Logo Lanfrica

Aml-Asd/Tazkarti--Expert--Dashboard

Domaine:

mobility

Type de record:

softwaremodel
Créateur:
Aml
Hôte:
Egypt Transport AI — Discrete Event Simulation & ML delay-prediction dashboard for Upper Egypt road routes, built with SimPy, scikit-learn & Streamlit. # 🚌 Egypt Transport AI — Simulation & Prediction Dashboard A Discrete Event Simulation (DES) system combined with an ML delay-prediction pipeline, modelling high-traffic public transport across 4 Upper Egypt road routes. Helps administrators find the optimal number of servers/buses to minimise waiting times and user abandonment. --- ## 🗺️ Routes Covered | Route | Location | Base Headway | |---|---|---| | Route 1 | Western Desert Road (Edfu) | 120 min | | Route 2 | Qena–Luxor Agricultural Road | 300 min | | Route 3 | Aswan Eastern Road | 60 min | | Route 4 | Marsa Alam Coastal Road | 540 min | --- ## ✨ Features - **DES Engine** — SimPy-based Poisson arrival modelling to simulate realistic traffic load - **ML Delay Predictor** — scikit-learn pipeline trained on 300 real route records, serialised with joblib - **Dual-Portal Streamlit Dashboard:** - 👤 **Commuter Portal** — compare live delay status across all 4 routes, get the fastest recommendation, view route map - 🏢 **Ops Command Centre** — KPI metrics, What-If headway simulation, projected delay reduction bar charts - **SLA Validation** — automatically checks if a configuration meets wait-time and abandonment thresholds - **AWS Cost Modelling** — estimates hourly infrastructure cost per configuration --- ## 🛠️ Tech Stack Python · SimPy · scikit-learn · joblib · Streamlit · pandas · matplotlib · seaborn · Queuing Theory --- ## ⚙️ Setup ```bash git clone github.com cd Tazkarti--Expert--Dashboard pip install -r requirements.txt # Place final_transport_model.pkl and Cleaned_Data_300_Rows.csv in the project root streamlit run main.py ``` --- ## 📊 ML Pipeline ``` Input features: hour, passenger_count, weather_severity, pressure_index, delay_range ↓ scikit-learn Pipeline (preprocessing + regressor) ↓ Predicted delay (minutes) per route ↓ Status: 🟢 Smooth (<15 min) | 🟡 Moderate (<45 min) | 🔴 Severe (≥45 min) ``` --- ## 👩‍💻 Author **Aml Abdelrhman Ahmed Mohamed …