DreamHome AI: A Rule-Based and Constraint-Based Home Construction Plan Designer with Parking Planning and Cost Estimation

Preliminary home planning is usually manual and fragmented: land details, room layout, parking and cost are handled separately by architects, draftsmen and contractors, so every change in requirements forces redrawing and recalculation. This paper presents DreamHome AI, a web-based home construction plan designer that brings requirement validation, 2D floor-plan generation, parking planning and construction cost estimation into one workflow. A constraint-based algorithm validates the user’s inputs, a rule-based spatial allocation algorithm arranges the rooms within the plot, and geometric checks test room dimensions, plot boundaries and overlapping placements. Cost is estimated from the built-up area and a cityspecific construction rate. The application is implemented with a Next.js and TypeScript frontend, a Python FastAPI backend and a PostgreSQL database. A worked example for a 30 ft × 40 ft east-facing plot produces a two-floor, three-bedroom layout and a cost estimate for three quality tiers. The generated plans and estimates are preliminary and must be reviewed by qualified professionals before construction.