Skip to main content
Loading navigation
Back to challenges

Plug-and-Produce MTP Module Orchestration with AutoGen and Arcee AI

Integrating modular process units via Module Type Package (MTP) standards requires automated semantic matching between skid capabilities and plant recipes. Build a multi-agent orchestration team using AutoGen and Arcee AI to parse MTP manifest XMLs and configure process topologies. Achieve 95% recipe execution fidelity across heterogeneous process modules.

EngineeringHosted by Vera
Sign in free to start

What you are building

Configure AutoGen conversational agents powered by Arcee AI specialized models to parse MTP XML descriptions and auto-assemble modular process plants.

How work is evaluated

Evaluation

Evaluates multi-agent group conversation accuracy in mapping recipe demands to available MTP process modules.

Datasets

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

How submissions are scored

Dimensions
2 scoring checks
Binary
2 pass or fail dimensions
Ordinal
0 scaled dimensions
Dimension 1

sequence_completeness

Verifies every recipe step has a valid MTP service mapping.

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Dimension 2

recipe_execution_fidelity

Fidelity ratio of correctly mapped MTP parameters to total target recipe steps. • target: 0.95 • range: 0-1

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Learning goals

  • Configure AutoGen multi-agent roles: MTP Descriptor Agent, Recipe Allocator, and Safety Inspector

  • Parse VDI/VDE/NAMUR 2658 MTP XML manifests to extract services and OPC UA interfaces

  • Use Arcee AI models to match plant process recipes with candidate module capabilities

Resources

Dataset notes

Standardized NAMUR MTP XML manifests representing dosing, mixing, and thermal process skids.

Challenge setup

Evaluates multi-agent group conversation accuracy in mapping recipe demands to available MTP process modules.

Challenge input

JSON with recipe requirements and list of MTP manifest XML strings

AutoGen

Multi-agent conversational framework

Arcee AI

Domain-adapted specialized model host

Evaluated output

JSON execution sequence detailing mapped service_id, module_id, and OPC UA endpoint

Evaluation checks
  • Verifies every recipe step has a valid MTP service mapping.
  • Fidelity ratio of correctly mapped MTP parameters to total target recipe steps. • target: 0.95 • range: 0-1
Configured evaluation evidence
  • Benchmark: Modular Process Automation Optimization
  • Recipe Execution Fidelity target: 0.95
  • 1 public reference case
Configured runtime
  • Python execution harness
  • Python sandbox (unavailable on Versalist)

Shows the first task, up to three tools, and up to three evidence items.

View technical recipe

Configured tools

Action Space
  • AutoGen · Required
  • Arcee AI · Optional
  • Microsoft AutoGen · Optional

Evaluation contract

  • sequence_completeness · Weight 1
  • recipe_execution_fidelity · Weight 1

Metrics

  • recipe_execution_fidelity · Target 0.95

Benchmarks

  • Modular Process Automation Optimization
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON with recipe requirements and list of MTP manifest XML strings

    Output: JSON execution sequence detailing mapped service_id, module_id, and OPC UA endpoint

    Execution harness: Python

Evaluation tools

  • AutoGen · framework
  • Arcee AI · ai-workflow-automation

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Plug-and-Produce MTP Module Orchestration with AutoGen and Arcee AI