stilhere4huniid/Harare-Asset-Intelligence-Engine
A dual-mode Real Estate Intelligence Dashboard for Zimbabwe's top developers. Features automated Board Reporting (PDF), Leasing Feasibility Simulations, and Revenue Risk Detection using Python & Streamlit.
πΏπΌ Harare Asset Intelligence Engine
A professional-grade Real Estate Intelligence Dashboard designed for the Zimbabwean market.
This tool serves two distinct profiles:
- Terrace Africa (Operational Defense): Protecting revenue by identifying at-risk tenants using Survival Analysis.
- WestProp Holdings (Development Offense): Simulating leasing velocity and pre-let occupancy for the Mall of Zimbabwe.
π Key Features
- Automated Board Reporting: Generates instant, PDF-ready "Board Packs" for executive meetings.
- Revenue Risk Detection: Uses statistical modeling to flag tenants with "Late Pay" or "Low Footfall" risks.
- Leasing Simulation Engine: A "What-If" simulator that projects occupancy rates if pipeline deals close.
- State Management: Robust session handling ensures data persists when switching between profiles.
- Tenant Mix Analysis: Visualizes sector exposure (Retail vs. Food vs. Services) using Sunburst charts.
πΈ Visual Walkthrough
Scenario A: Revenue Protection (Terrace Africa)
Identifying $273k in monthly revenue at risk across the portfolio.
| Live Dashboard Risk Detection | Automated PDF Risk Report |
|---|---|
![]() |
![]() |
Scenario B: Asset Deep Dive (Highland Park)
Drilling down into specific assets to view tenant mix and lease expiries.
| Asset Level View | Asset Performance Report |
|---|---|
![]() |
![]() |
Scenario C: Development Feasibility (WestProp)
Simulating a 100% conversion rate on the leasing pipeline to secure bank funding.
| Leasing Simulator (Pre-Simulation) | Feasibility Report (Simulated) |
|---|---|
![]() |
![]() |
Scenario D: High-Value Tenant Filtering
Filtering the pipeline for "Blue Chip" Anchor tenants with fit-out budgets > $3M.
| Anchor Tenant Tracker | Executive Budget Report |
|---|---|
![]() |
![]() |
π οΈ Tech Stack
- Core Logic: Python 3.9+
- Frontend: Streamlit
- Data Visualization: Plotly Express
- Reporting Engine: FPDF (PDF Generation)
- Data Manipulation: Pandas & NumPy
- Statistical Modeling: Lifelines (Cox Proportional Hazards)
Harare-Asset-Intelligence-Engine/
β
βββ assets/ # Folder for all your screenshots
β βββ dashboard_terrace_risk.png
β βββ report_terrace_risk.pdf.png
β βββ dashboard_highland_park.png
β βββ report_highland_park.png
β βββ dashboard_westprop_anchors.png
β βββ report_westprop_anchors.png
β βββ dashboard_westprop_sim.png
β βββ report_westprop_sim.png
β βββ ... (other images)
β
βββ venv/ # Virtual Environment (Do not commit to GitHub)
βββ .gitignore # Git Ignore file
βββ app.py # Main Application Script
βββ terrace_africa_v2.csv # Generated Data
βββ westprop_v2.csv # Generated Data
βββ data_generation.ipynb # Renamed from 'Untitled.ipynb' for clarity
βββ requirements.txt # Dependencies
βββ README.md # Main Landing Page
βββ METHODOLOGY.md # The Math & Logic
βββ DOCUMENTATION.md # User Manual
βββ LICENSE # MIT License
π» Installation & Usage
-
Clone the Repository
git clone [https://github.com/stilhere4huniid/Harare-Asset-Intelligence-Engine.git](https://github.com/stilhere4huniid/Harare-Asset-Intelligence-Engine.git) cd Harare-Asset-Intelligence-Engine -
Install Dependencies
pip install -r requirements.txt
-
Run the Data Generator (Optional)
If you want fresh random data, run the Jupyter Notebook:- Open
data_generation.ipynband run all cells to update the CSV files.
- Open
-
Launch the Dashboard
streamlit run app.py
β οΈ Disclaimer
This is an independent Data Science portfolio project created strictly for educational and demonstration purposes.
I am not affiliated with WestProp Holdings or Terrace Africa. All financial figures, rental assumptions, tenant names (outside of known anchors), and construction estimates are hypothetical simulations used to demonstrate financial modeling capabilities. This tool does not constitute professional investment advice, and the creator assumes no liability for decisions made based on its outputs.
π¬ Contact
Adonis Chiruka
Data Science & Financial Modeling
- π§ Email: stillhere4hunnid@gmail.com
- π LinkedIn: Adonis Chiruka
- π GitHub: stilhere4huniid
π License
This project is licensed under the MIT License - see the LICENSE file for details.
MIT License Β© 2025 Adonis Chiruka







