🚀 RPA Observability Platform

Built a full Automation Observability Platform to monitor, classify, and manage RPA incidents across enterprise environments. An end-to-end Automation Observability & Incident Management Platform designed to monitor, analyze, and manage RPA processes across UiPath environment.

---

📌 Overview

This project provides a centralized monitoring classification layer that:

• Detects failures and anomalies in RPA processes
• Classifies alerts based on business severity
• Generates and manages tickets automatically
• Tracks full alert lifecycle (NEW → ACTIVE → RESOLVED)
• Builds a complete process health intelligence layer
• Feeds operational dashboards (Power BI / ELK)

---

🧠 Architecture

The platform is built around three core engines:

1️⃣ Alert Engine

• Extracts data from:

  • UiPath Orchestrator API / DB

• Detects:

  • Faulted jobs
  • Stale jobs
  • Robot offline
  • Browser errors

• Normalizes alerts into a unified format

---

2️⃣ Ticket Engine

• Applies business rules:

  • Severity (CRITICAL / MAJOR / MINOR)
  • Threshold-based escalation

• Generates tickets automatically

• Tracks lifecycle:

  • NEW → ACTIVE → RESOLVED

• Stores tickets in SQL for persistence

• Produces agent-ready outputs

---

3️⃣ Process Health Engine

• Builds full inventory of all processes

• Enriches with:

  • Business mapping
  • Scheduling logic (cron parsing)
  • Queue activity
  • Execution history

• Calculates:

  • Health score
  • Process state (Healthy / Warning / At Risk / Stopped)

• Outputs datasets for dashboards and analytics

---

📊 Key Features

• 🔍 Real-time alert monitoring
• 🧾 Automated ticket generation
• 📈 Process health scoring
• 🔄 Incremental event processing
• 🧠 SLA-aware classification
• 🗂 Full process inventory (not only active ones)
• ⚙️ Config-driven architecture (no hardcoding)
• 🔗 Multi-source integration (API + DB + Logs) s

---

📂 Project Structure

rpa-observability-platform/
│
├── src/
│   ├── alert_monitor.py        # Alert detection & normalization
│   ├── ticket_engine.py        # Ticket lifecycle & rules engine
│   └── process_health.py       # Process health ETL & analytics
│
├── config/
│   └── config.xlsx             # Environment & system configuration
│
├── data/
│   └── business_mapping.xlsx   # Business severity mapping
│
├── outputs/
│   ├── process_health/
│   └── tickets/
│
├── docs/
│   └── architecture.png        ## 🧠 Architecture

![Architecture](docs/architecture.png)
│
├── requirements.txt
├── .env.example
└── README.md

---

⚙️ Configuration

All sensitive and environment-specific values are externalized:

• SQL Server connection
• Orchestrator API URLs
• Output paths
• Logging directories

Example:

SQL_Server=xxxx
SQL_Database=xxxx
SQL_UID=xxxx
SQL_PWD=xxxx
BASE_DATA_LAKE=./data_lake

---

▶️ How to Run

1️⃣ Install dependencies

pip install -r requirements.txt

---

2️⃣ Run Alert Engine

python src/alert_monitor.py

---

3️⃣ Run Ticket Engine

python src/ticket_engine.py

---

4️⃣ Run Process Health ETL

python src/process_health.py

---

📊 Outputs

The platform generates:

• UiPath_Process_Health.csv
• Queues_Summary.csv
• tickets_to_agent.txt
• tickets_history.csv

These can be used in:

• Power BI dashboards
• ELK stack
• Internal monitoring tools

---

🏢 Real-World Use Case

Designed and implemented for enterprise RPA environments to:

• Reduce incident detection time
• Improve automation reliability
• Provide full visibility over automation landscape
• Enable proactive monitoring instead of reactive fixes

---

🔮 Future Enhancements

• 🔔 Real-time streaming (Kafka / Webhooks)
• 📊 Advanced anomaly detection (ML models)
• 🤖 LLM-based root cause analysis
• 🌐 Web dashboard (Flask / React)
• 🔗 Integration with ServiceNow / Jira

---

👤 Author

Batool Kayed AI Engineer – LLM & Automation

---

⭐ If you find this useful

Give the repo a star ⭐ and feel free to connect!