Frontline’s MCP Server is more than just a REST API wrapper. It serves as a complete AI-powered infrastructure for analytical modeling, bringing together model creation, discovery, execution, and interpretation into a single, unified experience. Rather than requiring users to manually build and manage models step by step, the MCP Server enables an AI assistant to guide the workflow from start to finish.
At its core, the server provides access to a searchable library of nearly 200 example models spanning optimization, simulation, data science, and decision analysis. These examples can be used as starting points, allowing users to quickly adapt proven model structures to their own business problems. Built-in templates also help users generate new models from scratch with the correct structure and components already in place.
Beyond model creation, the MCP Server supports structural analysis and result interpretation. An AI assistant connected to the server can explain how a model is constructed, identify variables and constraints, and provide meaningful insights into solve results. This creates a closed-loop workflow where users can describe a problem, generate a model, solve it, and receive actionable recommendations for improvement.
What Can You Do With It?
The MCP Server enables a wide range of capabilities that go far beyond simple model execution. It supports the full lifecycle of analytical modeling, from exploration to optimization and continuous improvement.
- Explore pre-built examples — Search a rich library of ready-made models for common business problems such as portfolio optimization, scheduling, routing, simulation, and decision analysis.
- Build models from templates — Use structured templates for optimization models, simulations, decision tables, and other model types.
- Analyze model structure — Inspect how inputs, variables, constraints, objectives, and outputs interact.
- Solve models — Run optimization models, Monte Carlo simulations, data science workflows, and decision models.
- Interpret results — Review structured results and receive insights that help guide decision-making.
- Run goal-oriented workflows — Describe an entire task, such as building, solving, and analyzing a model, and let the system carry out each step.
- Manage models — List, upload, version, promote, and clean up models in a RASON account.
Example Tasks and Prompts
| Example Task | Example Prompts |
|---|---|
| Explore nearly 200 ready-made examples |
|
| Build models from templates |
|
| Analyze and understand models |
|
| Solve optimization, simulation, data science, and decision models |
|
| Go beyond simple commands |
|
| Manage models in your RASON account |
|
