Recently I attended the AGC BIM forum meeting in Boston (before the Red Sox won the World Series for the second time). There were many presentations about challenges and experiences with the current use of BIM in the AEC industry. The meeting was attended by architects, contractors, fabricators, subcontractors, and even lawyers.  One of the most discussed topics at the forum was the challenges of the modeling process.  Building the right model for coordination, estimating and scheduling at different stages of the construction process is not an obvious task.

I have seen three different approaches in the industry. The model can be created by the designer, in house by the contractor or using a service provider. The creator of the model can be defined by the project type, the delivery method or the contract, and finally the use case of the model and its intended purpose. It can also shift from one team member to the other depending on skills available in the various organizations.

When the model is created by the designer there is usually a very close partnership and trust established between the designer and the contractor.  Sometimes a sophisticated client will request the architect and the contractor to share information and adopt a model based approach. The model is mostly used for coordination by the design team, engineers and the contractor. Without clear specifications for modeling, the process can break down very easily and create friction within the team. The disadvantage of this approach is that although the model might represent the design well, it might not represent the way the building will be constructed.  Additionally, different Level of Detail (LoD) are frequently used by designing when not modeling to a spec.  In these cases, the contractor often has to spend more time adjusting the model for downstream use than it takes to model the building from scratch.  Although it seems counterintuitive that it could take longer to edit something than to recreate it, we have found this be true in our own internal tests when the models weren't created according to predefined specifications. The advantage of working together with a predefined spec is that there is a closer partnership between project team members and less rework.  Undoubtably, this approach will ultimately prove to be the most efficient, but it is also the most difficult way to begin, particularly if both the designers and contractor do not have the experience with modeling to know how a spec would need to be designed to fulfill their collective requirements.  I recommend that companies develop their own expertise first and then work on collaboration.

Modeling in-house requires hiring new people or finding the right ones within the organization. Many of our larger clients were able to set up BIM enabled preconstruction or construction teams. The advantage of this option that contractor can develop modeling as an in-house skill, and be more competitive in providing preconstruction services for example. The disadvantage is that it is expensive to hire and train people and takes more time to get started with BIM, than using models that others create.  Ultimately, I believe all firms will need in-house expertise in modeling even if other parties create the model.  The two primary reasons are that the firm will need to extend the model to fully suit their purposes and to be able to perform what-if scenarios and respond to change orders quickly.

Modeling as a service is available today from multiple suppliers. Modeling as a service could is perceived to cost more but can be part of the project budget and the cost of the model can be justified through the direct benefits it provides to the contractor and the owner. There are three different use cases for a model: Modeling for Coordination, Modeling for Estimation and Modeling for Scheduling. The contractor can decide which use case they would like to focus on and gain experience in using the model. Deliverables usually are ready in a very short period of time. I also believe that services should be combined with knowledge transfer, and I found it an efficient and relatively painless way to get started even on more complex and demanding projects.

Eventually, I believe the industry will achieve the vision of smooth dataflow between parties, but I think many current efforts at achieving the benefits of BIM are failing because the participants are trying too hard to solve the dataflow problems and losing sight of the key objectives and benefits.  Companies need to start working with models as any of the three approaches described above produce significant benefits.  Starting with services, moving to internal modeling, and ending with collaborative modeling seems to be the easiest path to change the traditional practices for the vast majority of firms.  Even firms that have begun with collaborative modeling have often not spread the process to a majority of their projects that could have benefited from modeling.  In other words, a staged approach makes it easier to spread the benefits to a majority of projects and a company may have projects using all three stages as they transform their internal processes and supply chain.

All options above require the design team, the contractor and the service provider to develop a process for creating the model and other deliverables and set up protocols for collaboration. Developing the process and implementing it at the same time are daunting tasks. It is not simple to develop a process that facilities fast and efficient modeling that integrates estimating and scheduling data with the model and can be used for coordination. The information captured in the model has to be well structured, and documented properly. This is the first step towards a more efficient flow of information and a more predictable construction process.

The question today is not whether you start or not, it's more like when and how?