mes-mcp-server

DohyeonKong/mes-mcp-server

3.2

If you are the rightful owner of mes-mcp-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to dayong@mcphub.com.

An educational Model Context Protocol (MCP) server for Manufacturing Execution System (MES) data analysis focused on footwear manufacturing.

Tools
4
Resources
0
Prompts
0

MES MCP Server

An educational Model Context Protocol (MCP) server for Manufacturing Execution System (MES) data analysis focused on footwear manufacturing.

Overview

This project demonstrates how to build a custom MCP server that analyzes real-world footwear manufacturing data. It's designed for educational purposes to help students learn:

  • What MCP is and how it works
  • How to define resources (documents, PDFs, and data)
  • How to implement tools for data analysis
  • Real-world manufacturing KPIs (Build to Schedule metrics)
  • Manufacturing process understanding (Injection Phylon operations)

Features

Resources

The server exposes two types of resources:

Documents (Markdown and PDF files):

  • mes://documents/kpi-definitions - BTSv and BTSm KPI formulas and descriptions
  • mes://documents/data-schema - Documentation of footwear production data structure
  • mes://documents/report-templates - Available report configurations
  • mes://documents/injection-phylon-operation - IP manufacturing process overview
  • mes://documents/manufacturing-process - Complete footwear manufacturing process diagram (PDF)

Data (CSV files):

  • mes://data/production-plan - Scheduled production by model, color, and shoe size (3,395 records)
  • mes://data/production-results - Actual production with MCA02 machine degradation scenario

Tools

The server provides four main tools:

  1. calculate_kpi - Calculate BTSv or BTSm for specific zones, machines, or shifts
  2. calculate_all_kpis - Calculate all KPIs for a dashboard view
  3. analyze_kpi_trend - Analyze KPI trends over time (identify machine degradation)
  4. generate_kpi_report - Generate formatted reports with analysis

Supported KPIs

  • BTSv (Build to Schedule - Volume) - (Total Units Produced / Total Units Planned) × 100%

    • Measures whether total production volume matches the plan

  • BTSm (Build to Schedule - Mix) - (Matched Production / Total Units Planned) × 100%

    • Measures whether the right products (model, color, size) were produced
    • Matched Production = sum of min(actual, planned) for each item

Project Structure

mes-mcp-server/
├── src/
│   └── index.js                        # Main MCP server implementation
├── docs/                               # Document resources
│   ├── kpi-definitions.md              # BTSv and BTSm KPI definitions
│   ├── data-schema.md                  # Footwear production data schema
│   ├── report-templates.md             # Report templates
│   ├── injection-phylon-operation.md   # IP process documentation
│   └── manufacturing-process.pdf       # Manufacturing process diagram
├── data/                               # Data resources (CSV)
│   ├── production-plan.csv             # 3,395 production plan records
│   └── production-results.csv          # Results with MCA02 degradation
├── tests/                              # Test files (to be implemented)
├── package.json
└── README.md

Installation

  1. Prerequisites:

    • Node.js 18 or higher
    • npm or yarn

  2. Install dependencies:

    cd mes-mcp-server
npm install

  3. Test the server:

    npm start

Usage with Claude Desktop

To use this MCP server with Claude Desktop, add it to your Claude configuration:

On macOS: Edit ~/Library/Application Support/Claude/claude_desktop_config.json

On Windows: Edit %APPDATA%\Claude\claude_desktop_config.json

Add the following configuration:

{
  "mcpServers": {
    "mes": {
      "command": "node",
      "args": ["/absolute/path/to/mes-mcp-server/src/index.js"]
    }
  }
}

Replace /absolute/path/to with your actual project path.

After updating the configuration, restart Claude Desktop.

Example Usage

Once connected to Claude Desktop, you can:

View KPI definitions:

Can you show me the BTSv and BTSm KPI definitions?

Calculate KPIs:

Calculate BTSv and BTSm for machine MCA02 on 2025-10-08

Analyze machine degradation:

Show me the BTSv trend for machine MCA02 over all dates

Generate reports:

Generate a machine deep dive report for MCA02

Access data directly:

Show me the production plan for ZONE 1

Learn about manufacturing process:

Explain the Injection Phylon operation process

For Students: Learning Activities

Activity 1: Explore Resources

  • Use Claude to read all document resources
  • Understand the KPI formulas and how they're calculated
  • Review the data schema

Activity 2: Calculate KPIs

  • Calculate BTSv and BTSm for different zones and machines
  • Compare KPIs across different shifts (Shift2, Shift3-1)
  • Analyze which machines and zones perform best

Activity 3: Trend Analysis

  • Analyze BTSv and BTSm trends over the past week
  • Identify machine degradation patterns (e.g., MCA02)
  • Find correlations between volume and mix performance

Activity 4: Generate Reports

  • Generate different report types
  • Compare daily vs. weekly reports
  • Understand how reports aggregate data

Activity 5: Modify and Extend

  • Add a new KPI definition to docs/kpi-definitions.md
  • Add sample data to the CSV files
  • Implement a new tool in src/index.js
  • Create a custom report template

Data Format

Production Plan CSV

date,shift,zone,machine_id,model_name,mcs_code,color,shoe_size,quantity
2025-10-01,Shift2,ZONE 1,MCA01,AIR MAX BOLT (PS),IU 007,WHITE,8,58

Production Results CSV

date,shift,zone,machine_id,model_name,mcs_code,color,shoe_size,planned_quantity,actual_quantity
2025-10-01,Shift2,ZONE 1,MCA01,AIR MAX BOLT (PS),IU 007,WHITE,8,58,57

Data Scenario:

  • Dates: 2025-10-01 to 2025-10-08
  • Machine MCA02 degrades from 98% to 40% performance
  • Other machines maintain ~98% performance
  • Demonstrates BTSv and BTSm anomaly detection

Extension Ideas

Students can extend this project by:

  1. Adding Google Sheets Integration:

    • Replace CSV files with Google Sheets API
    • Learn about OAuth2 authentication
    • Handle real-time data updates

  2. Adding New KPIs:

    • MTBF (Mean Time Between Failures)
    • MTTR (Mean Time To Repair)
    • TEEP (Total Effective Equipment Performance)

  3. Enhanced Reporting:

    • Generate actual chart images using charting libraries
    • Add AI-generated insights and recommendations
    • Export reports to PDF

  4. Data Visualization:

    • Create a web dashboard
    • Add real-time monitoring
    • Implement alerts for threshold violations

  5. Advanced Analytics:

    • Predict equipment failures using historical data
    • Optimize production schedules
    • Identify root causes of quality issues

Troubleshooting

Server won't start:

  • Check that Node.js 18+ is installed: node --version
  • Ensure dependencies are installed: npm install
  • Check for syntax errors in src/index.js

Claude can't connect:

  • Verify the path in claude_desktop_config.json is correct
  • Use absolute paths, not relative paths
  • Restart Claude Desktop after config changes
  • Check Claude logs for error messages

No data returned:

  • Verify CSV files exist in the data/ directory
  • Check CSV format matches the schema
  • Ensure date formats are correct (YYYY-MM-DD)

Learning Resources

License

MIT License - Free for educational use

Contributing

This is an educational project. Students are encouraged to:

  • Fork the repository
  • Add new features
  • Share improvements
  • Create teaching materials

Happy Learning! 🏭📊

For questions or issues, please refer to the MCP documentation or consult your instructor.