Introduction

The decade since the publication of the Government Construction Strategy in 2011, revealing the plan to require a fully collaborative 3D BIM on projects, marks a period of rapid shift towards digitalisation within the AEC sector, leading to the development of the ISO19650 suite of international standards, that aim to unify the information management approach when utilising BIM on projects.

During that period, BIM adopters have been seeing benefits increasingly, such as improved coordination and reduced risk, as reported in a series of surveys published in the annual NBS BIM reports, which document the level of BIM adoption over time.

While many organisations have a unique approach when adopting and implementing BIM standards, a survey in the latest annual NBS BIM report found that 38% of over 1000 respondents are using Uniclass 2015 to classify information. (https://www.thenbs.com/knowledge/national-bim-report-2020) Uniclass 2015, which is developed and periodically updated by the NBS, is the UK implementation of a classification system aligned to the requirements of ISO12006-2:2015 (Building construction — Organization of information about construction works — Part 2: Framework for classification).

So, how is this significant? Well, information requirements can vary largely between different clients, depending on their whole life asset management strategies as well as the maturity of their digital evolution. In my experience, I have come across clients who are still evolving and learning as they do, as well as clients who have very well-established asset information requirements. I also come across clients who are still in the early days of digitalising their processes and needing a good starting point when defining asset information requirements.

Regardless of maturity, utilising a unified classification system, which describes BIM model elements consistently in a common language, offers a lot of potential when it comes to specification, scheduling, cost estimation and asset management. One clear benefit from utilising Uniclass 2015 as such, would be to make use of The NBS BIM Toolkit. Initiated by the BIM Task Group and funded by Innovate UK, The NBS BIM Toolkit is a free tool for defining and managing project and asset related information.

The NBS Toolkit

The toolkit includes a library of over 5700 construction definitions for specifying Level of Detail and Information which is a key ISO19650 requirement. Users can access the library online at https://toolkit.thenbs.com/definitions where the information is available freely, and search the database for activities, spaces/locations, elements/functions, systems and products. An additional benefit which is built into the toolkit would be the use of the ‘Manufacturer Product Data Templates’ available for system and product definitions. The ‘Manufacturer Product Data Templates’ are a series of electronic spreadsheets, customised to the requirable LOI datasets for built assets, based on their classification code. The templates provide the supply chain with an easy format to work with when providing asset information on a BIM enabled project.

The Question (Presenting problem)

It is certainly worth exploring the benefits of utilising this well-defined resource when specifying product or system asset information requirements. However, have you wondered what the definitions contained within the database represent in their entirety, in an aggregated view and what that means for a project or across a portfolio? Along these lines, wouldn’t it be helpful to have the ability to zoom out and have a bird’s eye view of the requirable asset datasets on a project as early as the design stages?

Such questions can be particularly relevant to asset owner/operators when putting strategies in place for managing the whole information lifecycle and to facilitate change management during trigger-related events. The maintenance of the rich information model that is generated on a BIM project begins at handover and many considerations need to be made about its management. With this in mind, I set out to explore the database to gain a better understanding on its complexity to arm myself with the ability to predict and plan for the overall asset information model generated on projects which are aligned to the toolkit’s definitions and LOIs. My exploration focusses on definitions for items listed in the Uniclass Product (pr) and Systems (Ss) tables in particular.

The Workings: An aggregated view of the Uniclass Product (Pr) and System (Ss) definitions and Level of Information

The NBS Toolkit library comprises a series of tables nested within a well-structured set of URLs. By well-structured, I mean that the webpages follow a consistent naming approach. For example, the LOI5 datasets specified for Pr_25_57_51_51 can be accessed on https://toolkit.thenbs.com/definitions/Pr_25_57_51_51/?type=loi&detailLevel=5

By simply replacing the product (or system) Uniclass code with another, the table with information specified at LOI5 for the new product/system can be accessed. Similarly, by simply replacing the last character specifying the Level, ‘5’ in this example with 2,3,4, or 6, the tables with the datasets the respective LOIs can be accessed.

The data can be fed into Power BI to blend it into an interactive dashboard, together with a list of classified digital asset deliverables on a project, providing a summary of the definitions and datasets specified at each LOI for and an opportunity to evaluate compliance of the delivered information against the NBS model.

The results were surprising in several ways. Mostly, by the sheer volume of requirable unique datasets as well as the lack of definitions for many. It is key for asset owner/operators to have an efficient strategy about housing and managing the digital assets acquired on projects in a structured way to be able to maximise their potential during FM.

To review the data, I used a series of calculated tables, columns and DAX expressions in Power BI, looking at hierarchies and relations between the different data feeds. The overall volume of requirable product and system data fields was unsurprisingly substantial, resulting in a count of 89,760 system datasets and 310,642 product datasets.

Something else that became apparent from this exercise was the number of Uniclass products and systems for which no definition exists as of the time writing this article. Although, this is not necessarily an obstacle, extra care should be taken when specifying information aligned to the toolkit to check whether this might lead to any gaps on your projects. It is best to do this in advance, to avoid any ambiguity and frustration amongst supply team members later in a project when teams are working at their full capacity.

The next finding that surprised me, was the sheer volume of unique asset datasets that could be generated on a project. If you consider BIM as just one component within the wider digital ecosystem of a client’s organisation, it’s primary use would be to serve as a base feeding data into other platforms, such as CAFM, BMS or Building Advisory systems, providing the ability to rationalise that data into useful information to drive the decision making process and improve our knowledge. So, it is important to predict the volume of data deliverables when deciding their format and future lifespan. Also, with this sheer volume in mind, think about whether you might be falling in the pitfall of over-requiring information more than your needs. We have all heard the saying that no information is better than too much information, which becomes the reality when a client is faced with a substantial volume to data to make sense of at the end of a project.