Evaluate and categorize plans with artificial intelligence.
About the product
This product, realized jointly with our partners, the Institute for Enterprise Systems of the University of Mannheim (InES), Loidl GmbH, The Chainless GmbH and bwcon research gGmbH, enables the representations and information of a construction plan to be recorded and evaluated. With the help of the web-based “virtual planning office” and a specifically further developed, artificial intelligence – the “BauKI” – construction plans can be provided more promptly and categorized more extensively than with conventional working methods.
Task
In the construction industry, thousands of plans are created and used on a daily basis. With the help of structured data rooms and recorded additional information, the effort should be reduced. The collection of additional information, which in practice can be time-consuming, error-prone, incomplete and highly user-dependent in terms of content, is to be optimized and accelerated. And the findability and use of the plans should be made possible for all project participants without the additional help that is otherwise often required.
Functionality
- The BauKI can differentiate between plan content and plan header in a plan.
- It recognizes the type of representation of a plan in the plan content. Information such as description, scale, index, etc. is read out in the plan header. The position of the plan header in the plan is not relevant.
- Plan recognition also works with scanned plans, which means, for example, that paper plans can be recorded and evaluated.
- The construction AI evaluates the provided plans individually and transfers the read metadata back to the platform.
- The platform then maps the metadata to the appropriate plan.
- The BauKI can also be integrated into other systems as a white label solution.
Visualization
- The Gradcam visualization makes the areas recorded and analyzed by BauKI visible.
- The BauKI recognizes, for example, whether it is a view and/or a section.
- If several representations are recognized, the analyzed plan is assigned several times accordingly.
Benefits
- With the “Virtual Planning Office” – a web-based platform – users have access to the management of required plans.
- The BauKI analyzes uploaded plans and then transmits the results of their categorization and evaluation.
- This results in the following advantages:
- Central storage location for all plans.
- Accessible to everyone involved at all times.
- Fast entry of large planning inventories.
- Categorizations of several plan types.
- Evaluations in ongoing projects.
- Significantly lower error rates than with manual evaluations.
Development partner
- The Chainless GmbH : Know-how in the field of deep learning-based classification, regression and semantic segmentation tasks, as well as in cluster methods of high-dimensional data and in the development of metric learning methods (few-shot) to reduce training complexity .
- Loidl GmbH : Role of consultant in the field of digital construction. In connection with the increased requirements for the realization of commercial buildings, Loidl provides his expertise as a data, plan and project manager in the realization of large and very large construction projects.
- University of Mannheim – Institute for Enterprise Systems (InES) : The goals of the InES are to achieve high-quality research results while pursuing an interdisciplinary research approach and the transfer of knowledge into practice with a focus on manufacturers and users in the SME and enterprise sector, as well as the qualification of the scientific offspring.
- bwcon research gGmbH : The non-profit research organization, which in principle does not pursue any economic interests, has the appropriate methods and experience to successfully coordinate complex projects. In addition, she provides training in thinking in data-based business models as well as associated AI experts.
Sponsor
- This product development was part of the “AI Innovation Competition Baden-Württemberg” from
Funded by the Baden-Württemberg Ministry of Economics, Labor and Tourism .