requires data to be structured, consistent and available company-wide. Data-intensive topics such as ESG show where the limits lie. Another important point is collaboration across company boundaries. The property industry has always been highly networked. Greater data readiness is crucial to making this collaboration more efficient, digital and automated in the future.

    What are the most common misconceptions about the digitalisation of property portfolios?
    The effort and complexity are often underestimated. In particular, the provision and preparation of the required data is estimated too optimistically. Although data is available, it is often not of the necessary quality or structure. This leads to delays, additional costs and extra work for the specialist departments that should actually be doing their core business. Another misconception is that digitalisation projects can be implemented “on the side”. Professional project structures and the corresponding expertise are often lacking. This has a negative impact on motivation, acceptance and ultimately the success of the project.

    Which technological developments will change the property sector the most in the next 5-10 years?
    Basically, we distinguish between two levels: the digitalisation of the property itself and the digitalisation of the companies that operate these properties. At building level, we are seeing major advances in the cloud, IoT and digital models. The digital mapping of properties, often referred to as the digital twin, is increasingly becoming the standard and enables new forms of automation, while at company level, development will be strongly characterised by AI and process digitalisation. A small number of core applications, combined with flexible low-code platforms, will make it possible to automate processes efficiently and across companies.

    Many companies are experimenting with AI. Where do you see realistic fields of application in the next 24 months?
    In the short term, there is great potential in analysing and evaluating documents and unstructured data. Content can be summarised, evaluated and created more quickly. Another important step is the integration of AI tools into everyday working life, for example as assistance solutions. The next step will be to increasingly link these systems with company-specific data. Reporting and analyses will also change: instead of fixed reports, information will be compiled according to the situation and needs.

    What risks do you see in the use of AI in the property industry?
    We see the biggest challenges less in the regulatory area and more at a cultural and technological level. Many companies do not yet have the necessary skills and structures in place to deal with data and technology. In addition, the Swiss market is highly fragmented and heterogeneous, which makes it difficult to introduce standardised solutions. The property industry is project-orientated and has little serial logic. This further slows down the introduction of new technologies.

    Rate of adoption instead of technology: what are the biggest cultural hurdles?
    A key hurdle is that digitalisation is not yet anchored as a strategic topic in many companies. A lack of expertise, unclear responsibilities and the expectation of implementing digitalisation “on the side” are slowing down implementation. In addition, there is often a lack of willingness to consistently scrutinise and change existing ways of working.

    How is digitalisation changing roles in real estate companies?
    Digital skills will be part of the basic qualification of many roles in the future. The focus will be less on a deep understanding of technology and more on secure user knowledge. At the same time, new roles are emerging, for example for the management of digitalisation projects and digital platforms. These functions ensure that systems are used, developed and operated sensibly.
    This allows asset, property and facility managers to continue to focus on their core business.

    What makes a digitally mature company?
    A digitally mature company anchors digitalisation, technology and data at the highest management level. There is a clear strategic stance, defined goals and responsibilities. Digitalisation is not delegated to IT, but is seen as an entrepreneurial task.
    Such a company also has the necessary roles, processes and expertise to continuously develop digital solutions and adapt them to changing conditions.

    ESG and PropTech are growing together. Which technologies are already creating real impact?
    ESG is a strongly data-driven topic. There are now functioning solutions along the entire data chain, from measurement to key figures. Smart meters, automated analyses of energy bills and the consolidation of data across several buildings are technically feasible. The challenge lies less in individual components and more in end-to-end integration and automation within companies. We do not yet see a comprehensive all-in-one solution.

    Where do you see the biggest gaps between requirements and reality on the market?
    The biggest gaps arise where requirements are only implemented selectively without considering the entire value chain. Data-driven topics in particular show that technical possibilities are available, but organisational and structural requirements are often lacking.

    How do you assess the maturity of the Swiss PropTech market in an international comparison?
    Switzerland has a very lively and innovative PropTech scene. Many solutions are internationally successful. The biggest challenge lies in scalability due to the size of the market and the federal structures. Overall, however, the level of maturity is high and competitive.

    Which PropTech areas are underdeveloped and which are overheated?
    The ESG sector is currently very overheated. There are a large number of solutions, which leads to a certain disillusionment. In international comparison, the consistent use of BIM across the entire property life cycle is particularly underdeveloped. Other countries are further ahead here, especially when it comes to institutional investors.

    Where do you see potential for partnerships between established companies and start-ups?
    Partnerships offer great potential, but are challenging. Established companies think long-term, start-ups are dynamic and innovation-driven. Cooperations are successful where there is mutual understanding and clear expectations are defined, be it in projects, partnerships or targeted funding models.

    Which developments currently surprise you in particular, both positive and negative?
    The AI hype is both positive and challenging. Positive because it promotes innovation, efficiency and new ways of thinking. Negative because expectations are often overestimated in the short term. Sustainable success requires an in-depth examination of data, processes and governance.

    If you could change one thing in the industry immediately, what would it be?
    I would like to see more consistency across the entire property life cycle.
    The project-based, highly individualised way of working makes it difficult to use scalable digital solutions. Approaches such as prefabrication and standardised construction methods could help to enable technological leaps without losing quality and design freedom.

    What drives you personally to drive forward the transformation of the industry?
    I am motivated by change, new projects and the opportunity to develop things further.
    Property is a particularly exciting field because it shapes our daily lives, from living to working. I find shaping digitalisation and transformation in this context to be meaningful and highly relevant.

  • How artificial intelligence secures the construction process

    How artificial intelligence secures the construction process

    According to a press release, the Benetics AI email assistant is designed to help prevent one of the most costly sources of error in everyday construction work: incorrect execution due to outdated plans. This assistant was developed by Benetics AG. Founded in 2022, the Zurich-based software company will be presenting it as a world first at Swissbau in Basel from 20 January.

    Following the voice assistant for the skilled trades, also based on artificial intelligence (AI), which will be launched in 2024, “the AI email assistant is the second world first from Benetics AI,” says CEO Ferdinand Metzler. “This brings us another step closer to our vision: less time-consuming administrative work and more focus on what makes the skilled trades strong: productive work.”

    The email assistant recognises construction plan PDFs in incoming email attachments and compares them with existing plans in all current projects. It automatically notifies users when a new plan version has been received. And at the touch of a button, it ensures that no one is still working on the old plan version.

    The AI assistant is embedded directly in Microsoft Outlook and can be integrated from there into other systems such as SharePoint or other DMS and CDE solutions. According to the information provided, the open API of Benetics AI and new connectors on Make.com and Zapier enable flexible integrations. “This creates a continuous digital plan flow all the way to the fitter on the construction site for the first time,” the press release states.

    “The Outlook integration brings even more structure to the construction process,” says Bledar Beqiri. He is head of installation in plant engineering at Basler Rosenmund Haustechnik AG and a user of the AI email assistant. “New plan statuses reach our teams faster and without detours. This reduces sources of error and gives our project managers more certainty in execution.”

  • Million-euro financing accelerates autonomous construction site technology

    Million-euro financing accelerates autonomous construction site technology

    Venture capital firms from three continents have invested in Gravis Robotics: The spin-off of the Swiss Federal Institute of Technology in Zurich, founded in 2022, has raised 23 million dollars in an early financing round, according to its information.

    The round was led by the venture capital companies IQ Capital from London and Zacua Ventures from San Francisco. Pear VC from Palo Alto, California, Imad Ventures from the Saudi Arabian capital Riyadh, Sunna Ventures from Miami and the Zurich-based company Armada Investment as well as the globally active cement manufacturer Holcim from Zug also participated.

    Gravis Robotics offers autonomous earthmoving machines that combine artificial intelligence, machine vision and human interfaces to increase throughput, reduce waste and improve safety on construction sites, whether the operators are in the cab or coordinating the work remotely.

    With the recent funding, Gravis now has the technology, partnerships and global distribution channels across the industry to drive the adoption of true autonomy on a large scale, the company said. In addition, it also announced “a wave” of new industry partnerships. For example, Gravis Robotics has partnered with Taylor Woodrow in the UK on a major infrastructure project at Manchester Airport, carrying out the first autonomous excavation work on a major active construction site in the country. There are also agreements with Holcim and the South Korean company HD Hyundai.

    The fastest route to autonomy is through increasing productivity, CEO Dr. Ryan Luke Johns is quoted as saying. “By providing operators with real-time 3D intelligence and the ability to seamlessly switch between autonomy and advanced control, we are covering more of the work, accelerating the application and creating the data pipeline needed to learn new skills from the industry’s most challenging tasks. The company considers it an advantage to have its Zurich headquarters “at the heart of the renowned robotics and automation ecosystem”.