Scope of Work
The development scope encompasses six core system modules, comprehensive AI model training, and full-stack integration. The platform will be built using modern web technologies with a focus on scalability, security, and user experience. All modules will be developed with API-first architecture to ensure seamless integration with existing Sela systems and future extensibility.
Core System Modules
The Sela Estimation Platform is built on six interconnected system modules that form a comprehensive ecosystem for construction cost estimation. Each module serves a distinct purpose while maintaining seamless data integration across the platform, ensuring accuracy, efficiency, and scalability throughout the estimation workflow.
System Modules:
Dashboard
Project overview with filtering and search
Smart Estimation
AI-powered cost estimation engine
Data Lake
Centralized file repository for the received RfPs
Knowledge Hub
Library for Market Benchmarks
Financial Analysis*
System Integrations**
- Oracle Integration: Valuable historical data such as quotations will be pulled out manually through document controllers and added to the AI system for training purposes.
- CostX Integration: Integration might be limited based on the version that Sela has. Latest version 7.2 is required. The integration will be based on the available public API endpoints provided by CostX.
Development Approach
The platform employs an API-first architecture where all business logic is exposed through well-documented RESTful endpoints. This approach ensures that each module can operate independently while maintaining seamless integration with the broader system ecosystem.
Database architecture is designed with normalization principles to ensure data integrity, with dedicated schemas for estimation data, user management, project archives, and market intelligence. All database interactions are optimized for performance with proper indexing and query optimization.
Frontend development follows component-based architecture using modern React patterns, ensuring reusability and maintainability. The UI framework is built with responsive design principles to support desktop and mobile access across all user roles.
AI Drawing Analysis Limitations
Current AI technology has limitations in comprehensive drawing interpretation. The platform's drawing analysis capabilities will be limited to extracting basic information such as project type, building dimensions, floor count, and general layout characteristics. This information will be helpful in determining the appropriate Sela Class of accuracy for the estimation approach.
The system will not calculate detailed cost estimations directly from drawings. Instead, it will use the extracted basic information as supplementary input alongside other data sources (historical projects, market benchmarks, RfP specifications) to generate comprehensive cost estimates through the AI estimation engine.
Key Implementation Notes
Data Integration Critical
Success depends on early data integration and structured data availability
Ongoing Testing
Continuous testing and validation throughout implementation to ensure optimal performance
Historical Data Training
AI models trained on Sela's historical project database for accurate estimations
Success Factors
→Consistent and organized input data ensures accurate system outputs
→Early data integration accelerates implementation and improves precision
→Integration performance depends on data availability and API capabilities