Add Cognitive Manufacturing Environment

src/envs/cognitive_manufacturing/.gitignore

@@ -0,0 +1,50 @@
+# Documentation files (keep in separate docs repo)
+*.md
+!README.md
+
+# Python cache
+__pycache__/
+*.py[cod]
+*$py.class
+*.so
+.Python
+
+# Virtual environment
+venv/
+env/
+ENV/
+
+# IDEs
+.vscode/
+.idea/
+*.swp
+*.swo
+*~
+
+# Database files
+*.db
+*.sqlite
+*.sqlite3
+
+# Data directories
+data/
+exports/
+
+# Logs
+*.log
+server.log
+nohup.out
+server.pid
+
+# OS
+.DS_Store
+Thumbs.db
+
+# Testing
+.pytest_cache/
+.coverage
+htmlcov/
+
+# Temporary files
+*.tmp
+*.bak

src/envs/cognitive_manufacturing/README.md

@@ -0,0 +1,224 @@
+# Cognitive Manufacturing Environment
+
+An OpenEnv-compliant environment for AI agents to control and optimize manufacturing operations with 30 specialized tools across quality management, inventory control, energy optimization, and predictive analytics.
+
+**Author**: Mohammad Mowas | is.mohammad@hotmail.com
+
+## Overview
+
+The Cognitive Manufacturing Environment provides a comprehensive simulation of a production facility where AI agents must balance multiple objectives:
+- **Safety** (highest priority): Prevent failures, overheating, excessive vibration
+- **Throughput**: Maximize production output
+- **Quality**: Minimize defects
+- **Cost**: Optimize operational expenses
+- **Sustainability**: Improve energy efficiency
+
+## Features
+
+- **30 specialized tools** for complete manufacturing control
+- **Multi-machine production line** simulation (4-stage pipeline)
+- **Database persistence** (PostgreSQL/SQLite support)
+- **ML-powered analytics** (predictive maintenance, anomaly detection, quality prediction)
+- **Reinforcement learning** optimization (Q-learning)
+- **Quality management** with defect tracking
+- **Inventory control** and material ordering
+- **Energy monitoring** and power optimization
+- **Scenario simulation** and schedule optimization
+
+## Quick Start
+
+### Installation
+
+```bash
+# Install OpenEnv
+pip install -e .
+
+# Install cognitive manufacturing dependencies
+pip install sqlalchemy sentence-transformers pandas scikit-learn numpy scipy
+```
+
+### Basic Usage
+
+```python
+from envs.cognitive_manufacturing import CognitiveManufacturingEnvironment, ManufacturingAction
+
+# Create environment with all features enabled
+env = CognitiveManufacturingEnvironment(
+    multi_machine=True,       # Enable 4-machine production line
+    enable_database=True,     # Enable data persistence
+    enable_ml=True,          # Enable ML analytics
+)
+
+# Reset environment
+observation = env.reset()
+
+# Take an action
+action = ManufacturingAction(
+    tool_name="ReadSensors",
+    parameters={"machine_id": "M1"}
+)
+observation, reward, done, info = env.step(action)
+```
+
+### Server Mode
+
+```bash
+# Start HTTP server
+python -m uvicorn envs.cognitive_manufacturing.server.app:app --host 127.0.0.1 --port 8001
+
+# Server will be available at http://127.0.0.1:8001
+```
+
+### Docker Deployment
+
+```bash
+docker build -t cognitive-manufacturing:latest -f src/envs/cognitive_manufacturing/server/Dockerfile .
+docker run -p 8001:8001 cognitive-manufacturing:latest
+```
+
+## Available Tools (30 total)
+
+### Basic Control (5 tools)
+1. **ReadSensors** - Get real-time sensor data
+2. **CheckHealth** - Health diagnostics
+3. **AdjustSpeed** - Control machine speed (0-100%)
+4. **ScheduleMaintenance** - Schedule/perform maintenance
+5. **SendAlert** - Send notifications
+
+### Production Line (3 tools)
+6. **GetLineStatus** - Production line status
+7. **TransferMaterial** - Material transfer between machines
+8. **OptimizeLineSpeed** - Multi-machine speed optimization
+
+### Data Management (7 tools)
+9. **SaveProductionData** - Save runs to database
+10. **QueryProductionHistory** - Query historical data
+11. **ExecuteSQL** - Custom SQL queries
+12. **ExportToCSV** - Export data to CSV
+13. **ImportFromCSV** - Import data from CSV
+14. **SearchKnowledge** - Semantic search in knowledge base
+15. **AddKnowledge** - Add documents to knowledge base
+
+### ML Analytics (5 tools)
+16. **PredictMaintenance** - Failure prediction with Random Forest
+17. **DetectAnomaly** - Anomaly detection with Isolation Forest
+18. **PredictQuality** - Quality prediction with Linear Regression
+19. **OptimizeWithRL** - Reinforcement learning optimization (Q-learning)
+20. **ForecastDemand** - Time series demand forecasting
+
+### Advanced Management (10 tools)
+21. **InspectProduct** - Detailed quality inspection (visual/dimensional/functional)
+22. **RecordDefect** - Log and track defects with severity levels
+23. **UpdateQCThresholds** - Adjust quality control parameters
+24. **CheckInventory** - Inventory levels with reorder recommendations
+25. **OrderMaterials** - Material ordering with priority
+26. **UpdateStockLevels** - Update inventory (add/remove/set)
+27. **MonitorEnergyUsage** - Track power consumption and costs
+28. **SetPowerMode** - Adjust power modes (eco/normal/high_performance)
+29. **SimulateScenario** - What-if scenario simulations
+30. **OptimizeProductionSchedule** - Production schedule optimization
+
+## Environment Modes
+
+### Single Machine Mode
+```python
+env = CognitiveManufacturingEnvironment()
+```
+- 1 machine (M1)
+- Basic production simulation
+
+### Production Line Mode
+```python
+env = CognitiveManufacturingEnvironment(multi_machine=True)
+```
+- 4 machines (M1 → M2 → M3 → M4)
+- Material flow through buffers
+- Bottleneck detection and optimization
+
+### Full Feature Mode
+```python
+env = CognitiveManufacturingEnvironment(
+    multi_machine=True,
+    enable_database=True,
+    enable_ml=True
+)
+```
+- All 30 tools available
+- Database persistence
+- ML-powered analytics
+
+## Machine Simulation
+
+The physics-based simulator models realistic machine behavior:
+- **Temperature dynamics**: Heat generation from operation, cooling over time
+- **Vibration**: Increases with speed and wear
+- **Wear accumulation**: Gradual degradation during operation
+- **Failure probability**: Based on temperature, vibration, and wear levels
+- **Production output**: Speed-dependent with quality penalties
+- **Energy consumption**: Power usage based on speed and operating mode
+
+## Reward System
+
+Multi-objective reward calculation:
+- **Safety**: Penalty for dangerous conditions (overheating, high vibration)
+- **Throughput**: Reward for production output
+- **Quality**: Penalty for defects
+- **Cost**: Penalty for operational expenses
+- **Sustainability**: Reward for energy efficiency
+
+## Database Schema
+
+When `enable_database=True`, the environment persists:
+- Production runs with metadata
+- Sensor readings (temperature, vibration, speed, health)
+- Machine events (speed changes, maintenance, alerts)
+- Production units with quality scores
+- Knowledge base with vector embeddings
+
+Supports PostgreSQL and SQLite.
+
+## ML Models
+
+When `enable_ml=True`, the environment provides:
+- **Random Forest** for predictive maintenance
+- **Isolation Forest** for anomaly detection
+- **Linear Regression** for quality prediction
+- **Q-learning** for reinforcement learning optimization
+- **Exponential Smoothing** for demand forecasting
+
+Models train automatically from historical data when sufficient samples are available.
+
+## Architecture
+
+```
+CognitiveManufacturingEnvironment
+├── Simulator (physics-based machine dynamics)
+├── Production Line (4-machine pipeline with buffers)
+├── Database Manager (PostgreSQL/SQLite persistence)
+├── ML Service (5 ML models)
+├── Reward Calculator (multi-objective rewards)
+└── Tools (30 specialized tools)
+```
+
+## Dependencies
+
+- **Core**: OpenEnv framework
+- **Database**: SQLAlchemy
+- **ML**: scikit-learn, numpy, scipy
+- **NLP**: sentence-transformers (for semantic search)
+- **Data**: pandas (for CSV operations)
+
+## Examples
+
+See OpenEnv `examples/` directory for usage examples.
+
+## License
+
+MIT License - See LICENSE file for details.
+
+## Author
+
+**Mohammad Mowas**
+Email: is.mohammad@hotmail.com
+
+Developed as part of the OpenEnv framework for AI agent environments.

src/envs/cognitive_manufacturing/__init__.py

@@ -0,0 +1,34 @@
+"""Cognitive Manufacturing Environment - OpenEnv compliant manufacturing simulation.
+
+This package provides an environment for AI agents to manage and optimize
+a simulated manufacturing system with real-time sensor data, health monitoring,
+and multi-objective optimization.
+
+Phase 0 MVP Features:
+- Single simulated production machine
+- 5 tools: ReadSensors, CheckHealth, AdjustSpeed, ScheduleMaintenance, SendAlert
+- Multi-objective rewards: Safety, Throughput, Quality, Cost, Sustainability
+- Realistic physics simulation with temperature, vibration, and wear dynamics
+"""
+
+from .models import (
+    ManufacturingAction,
+    ManufacturingObservation,
+    ManufacturingState,
+    MachineStatus,
+    Alert,
+)
+from .client import CognitiveManufacturingEnv
+from .server import CognitiveManufacturingEnvironment
+
+__version__ = "0.1.0"
+
+__all__ = [
+    "ManufacturingAction",
+    "ManufacturingObservation",
+    "ManufacturingState",
+    "MachineStatus",
+    "Alert",
+    "CognitiveManufacturingEnv",
+    "CognitiveManufacturingEnvironment",
+]