Visum Thinker MCP Server

# 🚦 Visum Congestion Analysis Scripts ## 📁 Files 1. **`analyze-congestion.py`** - Standalone script (connects to Visum COM) 2. **`analyze-congestion-inline.py`** - Inline script (for Visum console/TCP) 3. **`send-congestion-analysis.js`** - Helper to customize and prepare scripts ## 🚀 Usage Methods ### Method 1: Standalone Script (COM Connection) Requires Visum to be open with a project: ```powershell python analyze-congestion.py ``` **Output:** - Console output with top 10 congested links - CSV file: `top_congested_links_AP.csv` --- ### Method 2: Visum Python Console (Inline) From Visum's Python console window: ```python exec(open('analyze-congestion-inline.py', encoding='utf-8').read()) ``` **Note:** Always specify `encoding='utf-8'` to avoid UnicodeDecodeError **Advantages:** - No COM connection needed - Faster execution - Can be copy-pasted directly --- ### Method 3: Customized Script Generation Generate a customized version with specific parameters: ```powershell # Default: AP period, top 10 node send-congestion-analysis.js S000009result_1278407893 # Custom: AM period, top 20 node send-congestion-analysis.js S000009result_1278407893 AM 20 ``` **Output:** - Creates `analyze-congestion-temp.py` with your parameters - Shows usage instructions --- ### Method 4: Via TCP Server (Future) When sending Python code via TCP server: ```javascript // Read the inline script const script = fs.readFileSync('analyze-congestion-inline.py', 'utf-8'); // Customize parameters const customScript = script .replace(/ANALYSIS_PERIOD = "AP"/, 'ANALYSIS_PERIOD = "AM"') .replace(/TOP_N = 10/, 'TOP_N = 20'); // Send via TCP await serverManager.executeCommand(projectId, customScript, "Congestion Analysis"); ``` --- ## 📊 Output Format ### Console Output ``` ================================================================================ 🚦 TOP 10 MOST CONGESTED LINKS (AP) ================================================================================ 📊 Statistics: • Total links: 227,508 • Links with traffic: 186,234 • Congested (V/C > 0.9): 3,421 • Over-capacity (V/C > 1.0): 1,234 Rank From To V/C Volume Capacity Length -------------------------------------------------------------------------------- 1 12345 12346 2.450 3,456.0 1,410.0 850.0m [SEVERE] 2 23456 23457 2.120 2,987.0 1,410.0 720.0m [SEVERE] 3 34567 34568 1.870 2,634.0 1,410.0 950.0m [OVERCAP] ... ================================================================================ ``` ### CSV Output (standalone only) File: `top_congested_links_AP.csv` ```csv Rank;FromNode;ToNode;Name;VC_Ratio;Volume;Capacity;Length_m;TypeNo;V0Speed;AnalysisPeriod 1;12345;12346;Via Roma;2.4500;3456.00;1410.00;850.00;1;50;AP 2;23456;23457;Via Milano;2.1200;2987.00;1410.00;720.00;1;50;AP ... ``` ### JSON Result (inline version) ```json { "status": "success", "analysis_period": "AP", "total_links": 227508, "links_with_traffic": 186234, "congested_links": 3421, "overcapacity_links": 1234, "top_congested": [ { "rank": 1, "from_node": 12345, "to_node": 12346, "name": "Via Roma", "vc_ratio": 2.45, "volume": 3456.0, "capacity": 1410.0, "length": 850.0, "type_no": 1 } ] } ``` --- ## ⚙️ Configuration ### Change Analysis Period Edit the script or use send-congestion-analysis.js: ```python ANALYSIS_PERIOD = "AM" # or "PM", "IP", "IP1", etc. ``` ### Change Number of Top Links ```python TOP_N = 20 # Instead of 10 ``` --- ## 🎯 Congestion Levels | Status | V/C Ratio | Description | |--------|-----------|-------------| | **SEVERE** | > 1.5 | Severe congestion, major delays | | **OVERCAP** | > 1.0 | Over capacity, significant delays | | **CONG** | > 0.9 | Congested, moderate delays | | **HIGH** | < 0.9 | High utilization, minor delays | --- ## 🔧 Requirements ### For Standalone Script - Python with `win32com.client` (pywin32) - Visum running with project opened - Completed PrT assignment ### For Inline Script - Visum project with Python console access - Completed PrT assignment - `visum` object available in scope --- ## 💡 Tips 1. **Run after assignment** - Ensure PrT assignment is complete 2. **Check period code** - Use correct analysis period (AP, AM, PM, etc.) 3. **Large networks** - May take 10-30 seconds for 200K+ links 4. **Export CSV** - Only standalone version exports CSV file 5. **Inline for speed** - Inline version is faster (no COM overhead) --- ## 📝 Example Workflows ### Workflow 1: Quick Check in Console ```python # In Visum Python console exec(open('analyze-congestion-inline.py', encoding='utf-8').read()) ``` ### Workflow 2: Generate Report ```powershell # From PowerShell python analyze-congestion.py # Opens: top_congested_links_AP.csv ``` ### Workflow 3: Custom Analysis ```powershell # Generate custom script node send-congestion-analysis.js S000009result_123 AM 20 # Run generated script in Visum console exec(open('analyze-congestion-temp.py', encoding='utf-8').read()) ``` --- ## 🔮 Future: MCP Integration When a generic `project_execute_python` tool is added to MCP: ```javascript visum_analyze_congestion({ projectId: "S000009result_123", analysisPeriod: "AM", topN: 20 }) ``` This would execute the inline script via TCP and return formatted results to Claude. --- **Created:** 2025-11-07 **Status:** ✅ Ready to Use **Tested:** Pending