This presentation outlines our vision on the D-Reader, the product our company develops. The product is focused on automated understanding of engineering drawings.
3. Digital transformation is changing the world by developing some major
technology trends.
Some of these trends are related to the way physical assets are created and
operated.
These trends improve society, create new industries and profoundly impact
major existing industries such as construction, urban development,
housing, transportation, manufacturing, energy, etc.
However, we believe that scaling to the mass market still lies ahead for
those trends and we believe we know what may spark this growth.
4. In all cases, the implementation of these technologies requires obtaining
digital models of the objects. These models must be detailed and precise,
being digital replicas of physical assets. The digital model is now the core
of the object lifecycle.
Creation of these models for the objects designed and built prior to the
digital revolution required a lot of professional labour. Automating this
process and decreasing the manual work required will have a major scaling
effect on those industries.
Automation of digitising assets
5. Key TechnologiesPhysical objects/assets
- Buildings
- Infrastructure
- Industrial facilities
- Industrial equipment
- Transportation facilities
- Tools, parts and components
- Digital modelling and prototyping
- Internet of things (IoT)
- Digital twins
- Smart facilities and infrastructure
- Smart cities
- Building information models (BIM)
6. A digital copy of the asset is at the very centre of the whole lifecycle
1. Design a digital copy in CAD
2. Build an object using applications and smart machinery
3. Use an object by interacting with its digital twin
4. Model its renovation using its digital copy
5. Plan its decommissioning with its model
7. The majority of the older assets’ designs exist only in paper drawing or 2-D
CAD models. Creation of 3-D replicas requires a lot of manual work.
Often the actual objects differ from their initial designs due to changes and
later remodelling. Those also require manual adjustments.
For large objects, that results in substantial time and resource requirement,
which complicates the implementation of new technologies.
2. Current problem
8. A software platform. An AI system combined with analytical algorithms for:
- Reading drawings – defining the symbols and assigning their meanings
- Recognising parts and shapes and connecting them into the model
- Converting 2-D drawings to 3-D models in the latest CAD formats
- Adding information from other sources (photo, videos, 3-D laser scanning)
to reflect the changes.
3. Solution
9. The system will be a cloud-based SaaS platform and used for industrial
solutions (equipment, facilities) and architectural objects in multiple
industries.
Human interaction is still required, but the level of automation increases
with time through further development and machine-learning.
The system will be compatible with major CAD and BIM platforms.
4. Product
10. - Remodelling, renovation, restoration or conversion design.
- Industrial facilities’ upgrade or automation projects. Smart infrastructure
and IoT projects at large facilities.
- Applying building information models (BIM) for pre-built assets.
- Smart city projects requiring digitalisation of all types of properties.
- Digital twin projects for buildings, equipment and cities.
- “Laser Scanning to BIM” projects.
- Engineering of decommissioning projects for large objects.
5. Practical examples
11. The team includes specialists in engineering management, engineers, AI
and programming, all with multiple years of experience.
Positioning as an innovative R&D-focused organisation. Strong
collaboration with research and academia.
“Developer partner” status planned with large CAD software companies.
The general revenue model is “Software as a service” (SaaS) for the
professional market. A service division will provide assistance and generate
additional revenue.
8. Organisation
12. Status of the project
- The project is in the early development stage. A seed round is planned.
- The first MVP is planned by the end of 2020.
- The service group is planned to be deployed after lockdown.
- A library of over 75,000 drawings, 2-D and 3-D models is available as a
dataset for training of AI.
Competition analysis
The system does not yet have direct competitors with an established
product. However, a number of academic and entrepreneurial projects are
attempting to address the same issues all around the world.
9. Status (as of April ‘20)