Scalable ESG reporting in property management
M&P is shifting its focus – from the question of tools to data-driven value creation.
Are you already talking to your building?
BIM@FM meets AI, or: how M&P BEGIS is using artificial intelligence in the context of BIM for building operations.
BIM has been one of the leading topics of the future in facility management for years. Solutions for ‘bridging the gap’ – that is, the integration of the construction and operational phases – have now not only been developed but are also in practical use. This ensures that no information from the construction phase is lost. No sooner has this been achieved than the topic of artificial intelligence (AI) opens up new perspectives. M&P BEGIS is already utilising AI within the framework of BIM and will consistently expand these capabilities as practical experience grows.
There are already many approaches to the application of AI. According to the 2024 GEFMA white paper ‘AI in Property Management’, the primary aim is to ‘reliably support people in demanding tasks and when making complex and difficult decisions’. According to the white paper, “so far, only the beginnings of” the integration of AI components capable of, amongst other things, carrying out analyses “can be discerned”.
M&P BEGIS trains artificial intelligence for use in BIM
This is where our solution, described below, comes in. It all began with the aim of simplifying operation. The integrated and appropriately ‘trained’ AI interface ‘translates’ voice inputs into system-compatible search queries. To display the results, our SAP experts developed dynamic views. This allows data to be queried individually and independently of the previously inflexible display options. As part of this process, the previously rigid 2D models were also made dynamic.
As a result, individual queries can be made intuitively. This enables user enquiries to be answered by retrieving and displaying data across floors, buildings or entire estates. Consequently, even complex user enquiries can be answered, for example:
- “Show me all buildings in estate X where ventilation systems of model Y are installed.”
- “Show me all the rooms in Building X where heating costs were above average last year.”
- “Calculate the average service life of component X in system Y.”
Centralised data storage for fast, error-free AI queries
Of course, the AI can only analyse data that is available. But why use a database or cloud solution as a single point of truth in the age of AI? Data of various types and structures is required. Examples include data from building automation, master data – and thus basic information on machinery and technical systems and/or the building itself – as well as (consumption) data collected via measuring points or meters, and so on. Storing this data in a central location, rather than in various different places, has many advantages. Apart from the (system) costs, duplicate data is avoided, the data structure is consistent, and the ‘right’ data is available for analyses, reports or even actions.
Looking towards the future:
the next steps
Once this has been done, data for ESG reporting, for example, can also be retrieved with AI support. This includes quantitative data – that is, numerical and measurable information on greenhouse gas emissions, water and energy consumption, waste generation and environmental pollution. As mentioned at the outset, the initial aim was to make the system easier to use. However, in practice, a wide range of additional use cases are emerging. For example, generating forecasts of expected conditions and situations based on the recorded data. This is done to predict when high utilisation will occur, where and when utilisation peaks are to be expected, or when maintenance work tailored to requirements needs to be carried out. Data from building automation systems, amongst other sources, is used for this purpose. The relevant data collected via sensors is transferred to the CAFM system, aggregated where necessary. The AI, which has been trained accordingly, identifies critical situations based on the status data and generates appropriate (warning) messages. These can either be dealt with manually or trigger automated actions.
This offers significant advantages both in space management and for the maintenance of technical systems, and, of course, for all types of reporting. M&P BEGIS has reached an important milestone in the use of AI in facility management. We are now moving forward.
