���Ǵ���

In recent years, two technological concepts have exploded, almost in parallel: One, called building information modeling (BIM), makes it possible for stakeholders to “build” construction projects in a 3D virtual environment before breaking ground on the jobsite. The other, big data, has become a talking point across industries. It’s used to refer to data sets that have more volume and variety of data generated at a greater velocity than traditional data management tools can typically process.

As BIM technology advances, the data sets powering those models have scaled up, too. And this has given rise to the concept of big BIM data. 

In other words, as BIM technology advances, the data sets fed to create the model have grown, and so have the data outputs created as users interact with that model. That’s moving users into a realm where big data concerns — like data hygiene and governance — become increasingly important. However, investing in strong data practices is worth it because a data-rich model can drive better project outcomes. 

Using BIM To Process Data Faster and Better

BIM can seem complex and even mysterious. In reality, though, it’s a tool the same as a hammer or a chalk line. It’s an instrument that teams can leverage to help them do their jobs better, faster, and more intelligently. 

Take, for example, a complex pharmaceutical facility complete with high purity piping. If the piping contractor makes isometric drawings (essentially, part drawings), they may include dimensional information on the drawings, but the drawings might not necessarily be to scale.

As a result, if the heat trace contractor comes in to create a pricing estimate, they might need to manually go through thousands of pages of those drawings. If, however, that piping is included in the BIM model, the model comes in handy. It can be used to generate a schedule of all of the pipe that requires heat trace in a matter of minutes. 

In other words, the more data that’s in the model, the more useful that model becomes in outputting the data teams need to do their jobs. 

What More Data Does for BIM

As technology advances, the data coming into, verifying and being generated by the model expands. Today, progress on the jobsite might be monitored with laser scanning or reality capture technology. Checking that against the model provides a verification process that ensures that the project is moving toward the planned goals. It also creates additional data that, in many cases, should be captured in the model. The data set grows. 

As the model gets built out with data, it becomes increasingly useful for not just current projects, but also future ones. The model can provide a source of information that estimators can tap to build conceptual estimates, for example. In the absence of other design information, model data from previous, similar projects can help create more accurate estimates. 

It also enables teams to finesse their processes. If, say, they consistently have conflicts with specialty contractors because everyone is vying for space on the jobsite, they can look to the model to help solve for that problem. Deploying the model — ideally, a 4D BIM model that shows project progress over time — allows teams to better schedule so that everyone has the room they need to operate. 

BIM gathers data from various sources in one place. Analyzing the model and the resulting large data set can help companies look for ways to modify their processes to solve their most commonly faced issues. As they leverage the model to cross off problem after problem, they build their competency as a digital builder. 

Optimizing Large Data Sets for BIM

With bad data, stakeholders can’t make good decisions. As data sets grow, it becomes increasingly important to implement good data practices. Otherwise, the model becomes less trustworthy or useful tool. 

Here are a few key areas of consideration for any company engaged in big BIM data.

Data Hygiene

Caretaking the large data set is important because, as the saying goes in data management, "good in" means "good out." The opposite is also true. 

Teams should take care to periodically review and clean their data. That means removing outdated records and duplicates. Say there are two sets of a certain door, for example, one with accurate pricing and another with an old price. The old door data should be removed. 

To maintain good data hygiene, including making it easier to review data and flag potentially problematic data, companies need to establish standardization for their data storage. Any elements that might get pulled into a model, for example, should be formatted and named to best function in the model authoring software.   

For strong standardization, companies might implement externally created guidelines. and the from the National Institute of Building Sciences are both options here. 

Data Governance

Companies dealing with big BIM data also benefit from enacting strong and consistent data governance. This helps to ensure data quality and foster trust in the data from the team using it. It also helps to protect data privacy and build in compliance with any applicable regulations. 

A key part of data governance comes down to determining who can access the data and what they can do with it. Some key questions to ask here include:

• How will tech tools be implemented, and who can use them?
• How will the solutions in the company’s tech stack be integrated with one another? 
• How will data be protected as it flows between solutions?
• Who has access to the model authoring platform?
• With whom will the model be shared, and how?

More granular data governance processes can be implemented as the company becomes more mature in its data governance. Being scrupulous with seemingly simple aspects like folder structure and file naming can drive consistency and, in turn, success.

That gets particularly important as the size of the data sets with which the team is working grows. 

Interoperability

Managing big data sets gets increasingly difficult as the number of software solutions in play grows. As a result, implementing processes that foster interoperability — or that eliminate the need for it — gets more important. 

That might mean implementing a BIM execution plan (BEP) that requires all stakeholders to use the same BIM authoring platform. Or it might mean requiring everyone to export their own portion of the model in industry foundation classes (IFC) formatting so the general contractor (GC) can create a federated model. 

Predictability is key, so having an interoperability plan from the outset goes a long way. 

Data Operators

The data operator is equally as important as the data. Identify people who are qualified and capable of maintaining data hygiene, complying with data governance and boosting interoperability. 

Companies may even choose to define specific agency across many different data operators: BIM managers, virtual design and construction (VDC) specialists, information technology users, or data analysts, just to name a few. As the data sets with which the team is working grows, have dedicated, competent people ready to helm the ship. This helps steer individual projects and the company’s overall trajectory in the right direction. 

A good data operator can reformat and re-curate data based on the end users. This, in turn, can bring more people into the BIM model, engaging more team members with this useful tool. 

Data Curation

Having strong BIM data operators means data curation gets easier. The data might exist in a certain package, but peeling off and reformatting it for various stakeholders means people get the information they need in a way they can digest. This helps to support communication and collaboration on any project. 

Data curation isn’t one-and-done. The data may change through various processes and get handed over to other operators through time. As a result, data operators get tasked with making sure appropriate data is delivered and the messaging is best tailored to the end user. That could mean using the large data set captured in the BIM model to create cut sheets or shop drawings or to pull information for submittals, for example. 

Challenges in Big BIM Data

As BIM data sets grow larger, existing issues expand and new problems enter the scene. Teams should be aware of and ready to manage potential obstacles.

Data Quality

Issues like duplicate data and struggles with version control can compromise data quality. As the data set grows, catching problems gets both more difficult and more important. Strong data governance and good data hygiene practices go a long way here. 

User Pushback

The word "data" can sometimes put people on edge, and some individuals might resist engaging with the data or the model. To help here, reframe the data as information — which is what it is. Help people understand that a bad set of drawings are, in essence, bad data.

The Allure of Flashy Technology

As companies adopt technology, their existing providers, along with salespeople from other tech platforms, might make big promises. They might suggest that “leading-edge” features and artificial intelligence (AI) will solve all of the company’s problems.

It’s important, though, to be aware of what the team actually needs and how potential additions would interface with their existing processes. Adding more technologies, especially when they don’t come with a strategic plan for implementation and integration, often makes a bigger mess.

Imperfections in the Model

Construction is messy, and perfect conditions rarely exist. The model often isn’t 100% accurate. Be honest about where it might have shortcomings. This way, the folks collaborating on using the model can understand what the tool can and can’t do. This helps build trust in the team and the data in the model. 

Difficulty Bringing the Model to the Field

As the data set behind any BIM grows, the model becomes increasingly useful to provide clarity and direction in the field. Still, deploying it there is often difficult, especially as more data-rich models come with issues like slower load times. Fortunately, solutions here range from streamlined BIM readers to augmented reality (AR) headsets that overlay the model with the real world. 

Lack of Clarity Around Owner Expectations

As owners get savvier, they increasingly leverage the model throughout the building’s lifecycle. If they know stakeholders have been using BIM during construction, they may expect a fully fleshed-out model to get handed over at closeout. Have a conversation with the owner about how they plan to use the model so that teams can plan accordingly and build in the right level of detail.

Courses about construction.
For construction.

Unlock your career potential with our free educational courses on Health & Safety, Data in Construction, and more.

Looking Forward With Large Data Sets and BIM

Why is management of BIM and big data important? It has the potential to make every project better — and to better support the people on those jobs. Because it can deliver stronger planning and management processes, it helps projects run better. As a result, the teams in the office and on the jobsite have a better experience. That motivates them to deliver better outputs. 

Data plays a key role in powering the model and, consequently, in supporting these improved outcomes. The people engaging with that data are the important piece, though. Teams benefit from making an effort to curate the data for the end user. Focus on the problem the team is trying to solve and the people who can use data to solve it.

Bring people along for the data maturity journey, too. Invest in education. Pair people who are technologically savvy with people who are resistant to the technology (those are often the same users who have extensive construction experience). Little by little, trust can be gained. And these pairings provide a way for the knowledge from the constructibility expert to be blended into the analysis of the data. This makes the model a truly powerful tool.

Written by