What challenges did JM Family Enterprises face before AI implementation?

Industrial automation is increasingly complex, with multi-step, interdisciplinary processes that exceed the capabilities of single AI agents. Need to coordinate, orchestrate, and scale multiple specialized AI tasks across manufacturing lines, compliance, customer support, and more. Existing processes can be bottlenecked by manual coordination, lack of modularity, and slow adaptation to changing requirements. Organizations face challenges integrating AI across disparate systems while maintaining reliability and control.

What was the impact of the AI solution?

Streamlined production and QA processes, reducing manual intervention and improving reliability. Accelerated software development life cycles for enterprise IT through collaborative agent ecosystems. Optimized bottling line performance, root cause detection, and predictive analytics in manufacturing. Established scalable, reusable AI agent architectures that adapt to evolving process requirements.

Source

Microsoft

Manufacturers automate production with multi-agent AI systems

Use case typeWorkflow automationUpdated Jun 13, 2026

Several manufacturers and Microsoft partners are leveraging the Azure AI Foundry Agent Service to develop multi-agent AI systems that automate and optimize production processes. The Azure AI Foundry Agent Service introduces capabilities such as Connected Agents, Multi-Agent Workflows, and Agent Catalog, allowing the orchestration of specialized AI agents for complex industrial use cases. Early adopters have implemented collaborative ecosystems of agents across tasks such as bottling line optimization, compliance monitoring, software development lifecycles, and enterprise customer support. These modular systems allow organizations to coordinate multiple AI models, reuse specialized agents, and integrate external tools or protocols to build scalable, resilient digital workforces. Real-world examples include companies like JM Family Enterprises streamlining software QA, and Sight Machine improving bottling line performance. Partner NTT DATA is highlighted for orchestrating complex deployments. Results include improved productivity, reduced manual intervention, measurable business analysis improvements, and potential cost efficiencies. The solution addresses the complexity of real-world industrial workflows, providing the ability to break tasks into modular, role-specific AI agents that can collaborate or operate in parallel. A2A and Model Context Protocol support offers interoperability across platforms, and an Agent Catalog accelerates deployment and adaptation.

Organization: JM Family Enterprises
Industry: Manufacturing
Location: United States
Published: May 2025

Reported outcomes

Strategic outcomes

New product / capabilityAutomated production and QA processesSpeed & agilityAccelerated software development lifecyclesBetter decisions & insightImproved root cause detection and analyticsScale & capacityEstablished reusable agent architectures

Primary read

Use case focus

Showing 3 of 3

1Multi-Agent Orchestrated Manufacturing Optimization
2Automated Quality Assurance with Agent Collaboration
3Scalable Agent Ecosystem for Industrial Automation

Industrial automation is increasingly complex, with multi-step, interdisciplinary processes that exceed the capabilities of single AI agents.
Need to coordinate, orchestrate, and scale multiple specialized AI tasks across manufacturing lines, compliance, customer support, and more.
Existing processes can be bottlenecked by manual coordination, lack of modularity, and slow adaptation to changing requirements.
Organizations face challenges integrating AI across disparate systems while maintaining reliability and control.

Implemented Azure AI Foundry Agent Service with Connected Agents and Multi-Agent Workflows.
Used no-code and SDK tools to configure, monitor, and debug agent orchestration.
Leveraged open protocols (MCP, A2A APIs) to enable integration with third-party agent frameworks and external tools.
Engaged Microsoft partners like NTT DATA and Sight Machine for domain-specific agent code samples and advisory.

Technologies

Azure AI Azure OpenAI Semantic Kernel

Streamlined production and QA processes, reducing manual intervention and improving reliability.
Accelerated software development life cycles for enterprise IT through collaborative agent ecosystems.
Optimized bottling line performance, root cause detection, and predictive analytics in manufacturing.
Established scalable, reusable AI agent architectures that adapt to evolving process requirements.

Architecture

The architecture consists of the Azure AI Foundry Agent Service orchestrating Connected Agents and Multi-Agent Workflows. Developers can compose and configure modular AI agents, each responsible for specific subtasks (e.g., data extraction, risk assessment, predictive analysis), and define workflow transitions and state management. The system includes a no-code portal and Python SDK, supports integration with external toolchains via Model Context Protocol (MCP) and Agent-to-Agent (A2A) APIs, and enables real-time collaboration and context sharing across agents. Example deployments integrate Azure AI, Azure OpenAI, Semantic Kernel, and domain-specific partner agents (like Sight Machine's bottling optimization).

Implementation partners2

NTT DATA Sight Machine

Sources & evidence1

Groundedness: Unavailable

Primary source

AI-generated summary. Verify important details with the linked sources before relying on this case.

Explore related AI use cases

Customer evidence