Europe Union

OkKast

Vision System for Industrial Precast Concrete Quality Control.

Automated AI-based system for measuring and detecting defects in prefabricated concrete elements.

A picture of OkKast, an app that does automated quality control for precast concrete elements

Automate Precast Concrete Quality Inspections with OkKast

OkKast is a comprehensive precast concrete quality control solution that combines industry-standard cameras with an AI algorithm that is trained to detect the features of precast elements. During the preparation of precast elements, it recognizes and signals any inconsistencies and stores the data for production documentation purposes.

Dimension measurement

Embedded elements detection

Surface defect detection

Digitalized snag list

Why you should introduce OkKast to your production process right now?

Standardise
quality control

Reduce waste
and recasting costs

Scale production without
scaling inspection teams

Stop shipping
defective products

Save time
on every casting cycle

Cut down on human
error and oversight time

Minimise the number
of complaints to zero

Prove quality with
visual documentation

Free up skilled labour
for critical tasks

a picture from OkKast app that detected a crack in precast concrete
an image showing a tablet with a dimensional measurement of precast concrete which is created automatically with computer vision

How OkKast Works?

The OkKast system comprises a camera, an algorithm and an application.

The camera is usually mounted on the ceiling of the production hall, capturing the entire casting bed while avoiding other elements of the production line. Depending on the production line’s height, lighting and other environmental conditions, our specialists select the appropriate camera. The aim is to obtain an image resolution that enables the measurement and identification of prefabrication parameters. When properly calibrated, the camera can achieve an accuracy of up to 5 milimeters at a height of 10–12 meters above the working area.

The camera image is transferred to the machine learning algorithm, which detects the prefabrication parameters. At the same time, the system downloads information about the element from your BIM system to compare the parameters with those of the project. If there is a discrepancy, the application informs the quality controller.

Additionally, OkKast creates a list of prefabrications per ID taken from the BIM system and stores photos and camera recordings in folders, providing visual documentation for each prefabrication.

Richard Kowalski
Richard Kowalski Precast Concrete Consultant
What Experts Say About OkKast
“AI is not just another technology layer. It is a strategic enabler. AI-powered visual inspection and real-time process adjustments yield near-zero defects in the produced elements. This is not a wish or prediction. This is a proven fact. This is real Construction 4.0. Try it in your factory.”

Without OkKast

  • Difficulty in aligning on-site elements with BIM data
  • Limited traceability of quality control issues
  • Manual quality control for precast elements

With OkKast

  • Automated dimension measurement and defect detection, calibrated to your production process
  • BIM integration for precise quality control of precast elements
  • Photo repository for centralized storage of documentation

Where to Start?

The Okkast system is flexible and can be adapted to meet your specific requirements.

Every production line is different. While OkKast has a set of core features and capabilities ready for implementation, we also understand the importance of flexibility and adapting the system to your production process’s unique needs.

For this reason, we will only define details such as which prefabrication parameters are important to you, the expected measurement accuracy, the exact camera placement and the integration with your existing systems after we have met and discussed your requirements.

See It Live. Get a Customized Demo.

Schedule your personalized demo by filling out the form below.

Contact us!

Send us an email: [email protected]