The Importance of Data Management in an OpenBIM Framework

By Nigel Rees and Eduard Mrazek

Effective data management is a fundamental and critical factor when dealing with information in any business process workflow. It is no different in a Building Information Modeling (BIM) focused model management platform such as Bimplus; some might say it is even more important. Why? A platform of this type is designed to ensure that there is data integrity and continuity, with intelligence built in to recognize change and manage it for the teams working on the project. A system based on a relational database is therefore more suited to this than one simply based on files – especially when bringing in data from multiple model sources and locations, as is the case on a construction project. If effective data management is the key to a smooth project workflow, then a relational database is the foundation of this process.

Bimplus was built from the ground up not as a viewer, but as a database first. The importance of this cannot be stressed enough – it means the data was the most important consideration from the outset. The advantage of a database over a file-based system is quite easy to see when you consider some of the strengths that databases bring with them.

Databases reduce data redundancy: they host information once – the single source of truth that is often referred to. Adopted correctly, this then makes this information available to other functions that need to access it, as opposed to file-based systems that typically have to make that data available over and over again and write to the files that will use that data. Copying information to multiple files allows for errors to occur easily, especially when changes happen frequently. In this respect, databases offer improved data integrity in comparison.

Data security is another big advantage of database systems. A great deal of work has been done to create layers around the data to ensure it can only be accessed by those authorized to do so. This is not always the case with file-based systems; it is quite easy to intercept, decompose, or corrupt the data written within.

So, what do we mean by a relational database? This is typically defined as a well-organized set of data with pre-defined relationships in existence that allow objects to be represented with all the attributes associated with them. We use this concept for Bimplus, which means we always build a rich data model with interdependencies no matter where the source of that model is or what information is related to it – such as documentation, for example. Within Bimplus, we do not distinguish between the resource of the object. Whether it comes from IFC, ALLPLAN, Sketchup, LandXML, or any other 3rd party application, it goes through a process to neutralize the data into a format that fits the purpose of content and data management.

Because we keep the data in a highly organized structure, it means it can be exported into the industry standard IFC format as required. However, we also make the whole data structure available through a very powerful API so that it can be connected to different software and become the data host for any process that requires it. For example, this could be the more traditional BIM processes of collaboration and issue management, or it could be used for applications such as the manufacturing of precast concrete elements.

All the services that we build around this database benefit from high data integrity – there is no need to reference dummy files, or make multiple checks on the data when it is being reused in these services, such as Revision Compare. The data structure within Bimplus allows us to compare models using real building objects within a federated model set.

By incorporating the principles of relational databases and customizing this for BIM-related activities, we ensure that the data in Bimplus is always true, always secure, and always available. The result is a reliable data source for any discipline requiring access to the data held within – and therefore, an efficient and effective data management process.