Open Source MCP Server · v8.0.0 · Free Beta

4 levels of optimization
intelligence for AI agents.

Solve scheduling, routing, and packing. Quantify risk with sensitivity, robust, and stochastic analysis. Find trade-offs with Pareto frontiers. Forecast demand and prescribe actions. All via MCP.

121 tests  ·  9 endpoints  ·  9 modules  ·  4 levels  ·  OR-Tools
Level 1

Deterministic Optimization

📅

Scheduling

Flexible Job Shop (FJSP)

Assign tasks to machines with precedence, time windows, and machine eligibility.

PrecedenceTime WindowsSetup Times4 Objectives
🚛

Routing

CVRPTW

Assign deliveries to vehicles with capacity, GPS coordinates, and time windows.

CapacityGPSTime WindowsDrop Visits
📦

Bin Packing

Multi-dimensional

Assign items to containers with weight, volume, and group constraints.

Weight + VolumeGroupsQuantities4 Objectives
Level 2

Optimization under Uncertainty

📊

Sensitivity Analysis

Parametric Perturbation

Find which inputs break the plan. Auto-detects critical parameters.

Fragility MapElasticityRisk RankingAuto-detect
🛡️

Robust Optimization

Worst-case Protection

Uncertainty ranges → scenarios → solutions that survive worst-case.

Worst CaseP90/P95Regret MinPrice of Robustness
🎲

Stochastic Optimization

Monte Carlo + CVaR

Probability distributions → Monte Carlo → risk-aware solutions.

4 DistributionsCVaR 90/95/99VaRRisk Recs
Level 2.5

Multi-objective Optimization

⚖️

Pareto Frontier

Trade-off Analysis

2-4 competing objectives → non-dominated solutions with correlation and spread analysis.

2-4 ObjectivesDominance FilterCorrelationWeighted
Level 3

Prescriptive Intelligence

🧠

Prescriptive Advisor

Forecast → Optimize → Advise

Historical data → demand forecast → optimal plan → risk assessment → prioritized actions.

4 Forecast Methods3 Risk AppetitesTrend DetectionAction Plans
9
Modules
11
MCP Tools
121
Tests
4
Intelligence Levels
16
Objectives
4
Distributions
4
Forecast Methods

How It Works

01

Discover

AI agent finds OptimEngine via MCP. Reads the schema. Sees 9 tools across 4 levels.

02

Optimize

Sends scheduling, routing, or packing problem. Gets optimal deterministic solution.

03

Quantify Risk

Runs sensitivity, robust, or stochastic analysis. Gets risk maps and CVaR metrics.

04

Get Advised

Sends historical data. Gets forecast, optimal plan, risk assessment, and action items.

See It In Action

From deterministic scheduling to prescriptive intelligence.

// POST /optimize_schedule
{
  "jobs": [{
    "job_id": "ORDER-001",
    "tasks": [
      { "task_id": "mixing", "duration": 30, "eligible_machines": ["M1", "M2"] },
      { "task_id": "filling", "duration": 20, "eligible_machines": ["M2"] }
    ], "due_date": 80
  }],
  "machines": [{ "machine_id": "M1" }, { "machine_id": "M2" }],
  "objective": "minimize_makespan"
}
// → Optimal schedule + metrics in ~10ms
// POST /analyze_sensitivity
{
  "solver_type": "scheduling",
  "solver_request": { /* scheduling problem */ },
  "parameters": [{
    "parameter_path": "jobs[ORDER-001].tasks[mixing].duration",
    "perturbations": [-50, -20, 20, 50, 100]
  }]
}
// → "mixing duration is CRITICAL. +50% → makespan +35%"
// POST /optimize_stochastic
{
  "solver_type": "scheduling",
  "solver_request": { /* scheduling problem */ },
  "stochastic_parameters": [{
    "parameter_path": "jobs[ORDER-001].tasks[mixing].duration",
    "distribution": "normal", "mean": 30, "std_dev": 8
  }],
  "optimize_for": "cvar_95", "num_scenarios": 50
}
// → Expected: 52. CVaR 95%: 68. "Moderate variability."
// POST /optimize_pareto
{
  "solver_type": "scheduling",
  "solver_request": { /* scheduling problem */ },
  "objectives": [
    { "name": "minimize_makespan", "weight": 2.0 },
    { "name": "minimize_total_tardiness", "weight": 1.0 }
  ],
  "num_points": 10
}
// → 6 non-dominated solutions. "Makespan and tardiness conflict (r=-0.72)."
// POST /prescriptive_advise
{
  "solver_type": "scheduling",
  "solver_request": { /* scheduling problem */ },
  "forecast_parameters": [{
    "parameter_path": "jobs[ORDER-001].tasks[mixing].duration",
    "historical_data": [
      {"period": 0, "value": 25}, {"period": 1, "value": 28},
      {"period": 2, "value": 30}, {"period": 3, "value": 35}
    ]
  }],
  "risk_appetite": "moderate"
}
// → Forecast: 33. Trend: increasing. Action: "Plan for +12% capacity."

Simple, Transparent Pricing

Free during beta. All 4 intelligence levels included.

Pro
€49/mo
Coming soon
  • 5,000 calls per day
  • All intelligence levels
  • Priority support
  • Higher solve time limits
  • Usage dashboard
Enterprise
Custom
Contact us
  • Unlimited calls
  • SLA guarantee
  • Dedicated support
  • Industry templates
  • Custom integrations

Find OptimEngine Everywhere

Connect in 10 Seconds

Add to your MCP configuration (Claude Desktop, Cursor, etc.)

{
  "mcpServers": {
    "optim-engine": {
      "command": "mcp-proxy",
      "args": ["https://optim-engine-production.up.railway.app/mcp"]
    }
  }
}