If you plan within an existing facility, you often work with incomplete CAD data, tape-measure readings, and old drawings. That's a risky foundation. Point clouds give you the actual as-is state, so you can assess collisions, clearances, routes, and utilities realistically. In VR, the hall layout becomes tangible, and decisions are made faster and with more confidence.
With a 3D scan of the hall, you see pipes, beams, doors, and racks where they actually are. How accurate the scan is depends on the method: stationary laser scanners usually deliver the highest accuracy, while mobile LiDAR systems and smartphone scans are faster but coarser. For layout purposes, a point spacing of 5–20 mm is usually appropriate, and a single hall quickly adds up to several gigabytes of data. To make this work in planning, you need the right format, clean preparation, and a data volume that runs smoothly.
Point clouds replace assumptions with measurement. They show the real as-is state, including deviations, extensions, and temporary installations. The difference from CAD: CAD represents the target state in detail, while the point cloud shows reality with its tolerances. By using both together, you make decisions about position, clearance, and collisions early and move into detailed design with CAD later.
In practice, you check aisle widths, beam heights, cable trays, and sprinklers directly in the existing facility. An aisle that is too narrow for material transport stands out as a bottleneck right away. In VR, you try out shifted racks and adjusted floor markings and check the solution at 1:1 scale. For new lines, you compare footprints, utility connection points, and safety distances. Conflicts with cable trays or columns become visible before the first anchor is set.
Route and distance analyses also benefit from the spatial impression. You walk pallet routes, assess visibility at intersections, and move buffers so that oncoming traffic no longer gets in the way. At the workstation, you check grasping heights and reaches directly in the existing facility; adjustments of 100–200 mm can be assessed right away. Because errors come to light earlier, rework, downtime, and costs go down.
In brief: a point cloud is a collection of three-dimensional measurement points from laser scanning, LiDAR, or photogrammetry. Each point carries coordinates and, optionally, color or intensity. Together they form an accurate replica of the as-is state that shows geometry, position, and clearances without surface modeling.
Split large halls into several sections with a moderate number of points so that loading and navigation stay smooth.
The accuracy of a factory scan depends on the sensor and, for mobile systems, also on path length and the stability of position tracking (SLAM). Choose 5–10 mm point spacing where connection dimensions matter and 15–20 mm in areas that only serve for orientation. Work in a local coordinate system and transform to UTM/ETRS89 only when needed. Check units, origin, and axis alignment before every export so that scale and orientation are correct later.
Preparation determines quality. Register the scans using targets or known points and document the residual errors. Remove noise and moving objects that would otherwise show up as false obstacles. Cut away irrelevant zones so that focus and performance are right. Thin out uniform surfaces more heavily and keep edges, passages, and connection points dense. Color values (RGB) help with orientation and communication, while intensity makes edges easier to read.
Data protection and occupational safety apply when scanning: mark off areas, choose quiet times, wear personal protective equipment, and make sensitive content unrecognizable before sharing.
Hidden areas, scan shadows, and uneven point densities are typical limits. Plan additional scans under platforms, behind racks, and above trusses, and close mobile LiDAR paths into loops to reduce drift. A short check before planning starts pays off:
For performance: load only the section you are working on instead of opening the entire hall. Keep the original data separate and work with derived sections.
Halocline is the planning platform for production layouts and workstations. With the Point clouds in VR add-on, you use your scan as a 3D template: you rebuild the as-is state with library elements, take measurements, create variants, and assess them, in VR or on the desktop, with no training required. The desktop view requires a dedicated graphics card.
This is how Schmitz Cargobull did it: the team scanned existing hall areas section by section with an iPhone equipped with a LiDAR scanner, rebuilt the as-is state in Halocline, and compared new arrangements of workstations, work platforms, and material areas directly with the scan. Jona Ostendorf, Industrial Engineer: "For us, Halocline is more than a planning tool – it is the enabler of our digital factory planning. The software has not only made us faster, but above all better."
For downstream processes, you reduce reality to reference geometry: planes for floor, walls, and ceiling, the column grid, openings, and outlines. Sections along aisles or around machine islands provide consistent dimensions. These lightweight references go to CAD or BIM for detailed design. For critical dimensions, add a denser scan or measure again on site. This takes you from quick checking in the virtual replica to clean detailed planning.
Want to plan your hall within the existing facility? Get started for free with Halocline Free and get to know 3D planning. To work with point clouds, you need the Point clouds in VR add-on; get in touch and we'll show you the right scope.
