A CAD automation system for building design that accelerates the generation of pre-design documentation.
Challenge
Large amount of routine work is performed while preparing technical assignment for designers of major repairs or construction of standard buildings. For example placement of equipment in rooms according to industry standards. When there are many buildings and their plans are similar, design staff’s eye becomes blurred, which leads to large number of errors, increases the number of iterations of the exchange of construction documentation between customer and designers.
Task
To automate the formation of technical specifications for designers, avoiding monotonous work, speeding up creation and reducing errors in the final technical specifications.
Solution
Robot& team developed an AI-based solution that generates technical specifications for designers based on a building plan. The customer uploads building plan in CAD format and receives technical specifications in the following format:
- Building plan (CAD file) with equipment automatically distributed among rooms according to industry standard: WiFi access points, SIP phones, RJ-45/USB/HDMI sockets.
- XLSX file with a list of hardware needed to equip this building. This information is immediately used for budgeting and planning the purchase of equipment.



What's under the hood
Parsing and saving CAD files
Parsing of room explication
AI model of semantic analysis for classifying rooms
Equipment distribution algorithms in accordance with the industry standard. Requirements for equipping rooms with this or that equipment are applied to it depending on the class of the premises.
Export to XLSX format of a complete list of equipment for budgeting, planning purchases and setting tasks for designers.
We are engaged
in innovation in creation
and taming of artificial intelligence
in innovation in creation
and taming of artificial intelligence
Team of eight worked on the project
8
Project lead
Anatoly Zakharov

Frontend development
Timofey Torgaev
QA
Alexander Tsoi
DevOps
Ilya Kostyuchenko
Backend development
Alexander Kvashnin
Anatoly Zakharov
Denis Fomin
Analytics
Gleb Sapunkov
Design
Tatyana Rapanovich



