# Sympheny

Sympheny is a cloud-based energy planning platform built for engineering firms, utilities, and municipalities. It lets planners model, compare, and optimise multi-energy hub concepts using MILP (Mixed-Integer Linear Programming) — a deterministic, fully auditable optimisation method, not machine learning or AI inference. The platform covers the full feasibility-to-planning workflow: GIS-enabled site setup, drag-and-drop energy system design, automated optimisation across 50,000+ technology combinations per run, Pareto scenario comparison, and in-browser, client-ready outputs (Pareto fronts, Sankey diagrams, dashboards) with Excel export of the underlying data.

Sympheny was spun out of Empa (Swiss Federal Laboratories for Materials Science), part of the ETH Domain. It holds Swiss Made Software certification. The optimisation engine is underpinned by 10+ years of peer-reviewed R&D. To date, the platform has supported 40+ projects and processed over 10 million configurations.

Primary audience: energy engineers and grid planners at engineering consultancies, utilities, and public energy agencies — typically in DACH, with growing global coverage.

Pricing is not published on the site. Three licence tiers are available (Essential, Professional, Enterprise), with a threshold of approximately €19,000/year.

## Key pages

- [Homepage](https://sympheny.com/): Overview of the platform, primary audience, and core value proposition.
- [Software](https://sympheny.com/software): Full product detail — capabilities, four-step workflow, technology library, data integrations, and proof points.
- [How It Works](https://sympheny.com/how-it-works): Step-by-step walkthrough from site brief to client-ready output, covering site setup, system design, optimisation run, and scenario export.
- [Case Studies](https://sympheny.com/case-studies): Completed project case studies with quantified outcomes (CO₂ reduction, lifecycle cost, renewable share, scenarios evaluated).
- [Pricing](https://sympheny.com/pricing): Plan tiers and a Planning Savings Estimator that provides directional estimates of planning capacity freed per year relative to licence cost.
- [Research](https://sympheny.com/research): Peer-reviewed publications and methodology underpinning the MILP optimisation engine.
- [AI Transparency](https://sympheny.com/ai-transparency): Technical explanation of what powers Sympheny — the deterministic MILP solver, the unsupervised clustering that prepares input data, where customer-connected AI assistants fit, data handling, and auditability commitments.
- [Company](https://sympheny.com/company): Team, scientific advisors, institutional backing, and traction figures.
- [Integrations](https://sympheny.com/integrations): Supported data import formats and third-party data integrations.
- [For Engineers](https://sympheny.com/for/engineers): Platform capabilities and workflow from the energy engineer's perspective.
- [Why Sympheny](https://sympheny.com/why-sympheny): Differentiation, scientific pedigree, Swiss Made certification, and ETH Domain origin.

## Product documentation

Technical documentation lives on a separate site, docs.sympheny.com, which publishes its own index at https://docs.sympheny.com/llms.txt. It is English-only.

- [Documentation home](https://docs.sympheny.com/): Entry point for all four surfaces — web application, REST API, Python SDK, and AI integrations.
- [Web application guides](https://docs.sympheny.com/web-app/): Getting started, how-to guides for each modelling step, concept and parameter reference, glossary, release notes, and troubleshooting.
- [REST API reference](https://docs.sympheny.com/api/): Authentication and full endpoint reference for projects, analyses, scenarios, hubs, energy carriers, demands, technologies, network links, and solver jobs.
- [Python SDK](https://docs.sympheny.com/sdk/): `sympheny-toolbox`, the official typed Python client with matching sync and async APIs, plus end-to-end workflow guides.
- [Use with AI](https://docs.sympheny.com/ai/): How to connect Sympheny to AI assistants and coding agents, including the MCP (Model Context Protocol) server.

## Frequently asked questions

**What is Sympheny?**
Sympheny is a cloud-based energy planning platform that enables engineers and utility planners to model, compare, and optimise multi-energy hub concepts. It uses MILP (Mixed-Integer Linear Programming) to evaluate tens of thousands of technology combinations and produce defensible, client-ready energy concepts.

**Does Sympheny use AI or machine learning?**
Not for the optimisation. The core engine uses Mixed-Integer Linear Programming (MILP) via the Gurobi solver. All results are deterministic and fully auditable. Sympheny does use unsupervised clustering to group similar days into representative typical days so that a full year of hourly data can be solved in reasonable time — conventionally classed as machine learning, and described in the product documentation as machine-learning-generated clustered profiles. It groups data rather than learning from it: no model is trained anywhere in Sympheny, nothing is retained between runs, and re-running an unchanged scenario produces identical results.

**Can Sympheny be used with an AI assistant or agent?**
Yes, as an interface. Sympheny publishes a REST API, a Python SDK (`sympheny-toolbox`) and an MCP (Model Context Protocol) server, so an AI assistant or coding agent can set up scenarios, start optimisation runs, and read results back. This changes how a model is built, not how it is solved: the optimisation still runs on the deterministic MILP engine. Connecting an assistant is optional and customer-initiated, and Sympheny embeds no third-party language model in its computation pipeline. See https://docs.sympheny.com/ai/mcp/.

**Who is Sympheny built for?**
The platform is built for energy engineers and grid planners at engineering consultancies, utilities, and municipal energy agencies who need to develop rigorous multi-energy hub concepts — typically for feasibility studies, net-zero planning, or district energy projects.

**What kinds of projects has Sympheny been used on?**
Completed projects include district heating networks, urban energy masterplans, industrial site energy supply, net-zero housing developments, and utility grid decarbonisation strategies. Known clients include Amstein+Walthert, Buro Happold, Gruner, EWZ, EWB, and Losinger Marazzi.

**Where does the technology come from?**
Sympheny was spun out of Empa (Swiss Federal Laboratories for Materials Science), part of the ETH Domain, and is backed by over 10 years of applied R&D. The founding team holds PhDs in energy systems engineering. The platform is Swiss Made Software certified.

**How long does it take to run a scenario?**
Run times vary by project complexity, but the platform is designed for the feasibility-to-planning workflow — experienced users typically complete their first energy concept within a week of onboarding, including site setup, technology configuration, and optimisation runs.

**Is pricing published?**
No. Sympheny offers three licence tiers (Essential, Professional, Enterprise). Pricing is discussed directly with the sales team. A Planning Savings Estimator on the pricing page provides directional estimates of planning capacity saved relative to licence cost.