A data-driven exploration into the patterns, causes, and insights behind train delays in India, using over 2.8 million data points collected across 3 months and 3,100 trains. From fog-induced disruptions to station-level bottlenecks — discover how data reveals the untold stories of our railways.
- 📦 Dataset: 2.8M+ rows of real-time train status logs
- 🛠️ Tools Used: Python, Pandas, Matplotlib, Seaborn, Frida, BurpSuite, ADB, Linux shell scripting
- 🔍 Scope:
- Delay variation by station, distance, time of day, weekday, and season
- Impact of fog and winter on punctuality
- Identification of high-delay stations and resilient nodes
- 📈 Output: A series of clean, publication-ready visualizations + LaTeX report
| 📌 Insight | 🧠 What We Found |
|---|---|
| Stations with extreme delays | Gumgaon delays 🚨, Virinchipuram early 🚆 |
| Delay vs Distance from Origin | Longer routes ⇒ higher delays |
| Delay vs Time of Day | Peak: 12–16 hrs ⏰ |
| Delay vs Day of Week | Mondays worst, Fridays best |
| Delay vs Month | Jan: 🔥 delays, March: ⛄ smooth rides |
| Fog Impact on Trains | 20,000+ 2hr+ delays in Jan due to fog 🌫️ |
- 📍 Prioritize delay-prone stations for audits
- ⏱️ Reschedule high-importance trains to off-peak hours
- ❄️ Plan fog-safe routing