Technology

3DIR: it’s time to join the search party

27 May 2015 | By Peter Demian, Kirti Ruikar, Anne Morris

Loughborough University’s Peter Demian and Kirti Ruikar from the School of Civil and Building Engineering, and Anne Morris from the Centre for Information Management, explain how their new app could help BIM professionals search for information more effectively.

BIM environments allow us to cram more and more information into our building models. Indeed, a central tenet of BIM is the “single version of the truth” – a single information repository (or Common Data Environment) which holds all building information, encompassing all asset life-cycle phases and all stakeholders. This ever-widening scope makes information management challenging, and information overload a serious threat, so maybe it’s time to stop and think.

It’s fair to say that construction professionals are accustomed to dealing with information overload. We don’t even question the number of times a day that we submit queries to web search engines, but what does this operation actually entail? Looking for information on the web means formulating a search query by entering keywords into a search engine; the search engine retrieves relevant URLs from its index, and returns them to the user in a text listing.

Yet when the information we are seeking is linked to a 3D artefact (such as a building modelled in a CAD or BIM environment), there are much more effective ways to express queries, retrieve relevant items and display search results.

The 3D visualisation of the building or geometric 3D data can definitely be exploited more fully. At Loughborough University, we are focusing on this opportunity through the Three-Dimensional Information Retrieval (3DIR) project

We conducted extensive research with construction professionals to ascertain the needs of BIM operators. People deal with information much more effectively when IT tools allow users to think visually and spatially, so 3DIR is exploiting these cognitive strengths to tackle information overload in BIM environments by developing BIM user interface components to make use of the 3D visualisation of a building when formulating search queries and displaying search results.

We are also extending the complex (but now standard) text retrieval computations to include 3D filters. In time, we want to index and allow retrieval by some geometric properties of building components.

As part of this project, the 3DIR app has been released as a free Autodesk Revit add-in (see video above). As an academic research effort, we are acutely aware of the issue of interoperability and aim to remain platform neutral. We would have liked to have developed the 3DIR tools within an open standard such as IFC. However, the Revit Application Programming Interface was found to offer functionality unavailable in more open IFC viewers or modellers.

The app works by creating an index of all text properties and the particular 3D objects in which they occur. It also follows the links in “URL” object properties (ie documents linked to 3D objects) and indexes those documents. This index works in the same way as that created by, for example, web search engines; it allows much more rapid searches than if the system actually searched the model itself each time a query was submitted. The 3DIR app offers multiple querying tools and multiple options for visualising search results.

Since its release in March, the app has been downloaded more than 100 times, but the team at Loughborough is keen for more BIM+ readers to download it, and importantly, to leave feedback in the Autodesk Exchange for Apps.

It’s time to join the search party so that, as BIM repositories get ever larger, your feedback can help the 3DIR team make BIM searching even better than web searching.

The 3DIR app is available for free from the Autodesk Exchange for Apps, via the project website at www.3dir.org/

People deal with information much more effectively when IT tools allow users to think visually and spatially, so 3DIR is exploiting these cognitive strengths to tackle information overload in BIM environments