Energy-aware production planning in a foundry
When does each furnace melt, when do heat treatment and the e-trucks charge, and how does the battery run — so the peak and the energy bill go down?
Drag to rotate, scroll or pinch to zoom. The schedule below plays the plan; hover a bar for details. Plan computed with a Timefold model.
The planning question
An iron foundry over one week in 15-minute slots: two 2.4 MW induction furnaces feed a molding line that pours every 90 minutes, a heat-treatment furnace runs unmanned, six e-trucks charge overnight, 1.5 MWp PV on the roof and a 2 MWh battery. Energy is bought at day-ahead prices, and the highest 15-minute peak sets the demand charge for the whole year.
How we plan it
Timefold plans production, flexible loads and the battery in one model: shifting a melt by 15 minutes or a heat-treatment batch into the night often saves more than the battery alone.
What the plan shows
| KPI | Optimized plan | Comparison: today, without battery |
|---|---|---|
| grid peak | 4,900 kW −1,753 | 6,653 kW |
| energy cost per year | 2,816 k€ −374 | 3,190 k€ |
| energy | 2,081 k€ −111 | 2,192 k€ |
| demand charge | 735 k€ −263 | 998 k€ |
Decisions
- furnace and start of every melt before its pour
- start of every heat-treatment batch inside its window
- charger and start of every e-truck
- battery settings: peak target and charge price limit
Constraints
- pour times of the molding line are fixed; large castings need both furnaces
- molten metal held hot at most 2 hours
- melting only while the melt shop is staffed
- grid import below 8 MW
- trucks back in the evening, out at 05:00
Objective
low energy cost and a low 15-minute peak, without touching the casting rhythm
Questions
Does production get slower?
No. Pour times and shifts stay fixed; the model only uses the freedom that already exists — melt a bit earlier, heat treatment at night, trucks after the evening peak.
Where do prices and loads come from?
Day-ahead prices from the exchange or the supplier, load profiles from the metering point and the machines; PV from a forecast.
More use cases
Your planning problem looks different?
Every operation has its own rules. Tell us about yours — we assess feasibility and expected impact in a free initial call.
Book a free call Write to us