Production cost simulation, security-constrained, on the live network.

CEForesight is the annual hourly production-cost twin of CEGridSight. Same 27-zone European fleet, same fuel and carbon assumptions — extended with full security-constrained unit commitment, reserve co-optimisation, and contingency-aware DC-OPF. Run any year, any policy, any fleet — and read out the realised cost, emissions, and LMP distribution as you would from a regulated SCUC tool.

Get on the early-access list CEAtlas Expert pricing

Why this exists

PCM should not be a separate spreadsheet from your forecast.

Production-cost models live on year-old datasets, run by consultancies, with answers that arrive months after the question. CEForesight is the same code path as the live CEGridSight forecast, extended to the SCUC and reserve-co-optimisation features that make it usable for resource adequacy, integration studies, and capacity-market clearing — and runnable on demand against today's fleet, not last year's snapshot.

What it does

Three pillars.

Security-constrained UC

Annual 8,760-hour unit commitment with min-up/min-down, ramp limits, must-run, and outage scheduling per unit. N-1 contingency screening as binding constraints, not post-processing.

Reserve co-optimisation

Energy and operating reserves (regulation up/down, spinning, non-spinning, frequency response) cleared as a single MILP with substitutability rules and locational reserve zones — exactly as a real ISO market clears them.

LMP decomposition

Every zonal LMP split into energy + congestion + losses, with sensitivity to each binding line constraint. Hand a trader, a regulator, or a planner the exact line that's pricing the position.

Methodology

For the quants in the room.

Plain-English first; this section for anyone vetting the maths. Skip if you're not running the study yourself.

Formulation

Rolling-horizon SCUC + DC-OPF.

Day-ahead SCUC with 24-hour look-ahead and 24-hour solved horizon, rolling daily. Hourly resolution, configurable to half-hourly. Network: DC-OPF with PTDFs computed from the same 27-zone topology used by CEAtlas. Generation tight constraints follow Damci-Kurt and Carrion–Arroyo formulations.

Contingencies

Iterative N-1 with screening.

N-1 enumeration filtered to a binding subset by an LODF-based screening pass, added as constraints in successive solves until no new contingency binds. AC backcheck via Newton-Raphson on the cleared dispatch. Cuts derived from binding contingencies cached and reused across hours.

Outages

Monte-Carlo on forced outages.

Forced-outage rates per technology from NERC GADS and ENTSO-E TYNDP. 200 Monte-Carlo replications per study by default. Common-mode weather-driven outage correlation modelled via copula on temperature and wind speed.

Reserves

Probabilistic with explicit substitution.

Reserve requirements set probabilistically against forecast error and N-1 contingency size. Faster reserves substitute for slower per market rules; cross-zonal reserve sharing enforced or forbidden per scenario. Solved jointly with energy as a single MILP — no two-stage sequential clear.

Status

In active build.

CEForesight is in active build, slated for integration into the CEAtlas Expert tier. Customers on early access are paired with the engineering team during build for spec and validation review against benchmark studies.

Request early access See CESentinel Back to all products