Point Cloud to BIM Workflow Explained Step by Step

Accuracy is the key to success in the world of architecture, engineering, and construction. However, there are always things unknown about existing buildings. Fortunately, Point Cloud to BIM comes to rescue and helps create a digital clone of the existing building.

Recording of existing buildings is probably the most common task that engineers face when working with constructions. Usually, there are no original blueprints and the available drawings are often inaccurate.

Point Cloud to BIM technology solves this problem. It captures the conditions of the real site with high accuracy. Then, it converts this raw data into intelligent models with Building Information Modeling (BIM).

This technology has a wide range of applications. It helps to document existing buildings before renovation and retrofit. It provides facility management teams with accurate data about buildings and helps to verify clashes in the models.

What is Point Cloud to BIM?

First of all, let us take a look at the definition of point cloud and BIM separately.

A point cloud is a dense group of 3-dimensional points captured by using laser scanners or LiDAR technology. The latest point cloud capturing systems allow capturing up to 2 million points per second. Every single point corresponds to the location on the real physical surface of a building.

Building Information Modeling (BIM) creates an intelligent model from the raw data containing the geometry and other information in a collaborative environment.

However, there is a critical difference between point cloud data and BIM. As we mentioned, point clouds are raw and have no semantic structure. They can’t be used for most of the engineering applications directly. Point clouds are simply a visualization of the scan without any additional information. Therefore, if the point cloud data isn’t converted into an as-built BIM model, this scan won’t have any practical applications.

Step-by-Step Workflow

 Planning the Project

Any successful scan-to-BIM project starts with the definition of the project’s aim. The team should define the necessary target areas, the desired accuracy, and Level of Development (LOD).

Is this project done for a full renovation requiring LOD 350? Or, does the team need information about the general structure of the building and, therefore, the LOD 200 would be enough?

At this step, the team defines the necessary LOD and saves time during the whole process of point cloud scanning.

 Scanning the Site and Capturing of the Data

During this step, laser scanners are mounted on the tripod or aerial drones or mobile scanners. This instrument emits light beams and measures the time to return back. This way, the scanner calculates the distance to a certain physical surface.

Full coverage is crucial for successful modeling. The placement of scanners is a very important step since scanners can’t capture inaccessible places and, therefore, several scanning points are needed.

 Point Cloud Registration

The result of individual scanning is several point clouds, which cover different parts of the site. At this step, the multiple point clouds are registered and aligned into one coordinate system.

It is very important to perform correct registration. Incorrect registration leads to incorrect spatial references. The whole modeling process will suffer from this mistake and will generate incorrect results. This process could be imagined as the assembling of the jigsaw puzzle where each piece should fit perfectly.

 Point Cloud Cleaning and Processing

The raw point cloud data contains noise, duplicate points, moving objects, and irrelevant background data. This step cleans this unnecessary information.

Also, point cloud processing includes the downsampling, which means reduction of the amount of the data while preserving geometric structure. The cleaned and cropped point cloud is ready to be used in BIM modeling.

 Import of the Point Cloud into BIM Software

The processed point cloud data is transferred into the BIM authoring software such as Autodesk Revit. The special software such as Autodesk ReCap Pro helps to translate the point cloud to make it usable in the BIM software.

The point cloud should be aligned to the project levels, grids, and coordinates.

 BIM Modeling from the Point Cloud

This is the stage when the actual BIM model is created. The modelers create the architectural, structural, and MEP elements based on the scanned data. Recently, there are a lot of new approaches, which increase the automation of this process.

The model is created according to the required LOD and project deliverables. For the complicated geometrical shapes, this stage requires a lot of skills and experience.

 Validation of the BIM Model

The validation of the BIM model is a very important step. The BIM model should be validated in accordance with the point cloud to check if there are some errors or missing elements.

Some software allows creating a color-coded heat map, which shows the place where the installed conditions are not matched with the model.

 Deliverables and Handover

There are a lot of deliverables at the end of this project:

  • BIM model.
  • 2D drawings.
  • IFC file for interoperability.
  • Comprehensive as-built documentation.

All of this helps the building owner to prepare the building for further maintenance and renovation. Nowadays, a lot of projects require point clouds as a deliverable for close-out documentation.

Advantages of the Scan-to-BIM Workflow

The advantages of this process include improved accuracy in recording existing conditions, reduced site rework, fast renovation and retrofit planning, improved coordination across disciplines, support of long-term facility management, etc.

Challenges in the Scan-to-BIM Workflow

Of course, there are a lot of difficulties that appear during the process of scan-to-BIM. The incomplete scan coverage is a very common problem. The errors of registration lead to inaccurate modeling. The heavy point cloud data affect the performance of software.

The misunderstanding of the scope or LOD requirements leads to over-modeling or under-delivering. The complicated geometry is difficult to model from the point cloud data.

That is why there are a lot of efforts to develop automated solutions in this field. Recently, a lot of frameworks using deep learning segmentation and unsupervised clustering have been developed . There are some open source tools, which automate wall and slab segmentation and support non-orthogonal geometries.

 

Why Professional BIM Modeling Services Are Important?

The professional BIM modeling services help to keep the process consistent and fast. Skilled specialists know how to deal with raw scan data and choose the appropriate LOD for each project.

Professional BIM services are especially useful in complicated projects and in active sites. The skilled team knows the BIM software ecosystem from ReCap to Revit and to Navisworks.

Conclusion

Point Cloud to BIM helps to connect the physical reality with the digital world of modeling. The described step-by-step workflow allows capturing site conditions, processing the data and creating an intelligent model.

The teams, which follow this process consistently, can build safer and more accurate projects. The scan-to-BIM technology is useful for renovation, facility management and construction. It creates a powerful digital asset for the building.

Point Cloud to BIM Services of Camellia Buildtech

Are you interested in the transformation of your existing building into the digital twin? Camellia Buildtech can offer you Point Cloud to BIM services for commercial, residential and heritage projects. Our specialists will handle everything related to the scanning of the site and delivering the model. Contact us now to talk about your requirements.