New European directives on energy efficiency, targeting a 55% reduction in greenhouse gas (GHG) emissions to be achieved by 2023, are triggering deep renovation building projects, which are largely responsible for these emissions. This high demand for the transformation of the existing building stock makes us consider the need to execute this type of renovation projects in the shortest period of time. Furthermore, it is important to offer an adequate cost/benefit balance for the proposed interventions.
And in this process of transition towards climate-neutral buildings, how can the use of new technologies and the application of methodologies such as Building Information Modelling (BIM) help in the implementation of deep renovation projects? The adoption of BIM models, traditionally used for new buildings, can provide important decision support when selecting solutions to be implemented in renovation projects. This was one of the main objectives of the H2020 BIM-SPEED Project, to improve deep renovation projects of residential buildings, reducing the time and costs associated with them, and promoting the use of BIM among the different stakeholders involved. To this end, standardised processes, with the creation of Use Cases, and different BIM‑based tools were developed as part of the BIM‑SPEED web platform ecosystem, as well as training materials on how to use these services1. To address interoperability issues, different ETLs (Extract, Transform and Load) and BIM connectors were implemented.
Interoperability framework between BIM tools and the BIM-SPEED web platform, showing the connection to the implemented ETLs and BIM Connectors. To ensure the reliability of the data, different Checker tools were applied
It was also possible to see how beneficial the combination of Machine Learning techniques with BIM models is for decision making in deep renovation projects, allowing the automatic selection of the most appropriate renovation option. This selection is based on national building envelope regulations, and also takes into consideration a number of user-defined input parameters on the limitations of its application2. The combination of the Scan to BIM process with the automatic creation of walls in BIM, using point clouds as input data, was also of great interest to end users3.
And now, what else?
The possibilities of using BIM models do not end with the renovation phase of the building. These models can also play a key role in the Operation and Maintenance phase. The development of Digital Building Twins based on BIM models can help in the optimisation and control of buildings to improve their energy performance. In line with this, projects such as BuildON, coordinated by CARTIF, and SMARTeeSTORY, the latter focused on monitoring and optimisation of the energy performance of non-residential historical buildings, are starting. We will keep you updated on further developments in future posts.
2 Mulero-Palencia, S.; Álvarez-Díaz, S.; Andrés-Chicote, M. Machine Learning for the Improvement of Deep Renovation Building Projects Using As-Built BIM Models. Sustainability2021, 13, 6576. https://doi.org/10.3390/su13126576
3 Álvarez-Díaz, S.; Román-Cembranos, J.; Lukaszewska, A.; Dymarski, P. 3D Modelling of Existing Asset Based on Point Clouds: A Comparison of Scan2BIM Approaches. In 2022 IEEE International Workshop on Metrology for Living Environment (MetroLivEn); IEEE, 2022; pp 274–279. https://doi.org/10.1109/MetroLivEnv54405.2022.9826964
The BIM approach (Building Information Modelling) is all around Architecture, Engineering and Construction professionals, but when it comes down, very few companies are founding their daily work on this paradigm and applications are really far from being homogeneous. BIM is many times (let’s say “usually”) incorrectly identified as a specific software package or a type of 3D digital model. However, BIM is much more than a newer version of CAD or a 3D visualisation tool.
The BIM approach provides a digital featuring of a building or infrastructure throughout its whole life-cycle, adding extra information to help making better and more-timely decisions upon a 3D model that allows a multidimensional analysis: 4D (evolution); 5D (costs); 6D (sustainability -including energy efficiency-); 7D (maintenance).
Although there is still a lack of knowledge on how BIM and associated digital innovations are applied across European countries, the European Directive 2014/24/EU imposes BIM Level 2 for government centrally procured projects. Level 2 refers a collaborative process of producing federated discipline specific models, consisting of 3D graphical data (those visually represented) and semantic data (those significant additions) as well as associated documentation (for instance: master plans). Information is exchanged using non-proprietary formats, such as Industry Foundation Classes (IFC).
Consequently the built heritage is subject to BIM for the purposes of documentation, conservation and dissemination, but the distinctiveness and sensitivity to meet heritage demands requires technological and methodological particularizations leading to the concept of Heritage-BIM (H-BIM). The purpose of H-BIM is to provide a 3D parametric model as a “container” of information generated all over time by different procedures, by different people, and from different sources (hw & sw). The model would capture the multidisciplinary nature of Heritage, far away from the simplicity and modularity of conventional construction, and would be very useful to study, evaluate the state of conservation and plan interventions on the assets in a profitable way. It is quite a challenge for a sector where digitization is a pending issue.
This technologically means facing many challenges, starting with the minimum amount of graphical and semantic data that would be adequate to support the activities of the sector. Two of the most important are:
The combination of 3D data with different types of images (thermography, high resolution photographs or multispectral recordings) to produce a really useful H-BIM model for exhaustive assessment.
The photorealistic texturing of 3D models for a rigorous representation of reality.
Both aspects are being worked by CARTIF to decisively help companies, managers and public administrations in the digitization of Cultural Heritage.
In 2025, the volume of data created and consumed worldwide exceeded 180 zettabytes (one zettabyte equals one billion terabytes), and this figure is expected to triple between 2025 and 2029. This global digital explosion has placed data centres at the heart of today’s critical infrastructure. From the perspective of the construction sector, data centres represent one of the most technically specialised building types. It is not simply a matter of erecting a large warehouse filled with equipment: the structure, the building envelope, the electrical installations, the cooling systems and the security controls must all function with almost surgical precision. And increasingly, these buildings must do so in a sustainable manner.
Every time we send an email, check the weather forecast or ask artificial intelligence for help, that request travels to a building that almost no one ever sees. A data centre is not your typical company server at the end of a corridor; it is a highly complex industrial infrastructure, with thousands of servers and a multitude of auxiliary systems. They are designed to store, manage and process vast amounts of data and ensure it is always available, supporting essential digital services: from cloud platforms to artificial intelligence, including banking, public administration and the connected industry.
Types of data centres: they are not all the same
The construction of a data centre varies dramatically depending on its purpose and scale. Designing a facility for an SME is not the same as designing a massive node for a tech giant. A distinction must be made between different types of data centres based on various characteristics:
Classification according to purpose:
On-Premises, built and managed by the company itself for its exclusive use, offering maximum control
Colocation (Colocation),where the provider rents out space, power and security to various companies, enabling them to outsource their infrastructure
Cloud Data Centers, virtualised infrastructure hosted by cloud service providers (e.g. AWS, Azure), designed for scalability and on-demand use
Hybrid, which combine physical infrastructure with cloud services, optimising flexibility and security
Managed services, which are those where the provider not only rents out space, but also actively manages the customer’s infrastructure.
Clasification according to size and capacity:
Hyperscale, which are large-scale facilities designed for massive cloud computing and big data, operated by tech giants (Google, Meta, AWS)
Edge Data Centers, which are smaller, decentralised facilities located close to the end user to reduce latency
Micro Data Centers, which offer compact, often prefabricated solutions for specific requirements or confined spaces.
Classification based on their level of redundancy and guaranteed availability:
This classification is based on the Uptime Institute’s Tier standard, which is the global benchmark.
Clasification
Availability
Redundancy
Typical profile
Tier I
99.671%
No redundancy
SMEs, offices…
Tier II
99.741%
Partial
Medium-sized companies
Tier III
99.982%
Maintenance without downtime
Deployment, cloud region
Tier IV
99.995%
Total, fault-tolerant
Banking, defence, hyperscale
Energy consumption: some staggering figures
One of the most critical aspects of data centres is their high energy consumption. A hyperscale data centre can consume as much electricity as a city of 100,000 inhabitants. Designing them properly is not just a technical matter: it is a collective responsibility. The International Energy Agency (IEA) estimates the annual energy consumption of data centres at around 450 TWh, accounting for almost 2% of global energy consumption. And estimates suggest that this consumption could double by 2030.
The energy efficiency of a data centre is measured using the PUE (Power Usage Effectiveness) metric. A PUE of 1.0 is the theoretical ideal – all the energy goes to the servers – and a PUE of 2.0 means that for every watt of useful power, another watt is spent on cooling, lighting and other auxiliary systems. The industry average is around 1.5, although the best modern centres already achieve figures between 1.1 and 1.2.
PUE (Power Usage Effectivenes) = total energy consumption of the facility / energy consumption of IT equipment
Cooling can account for between 35% and 45% of a building’s total electricity consumption. For this reason, innovation in cooling systems – liquid immersion cooling, two-phase heat exchangers, indirect evaporative cooling systems – is one of the most active areas of research and also one of the fields where we at CARTIF see the greatest potential for technology transfer.
The waste heat from a data centre should not be wasted; it can (and should) be reused for district heating systems or for nearby industrial processes. A data centre can literally act as the boiler for an entire neighbourhood. Recent European projects demonstrate that it is feasible to feed this heat into district heating networks to heat homes and commercial buildings in winter, turning a problem into an energy asset.
Challenges and solutions in data centre construction
From a construction perspective, data centres differ significantly from conventional buildings. Their design is determined by three key factors: availability, security and efficiency. Unlike conventional construction, the building process for a data centre is governed by critical concurrence: civil engineering works and the integration of complex systems (electrical and mechanical) must proceed in perfect synchronisation. The greatest challenges facing this type of specialised construction are:
Structure and floor load: server racks can exceed 1,500 kg/m². Floor slabs must be dimensioned to a much higher standard than in a conventional office building, and the structure as a whole must incorporate robust construction solutions designed to withstand external risks (earthquakes, flooding, fire).
Redundant power supply: dual mains connections, diesel or hydrogen-powered generators, and large-scale UPS (uninterruptible power supply) systems are required to ensure there are no micro-outages. Electrical rooms may occupy up to 30% of the total floor area.
Cooling: this is the major challenge, as the amount of heat generated by the servers is enormous. Systems range from precision air conditioning (CRAC/CRAH) to direct liquid cooling at the rack level, chilled/heated ceilings and free-cooling towers, which utilise outside air.
Physical security: another critical aspect that must include biometric access control, 360° cameras, electromagnetic shielding (Faraday cages), early-warning fire detection systems using aerosol or clean agents that do not damage equipment, and building materials with high REI ratings.
Cable and infrastructure management: raised access floors and suspended ceilings must accommodate kilometres of fibre-optic cables and other wiring, with strict compartmentalisation and redundant routes.
Building envelope and efficiency: it is essential to incorporate highly insulated façades, roofs that minimise solar gain, and a carefully considered orientation to harness prevailing winds for passive free-cooling strategies. Finally, the layout of plant rooms, the building’s orientation, and the integration of passive and active systems must optimise overall energy consumption.
The construction process for a data centre has specific characteristics that set it apart from other types of building projects, primarily due to the need to precisely coordinate multiple disciplines that rarely come together in a single project: heavy-load structural engineering, medium-voltage electrical systems, precision air conditioning, advanced BIM management and monitoring technologies. Following an initial phase of highly detailed planning and design – often supported by BIM methodologies – the construction phase is characterised by a tightly controlled sequence in which civil works and the installation of critical systems proceed in parallel.
Construction usually begins with a robust foundation (foundations and structure) capable of withstanding heavy loads and ensuring stability against vibrations. Subsequently, particular importance is attached to the installation of redundant electrical systems (transformer stations, generator sets, UPS systems) and HVAC systems, the integration of which requires specific technical spaces and extremely precise execution. In the final stages, exhaustive testing (commissioning) is carried out to verify that all systems operate in a coordinated manner under different operational scenarios, which is critical prior to commissioning.
The life cycle of a data centre – from conception through to operation and decommissioning – offers fertile ground for the application of emerging technologies. The following are the most significant from the perspective of construction and facilities engineering:
Digital twins and BIM. The BIM (Building Information Modelling) methodology is now an absolute must for data centre projects. It enables the precise coordination of installations across all disciplines before construction begins, identifying clashes and sequencing the work. The next step is the operational digital twin: a model updated in real time using installed sensors, which enables the simulation of failure scenarios, the optimisation of load distribution, and the management of predictive maintenance.
Industrialised and modular construction. Prefabricated data centre modules help to improve quality and reduce commissioning times. This trend is being adopted by leading data centres as a strategy for rapid scaling.
Artificial intelligence in management and predictive maintenance. State-of-the-art DCIM (Data Centre Infrastructure Management) systems incorporate machine learning algorithms that optimise the operation of cooling equipment in real time, manage load distribution between servers and predict when a component is likely to fail before it does.
Direct Liquid Cooling (DLC). Given the power density of new artificial intelligence processors – which can exceed 400 W per chip – air cooling is simply not enough. DLC systems circulate water or a dielectric fluid directly to the processor via cold plates, shifting heat management to the hydraulic system and enabling that heat to be recovered at usable temperatures.
Renewable energy and hydrogen. Major technology corporations have committed to operating on 100% renewable energy, and many data centres are incorporating rooftop solar photovoltaic systems or using PPAs (Power Purchase Agreements). Green hydrogen is emerging as an alternative to diesel generators for long-duration energy storage, with the first pilot projects already underway in Northern Europe.
The economic impact of data centres: why regions are competing to attract them
The importance of this infrastructure to the economy is undeniable. In Spain, planned investment for this year exceeds €8 billion, with projections reaching €67 billion by the end of the decade. A large data centre is not just a building: it is a major economic catalyst. The construction of a hyperscale centre creates between 400 and 2,000 direct jobs during the construction phase, with high demand for specialist roles such as industrial electricians, HVAC technicians, fibre-optic network installers and BMS (Building Management Systems) operators.
Once up and running, data centres create stable, well-paid jobs – systems technicians, facilities engineers, security operators – and pay electricity bills that make a significant contribution to the revenue of distribution companies and to local tax revenues. Over the last three years, Spain has experienced a wave of investment, particularly in Madrid (which is already one of Europe’s five largest data hubs), but also in regions such as Aragon, Navarre and Galicia, which have set up one-stop shops and introduced favourable electricity tariffs to attract major operators who, in turn, draw in technology companies seeking proximity to the infrastructure. In regions such as Castile and León, with land availability, access to renewable energy and favourable climatic conditions, there is a clear opportunity to attract this type of investment.
From a regional policy perspective, investing in this type of infrastructure contributes directly to the objectives of the European Digital Agenda (Digital Compass 2030) and to technological sovereignty, ensuring that the data of European citizens and businesses is not stored exclusively on infrastructure in third countries.
Opportunities and benefits for businesses and researchers
As we have seen, data centres can offer significant benefits, including the creation of direct jobs during the construction and operational phases, the development of highly advanced energy and telecommunications infrastructure, the attraction of technology companies and start-ups, and, more generally, an increase in regional competitiveness.
The data centre sector opens up numerous business opportunities where collaboration between industry and research centres is particularly valuable:
Specialist technical consultancy. The market is in demand for architects and engineers with specific training in data centres. The shortage of professionals with this profile in Spain presents a real opportunity for firms and consultancies that invest in training and certification.
Research into energy efficiency. Technology centres such as Cartif play a key role in developing and transferring solutions for energy optimisation, waste heat recovery and integration with smart grids.
Advanced building materials and systems. The industry demands high-efficiency building envelopes, water management solutions for evaporative cooling systems, and materials with rigorous environmental certifications (EPD, cradle-to-cradle).
Predictive maintenance and inspection. The use of drones equipped with thermal imaging, inspection robots and AI-based data analysis platforms to predict failures in critical infrastructure is a growing market offering high added value.
Industrial symbiosis with waste heat. Integrating data centres into district heating networks requires systems engineering, planning permissions and innovative business models, which provide fertile ground for applied research.
CARTIF’s role in promoting sustainable data centres
Data centres are, paradoxically, the most influential buildings of our time and, at the same time, the most invisible to the general public. They are built in industrial estates, hidden behind unassuming façades, and only feature in the media when something goes wrong. Yet every internet search, every bank transaction, every video call and every query to an artificial intelligence model passes through them.
At CARTIF, we see these projects as an exciting crossroads: the need to build faster, more efficiently and more sustainably, whilst demand is growing at a rate that defies all forecasts. The decarbonisation of the sector, the smart management of water resources in regions such as Castile and León, and the integration of data centres into the urban and energy fabric of cities are challenges that require precisely the kind of applied research and public-private collaboration that is our raison d’être.
When we enter a cathedral, stroll through a monastery, or visit a castle, we rarely think about everything that is happening “inside” them. We do not see how moisture slowly rises through the walls, how increasingly frequent and abrupt temperature changes generate invisible stresses, or how a millimetric crack can eventually become a striking fissure over time. And yet, that is often where the deterioration of heritage begins.
Preserving our historic buildings is not just about restoring them when a crack appears or cleaning them when a façade looks worn. Above all, it is about anticipation. It means understanding what is happening to them before the problem becomes evident. With this idea in mind, the Comprehensive Intelligent Monitoring and Predictive Risk Assessment Model for Cultural Heritage Assets (MIMER-BIC, Spanish acronym) was developed by the Cultural Heritage Area of CARTIF.
This model is based on something we can all agree on: to know what is happening to someone, we must first listen. And if that “someone” is something as valuable as our historic buildings, listening means measuring. Sensors record temperature, humidity, light (infrared, visible, and ultraviolet), air quality, crack growth, wall inclination, vibrations, the presence of insects that attack wood, the number of visitors, or even potential intrusions. However, the real innovation lies not in placing sensors, but in transforming that data into useful information. The model converts all these measurements into clear indicators and risk indices that, on a simple scale from 0 to 100, reveal whether a building is in a stable condition or requires priority intervention.
MIMER-BIC model graphic representation
Thanks to this methodology, it is possible to detect whether the indoor environment is endangering paintings or altarpieces, whether a structure is undergoing abnormal movements, whether excessive visitor numbers are affecting the microclimate, whether a weather event could accelerate external deterioration, whether a fire is starting, or whether someone has entered a restricted area. The focus is no longer on reacting once damage is visible, but on preventing it in advance and, above all, doing so with sound judgment.
Behind this advancement lie many years of research. The CARTIF team has worked intensively in different technologies such as:
3D surveying
HBIM
Preventive conservation
Structural analysis
Risk modelling
Advanced sensorization
Artificial intelligence applied to heritage
Yet this journey has not been undertaken alone. Close collaboration with companies in the sector (where the role of TRYCSA has been particularly noteworthy) has been key to ensuring that the model did not remain on paper but became a practical reality. Their hands-on experience, technical expertise, and commitment have made it possible to test, refine, and transform the methodological proposal into an effective and applicable tool.
The result is an original model with its own methodology (from the definition of architectural-functional typologies and major families of pathologies to the formulation of synthetic risk indices), protected under intellectual property regulations. This protection is not merely a legal formality: it is recognition that we are facing a pioneering proposal with high scientific and technological value for the heritage conservation sector. A model developed in the region of “Castilla y León”, yet with a clearly global vocation and positioned at the forefront of applied research in cultural heritage.
“We are facing a pioneering proposal with high scientific and technological value for heritage conservation sector”
At a time when climate change, tourism pressure, and limited resources are testing conservation capacities, having tools that enable prioritization, planning, and decision-making based on objective data is more necessary than ever. The MIMER-BIC model represents precisely that: a new way of caring for what belongs to us all, combining expert knowledge, technology, and collaboration between research and industry. Because, in the end, preserving heritage is not only about keeping old buildings standing. It is about protecting stories, memories, and a shared identity. And doing so intelligently today is the best guarantee that they will still be there tomorrow.
Forests are one of the most valuable pillars of our natural environment. Not only do they provide renewable raw materials such as wood and resin, but they also perform essential functions for life: they regulate the climate, act as carbon sinks, conserve biodiversity, protect the soil from erosion, and provide spaces for well-being and rural development.
However, these ecosystems face major challenges. Climate change, biodiversity loss, forest fires, and rural depopulation threaten their balance. Added to this are structural difficulties within the forestry sector itself, such as international competition, a shortage of skilled labor, and the need to improve operational efficiency. In this context, active forest management is more important than ever. A well-managed forest is a resilient forest, capable of withstanding pests, diseases, and, above all, fires. The devastating fires of recent years have highlighted the urgent need to modernize the sector, moving towards a digital and ecological transition that will transform traditional forestry into an innovative and sustainable bioeconomy.
In regions such as Castile and León, where forest areas represent a significant portion of the territory, sustainable management of natural resources has become a strategic priority. This requires not only technical knowledge, but also advanced digital tools that facilitate decision-making and process optimization. Forests are not just natural landscapes: they are complex ecosystems that provide incalculable benefits, from maintaining biodiversity to providing economic sustenance for rural areas. Furthermore, their role as natural carbon sinks makes them indispensable allies in the fight against climate change.
Modern forest management involves much more than conservation: it means planning, monitoring, and adapting land use, drawing on knowledge, technology, and collaboration between administrations, companies, research centers, and society. The ecological and digital transition in this area is an opportunity to improve efficiency, prevent environmental disasters, and consolidate a forest bioeconomy that generates employment and development in rural areas.
Innovation and digitization at the service of forests
The drive towards digitization and the use of advanced technologies—such as GIS systems, BIM models, and data spaces—are transforming the way we understand and manage forest ecosystems. These tools provide accurate, real-time information on the state of forests, optimize harvesting, and strengthen prevention against fires and other environmental risks.
One of the most notable innovations is the development of continuous forest inventory systems, which provide real-time information on the existence, growth, and carbon stock of forest areas. Thanks to remote sensing, artificial intelligence, and satellite image processing, it is possible to monitor millions of hectares, detect changes in land use, and plan actions more efficiently.
Digitization is also changing the way forest fires are tackled. Automatic detection systems using artificial intelligence, combined with meteorological and satellite data, enable dynamic risk maps to be generated and improve the coordination of firefighting teams. All of this reduces response times and increases the effectiveness of emergency management. The forest bioeconomy, understood as the comprehensive and sustainable use of forest resources, finds a decisive ally in technology. From mobile applications for real-time management to digital certification or traceability platforms, digitization is redefining the sector’s value chain. Artificial intelligence-based solutions enable the automation of processes, improved occupational safety, and optimized forest logistics, thus promoting a more competitive and sustainable model.
The European framework: investing in resilience and digitalization
The European Union has made a firm commitment to transforming the primary sector through the Recovery, Transformation, and Resilience Plan, financed by Next Generation EU funds. This is the framework for Order MAV/626/2025, issued by the Regional Government of Castile and León, which regulates subsidies for the implementation of the RetechFOR project, Technological and Territorial Network for forest monitoring and environmental disaster reduction as levers for the development of the forest bioeconomy, one of the initiatives of the RETECH program, Territorial Networks of Technological Specialization, a tool launched by the Secretariat of State for Digitalization and Artificial Intelligence. The project is worth €28.45 million, 75% of which is co-financed by the European Union and 25% by the autonomous communities of Castile and León and the Canary Islands.
This initiative seeks to modernize the management of forest resources through the intensive use of enabling technologies and the creation of an interoperable data infrastructure connecting administrations, companies, and research centers.
The RetechFOR project focuses on developing advanced solutions in three key areas. First, monitoring and prevention through the use of satellite data, sensors, and artificial intelligence algorithms for early warning and dynamic generation of fire risk maps. Secondly, data management through the creation of an interoperable forest data space, a crucial infrastructure for the flow of information between the administration, managers, and industry. Finally, precision forestry through the implementation of a continuous forest inventory that, through the use of digital twins, allows for the optimization of forest use and planning, ensuring an efficient and sustainable forest bioeconomy.
RetechFOR partners
CARTIF: innovation for digital forest management
As part of this initiative and with the aim of ensuring the best technological implementation, CARTIF is actively collaborating in the execution of the AG-RetechFOR project. This technology center is leading the development of forest data spaces that guarantee the interoperability, sovereignty, and traceability of information, which are fundamental elements for the success of the sector’s digital transformation.
CARTIF’s participation in the RetechFOR project focuses on three key areas that are essential for the modernization of the forestry sector. First, the design and implementation of interoperable connectors that facilitate the secure integration of heterogeneous data from multiple sources, allowing dispersed information to be consolidated and analyzed efficiently while maintaining the sovereignty and governance of the information sources.
Secondly, the development of digital platforms for natural heritage management, incorporating multi-layer geographic information system technologies that enable the visualization and integrated analysis of geospatial information. These tools are essential for understanding territorial complexity and making informed decisions about natural resource management.
Finally, the application of information modeling methodologies for construction in the digital management of critical forest infrastructure represents a significant innovation. This approach allows for the creation of accurate digital representations of forest facilities, facilitating their maintenance, optimization, and long-term planning.
CARTIF’s technical capacity, combined with its compliance with European standards on data spaces, positions the technology center as a strategic player in the sustainable digital transformation of the region. Its contribution is key to building a more resilient, efficient, and competitive forest ecosystem in Castile and León, demonstrating that technological innovation and environmental sustainability can and must go hand in hand.
