From BIM to building operation: where digital facility management stands today

Today, a building can be almost completely digitally planned. Rooms, systems, and components exist as data objects long before completion, technical properties are stored, and models are exchanged between specialized disciplines. After handover, however, a completely different phase begins: the building must be operated, maintained, cleaned, and optimised for energy efficiency. Suddenly, different information becomes important than during planning.

Now it becomes clear how far digitalization actually reaches. Can the data generated in the project be used directly for maintenance, space management, or operator obligations? Or are extensive BIM models available, but facility management lacks essential equipment identifiers, maintenance specifications, or reliable information about the built condition?

Three quarters of respondents still work with Office & Co

The gefma trend report Digitalization in the Real Estate Industry (opens in a new window) from 2025 shows how widely the industry varies in this regard. According to the report, three quarters of the surveyed users still work predominantly with Office applications or classic CAFM and IWMS systems. At the same time, the demand for more modern architectures is clearly increasing. For two thirds of the current project awards examined, a cloud variant is already preferred. The discrepancy is even greater for artificial intelligence. 75 percent of users name AI as an important digitalization trend, especially for data analysis, automated reporting, and energy optimization. However, only about one in five has invested in advanced AI solutions so far.

The Lünendonk study “Facility Service in Germany (opens in a new window)” from 2026 also describes a market in transition. For 30 percent of the surveyed providers, operational facility services are already running in an integrated system landscape, and 33 percent work with standardized digital process chains. Another 26 percent use individual digital tools. Partially automated or AI-supported solutions have so far been adopted by eleven percent. As the biggest hurdle for faster digitalization, 64 percent cite a lack of budgets or a low willingness to invest.

Technology is therefore developing faster than its widespread adoption. At the same time, the need for reliable building data is growing. Energy and resource management, ESG requirements, operator obligations, maintenance, or future AI-based applications only work if the underlying information is correct and accessible.

1 / 2

2 / 2

Not as much data as possible, but the right data

Mahmut Tümkaya, Managing Director of Piepenbrock FM Consulting and Chairman of CAFM RING, has been dealing with the interface between construction and operation for years. He identifies a fundamental problem: during planning, a great deal of data is easily generated without adequately clarifying beforehand which of it will actually be needed during operation. An extensive BIM model is of little help to facility management if it contains numerous properties but lacks relevant maintenance or operational information.

The Level of Information Need, or LOIN for short, addresses precisely this. It describes what geometric, alphanumeric, and documentary information is required for a specific purpose, at a specific time, and in what quality. The concept is now regulated in DIN EN ISO 7817-1:2024. The idea behind it is simple: not to collect as much data as possible, but the right data.

For this to succeed, the client must already define what information will be needed later. The client information requirements (Auftraggeber-Informationsanforderungen, AIA), serve this purpose. They define what information is to be delivered at what time and in what quality. The BIM execution plan (BIM-Abwicklungsplan, BAP), then describes how these requirements will be implemented in the project. BIM Deutschland, the federal government's central BIM contact point, refers to both as the central basis for BIM implementation.

For facility management, this means that operator requirements must not be formulated only shortly before the building handover. Anyone who later wants to digitally manage maintenance intervals, technical systems, or space structures must determine early on what information is needed in the model for this purpose. The market is also moving in this direction. A quarter of the facility service providers surveyed by Lünendonk in 2026 observe that clients want to involve their service providers in the planning phase of buildings in the future.

Open standards against new data silos

However, even neatly structured data is of little use if it only works within a specific software. Prof. Dr. Niels Bartels from TH Köln therefore sees open interfaces as an essential prerequisite for long-term usable information management. Bartels' research focuses, among other things, on cross-lifecycle data exchange, smart buildings, digital twins, and BIM and sustainability. Especially in facility management, information often has to last for decades. During this time, software products, service providers, and owners change. Building data should therefore not be tied to a single application.

One approach to this is CAFM-Connect. The open data standard developed by CAFM RING is based on IFC and works with so-called BIM profiles for specific use cases in building operation. Instead of transmitting all possible properties of a component, a BIM profile only contains the information needed for the respective process. Different data is needed for maintenance than for cleaning or space management. The basic profile includes, among other things, room usage types according to DIN 277-2, component types based on or as an extension of DIN 276, and document types according to GEFMA 198. Building data can be stored in IFC format and transferred to compatible CAFM systems.

In addition, there is BIMeta. This open cloud-based property server manages classes and properties for BIM data exchange and connects them with rules, standards, and the buildingSMART Data Dictionary. CAFM-Connect is also maintained there. The goal is a common machine-readable language for information that is to be passed on from planning to operation.

From BIM model to digital twin

A BIM model does not become a digital twin merely because it continues to be used after completion. The difference lies in the connection with the real building and its current data. If current information from the real building is linked to it, a digital twin can emerge. Sensors can, for example, provide energy consumption, temperatures, or operating states of technical systems. If this data is linked to the corresponding objects in the model, states can be analyzed, maintenance can be prepared more specifically, or operating strategies can be adapted.

In this context, 7D-BIM is often discussed. BIM Deutschland uses this term to describe the expansion of the model with information for facility management and building operation, for example, on technical systems, maintenance plans, manufacturer specifications, or upkeep. The term thus stands less for another spatial dimension than for the use of BIM data in the operational phase.

With AI, the data question gains additional importance. Systems can analyze large amounts of information, recognize patterns, automate reports or support predictive maintenance. Especially in facility management with its many recurring processes, this offers significant potential. However, current study figures show that practical use is still in its early stages. While AI is seen as an important trend by three quarters of gefma respondents, only about one fifth invests in advanced applications. According to Lünendonk, eleven percent of facility service providers already use partially automated or AI-supported solutions. AI does not solve a fundamental data problem. If assets are mislabeled, information is missing, or data is stuck in incompatible systems, a bad data basis will not become a good one through an algorithm. This development therefore rather increases the pressure to record building data in a structured way and keep it available long-term.

Digitization needs people who understand the data

In addition to technology and standards, the shortage of skilled workers remains a challenge. The job profile in facility management is expanding: knowledge of buildings and technical systems is increasingly no longer sufficient. Data management, BIM, CAFM systems, interfaces, cloud platforms, cybersecurity, and AI are being added.

At the same time, collaboration is becoming more important. A large part of the operational data needed later is not generated in facility management itself, but during planning, tendering, construction, and commissioning. For the information to be usable later, planners, construction companies, manufacturers, and operators must know early on what is needed from each other. Niels Bartels clearly articulates this organizational task: "We simply need to cooperate much more. Theory, research, and practice must be brought to the table more often. FM is not the most attractive topic in the BIM sector, but it is one of the most important."

What matters is how the data is used

The development in facility management can therefore only be measured to a limited extent by how many companies use BIM. What is more interesting is what happens with the information after the building handover. Cloud platforms are more frequently tendered, integrated system landscapes are spreading, and initial companies are automating operational processes with AI. In parallel, structures are being created with LOIN, AIA, IFC, CAFM-Connect, or BIMeta that enable more reliable data exchange. The real test, however, remains the building operation itself. If information from planning and construction can be used there without re-entry and subsequently kept up-to-date, a robust digital basis for maintenance, energy optimization, sustainability, and automated processes is created. The actual progress, therefore, does not lie in the most comprehensive BIM model possible. It lies in ensuring that the right information arrives at the right time where it is needed.