Skip to main content
Back to challenges
Challenge

Autonomous 'Dark Factory' Orchestration with Claude Opus 4.5 and CrewAI

As global manufacturing shifts towards fully autonomous 'dark factories,' the need for advanced AI orchestration becomes paramount. This challenge involves designing and implementing a multi-agent system that can autonomously manage and optimize operations within such a factory. Participants will create a CrewAI-based team of specialized agents responsible for predictive maintenance, supply chain resilience, production scheduling, and quality control. The system will rely on Claude Opus 5 for its superior long-context reasoning to analyze vast streams of IoT sensor data, historical performance logs, and complex supply contracts. Agents will communicate using an A2A (Agent-to-Agent) protocol, making decisions and executing actions across various simulated industrial systems. A central component will be the integration of the MCP to enable seamless, real-time interaction with enterprise resource planning (ERP) systems, SCADA (Supervisory Control and Data Acquisition), and IoT platforms, allowing agents to retrieve and update critical operational data. This solution will demonstrate the power of generative AI in orchestrating complex industrial processes, minimizing downtime, and adapting to unforeseen challenges in a truly autonomous 'dark factory' environment.

Workflow AutomationHosted by Vera
Sign in free to start
Challenge brief

What you are building

The core problem, expected build, and operating context for this challenge.

Orchestrate an MCP-driven multi-agent system using CrewAI with Claude Opus 4.5 for autonomous 'dark factory' management, emphasizing A2A communication and extended thinking.

Delivery guide

How work is evaluated

Evaluation

The evaluation will assess the autonomous factory system's efficiency, resilience, and decision-making quality under various simulated operational conditions, including equipment failures, supply chain disruptions, and dynamic production targets.

Datasets

Challenge datasets

Public datasets, plus the private datasets you have access to.

Learning goals

What you should walk away with

  • Employ advanced reasoning with Claude Opus 4.1 for predictive maintenance and supply chain optimization.

  • Integrate MCP-enabled tools for real-time interaction with simulated industrial control and ERP systems.

  • Utilize A2A protocol for seamless inter-agent communication and task delegation.

  • Design and implement a role-based multi-agent system for autonomous factory management.

  • Develop a system capable of adaptive planning and responding to dynamic factory conditions.

Resources and assets

Reference links and supporting material

Dataset notes

Sample data for 2 tasks

Challenge setup

The evaluation will assess the autonomous factory system's efficiency, resilience, and decision-making quality under various simulated operational conditions, including equipment failures, supply chain disruptions, an...

Challenge input

{ 'initial_production_targets': { 'product_A': 100, 'product_B': 50 }, 'sensor_data_stream': 'simulated_iot_data' }

Agent execution

No agent tool is listed for this challenge.

Evaluated output

{ 'actual_production': { 'product_A': 'int', 'product_B': 'int' }, 'resource_utilization_report': 'string', 'efficiency_score': 'float', 'actions_t...

Evaluation checks
  • System maintains at least 80% of production targets under normal operations.
  • System generates a feasible and timely mitigation plan for simulated disruptions.
  • Demonstrates clear evidence of inter-agent communication for task coordination.
Configured evaluation evidence
  • ProductionEfficiency target: 0.85
Configured runtime
  • Python execution harness

Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.

View technical recipe

Configured tools

No tool records are attached.

Evaluation contract

  • System maintains at least 80% of production targets under normal operations.
  • System generates a feasible and timely mitigation plan for simulated disruptions.
  • Demonstrates clear evidence of inter-agent communication for task coordination.
  • Successfully uses MCP-enabled tools to interact with simulated ERP/IoT data.

Metrics

  • ProductionEfficiency · Target 0.85
  • AdaptiveResponseQuality · Target 4

Workflow tasks

  1. Task 1

    Input: { 'initial_production_targets': { 'product_A': 100, 'product_B': 50 }, 'sensor_data_stream': 'simulated_iot_data' }

    Output: { 'actual_production': { 'product_A': 'int', 'product_B': 'int' }, 'resource_utilization_report': 'string', 'efficiency_score': 'float', 'actions_taken': ['string'] }

    Execution harness: Python

  2. Task 2

    Input: { 'current_state': 'factory_snapshot', 'event_trigger': { 'type': 'equipment_failure', 'details': 'Machine X failure' } }

    Output: { 'incident_report': 'string', 'mitigation_plan': ['string'], 'recovery_time_estimate': 'string', 'impact_assessment': 'string' }

    Execution harness: Python

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Autonomous 'Dark Factory' Orchestration with Claude Opus 4.5 and CrewAI