Skip to main content
All CollectionsConstructionUse Cases
How to identify and manage deviations in construction
How to identify and manage deviations in construction
Updated over a week ago

If you are using Sitemark's Solar Construction Data Product, our platform offers a powerful solution for automatically detecting deviations between the as-built structures and your original design.

By leveraging Sitemark's advanced deviation detection capabilities, you can ensure higher accuracy and quality control in your construction projects, leading to reduced rework, lower costs, and faster project completion times. Our platform's sophisticated algorithms and comprehensive data analysis tools provide you with the insights needed to make informed decisions and maintain the highest standards in your solar construction projects.

Currently, automatic deviation detection is available for piles. To use this feature effectively, it is crucial to ensure that all your design components are imported into the platform. Click here to learn more about how to set up your solar construction site.

Sitemark's system detects three primary types of deviations:

Unexpected Components

Components that are located more than 1 meter away from any design component will be flagged as unexpected.

Misplaced Components

Components detected within a range of 45 cm to 1 meter from their designated design locations will be marked as misplaced. As we continue to enhance our AI models, this threshold will be reduced from 45 cm to 20 cm over the coming months, allowing for even more precise detection. However, due to technical limitations inherent in the photogrammetry process, deviations of less than 10 cm cannot be reliably detected.

Missing Components

If a component is not detected, but its surrounding components are present, our system will label it as missing. This detection is powered by a proprietary advanced algorithm that analyzes the spatial relationships between design components and their detected counterparts. By cross-referencing the locations and ensuring that a certain number of neighboring components have been detected, the algorithm identifies missing components.


To see deviations in action and learn more about how our platform can benefit your projects, watch our demonstration video below:


​


​
​


​

Did this answer your question?