Model Context Protocol

Analyze queues by asking

Connect LINE to an AI assistant, describe a system in plain language, and get queue lengths, utilization, response times, and throughput.

Try asking “Analyze an M/M/1 queue with arrival rate 2 and service rate 5.”
1

Install LINE

This installs the solver and the line-mcp server.

pip install line-solver
2

Connect your client

Choose the client you use.

Claude Code

Run once in your terminal:

claude mcp add line-solver -- line-mcp

Check it with claude mcp list.

Claude Desktop

Add this to claude_desktop_config.json, then restart Claude:

{
  "mcpServers": {
    "line-solver": {
      "command": "line-mcp"
    }
  }
}
Command not found? Run which line-mcp and use the returned absolute path in your client configuration.
3

Ask a question

The assistant chooses the LINE tool and solver, then explains the result.

“Analyze an M/M/1 queue with arrival rate 2 and service rate 5.” Single queue
“Eight jobs circulate through stations with demands 0.5 and 0.3. Find the bottleneck.” Queueing network
“Sweep arrival rate from 0.5 to 2.5 and show how response time changes.” Parameter sweep

What you can do

Build

Create open, closed, mixed, multiclass, fork/join, and layered models.

Solve

Get averages, distributions, percentiles, transient results, or simulation traces.

Explore

Sweep parameters, compare solvers, inspect examples, and visualize topology.

Import & export

Work with LINE JSON, JMT, PNML, LQNX, MATLAB, pickle, and Python.

Tool reference

You usually do not need to call tools yourself. These groups are useful when you want to guide the assistant.

Quick analysis
analyze_queue, analyze_network, sweep_parameter
Model building
build_network, build_topology, build_from_gallery
Files & solvers
solve_model_file, find_solver, compare_solvers, convert_model
Discovery
list_gallery, list_examples, list_solvers, check_install