Energy Purchasing

Walloon food manufacturer · 1,746 MWh/year · Belgium

From one annual click to a smart forward strategy

The challenge

The client was locking in 100% of its annual volume in a single forward contract each year, always at the Endex Cal benchmark — no timing, no flexibility. Energy costs had become a margin risk with no levers to pull.

Epon mapped the full 2025 forward curve and identified 4 optimal entry windows
Core/Flex split: 70% locked early, 30% reserved for opportunistic timing
Quarterly volumes sized to actual consumption profile (Q1 heavy, Q3 lightest)

Results — 2025 full year

−€113K
Annual savings vs single-click benchmark
−43.6%
€84/MWh
Average purchase price achieved (vs €149 benchmark)
Core/Flex
4 clicks
Quarterly buy decisions — vs 1 annual click previously
+Momentum signal
Spot Exposure

Walloon industrial site · 1,746 MWh/year · Solar PV + BESS

Turning rooftop solar into a grid independence strategy

The challenge

The site had 2.24 MWp of solar PV but was still drawing heavily from the grid — especially in summer — and injecting surplus at near-zero value. Spot exposure in Q3 reached 32% of offtake, creating price volatility. Epon modelled what a battery storage upgrade could do.

Solar expanded to 4.274 MWp + BESS 2.5 MW / 5 MWh installed
Epon algorithm dispatches BESS to maximise self-consumption and minimise peak draw
Spot exposure managed quarter by quarter — highest risk period (Q3) nearly eliminated

Results — Before vs After (projected 2025)

−56%
Reduction in annual grid offtake (1,725 → 757 MWh)
PV + BESS
32% → 5%
Q3 spot exposure — peak summer risk eliminated
Q3 · Jul–Sep
827 MWh
Annual BESS dispatch — arbitraging solar production daily
~2.3 MWh/day avg
Demand Response

Flemish logistics operator · 3,100 MWh/year · Refrigeration + conveyor loads

Shifting load to the cheapest hours — automatically

The challenge

The site ran large refrigeration compressors and conveyor systems around the clock with no price signal integration. Peak-hour consumption (7h–9h and 17h–20h) was expensive and avoidable. Epon built an hourly attractiveness score to guide when to consume more or less.

Epon scores every hour 1–5 (Avoid → Ideal) based on spot price, solar, and network demand
15% of total load identified as shiftable without impacting operations
Peak-hour consumption reduced — also lowering the ORES capacity charge component

Results — Annual estimated impact

465 MWh
Annual load shifted from peak to off-peak windows
15% flex
~€13K
Annual gain from peak/off-peak spread capture
~€28/MWh spread
Q1 best
Winter spreads highest — most valuable quarter for DR
Jan–Mar peak
Peak Shaving

Flemish plastics manufacturer · 4,500 MWh/year · High demand variability

Taming monthly demand peaks to control the grid bill

The challenge

The ORES capacity term — based on the average of monthly HP demand peaks — was growing year-on-year as production expanded. Occasional spikes during start-up sequences were setting new monthly records and inflating the capacity charge permanently. The site had no visibility on when peaks were occurring.

Epon monitors real-time demand and flags HP peak risk 15 minutes in advance
BESS dispatched to shave spikes above 537.7 kW contracted PS threshold
Monthly HP peak stabilised — capacity term no longer compounding annually

Results — 12-month operational period

537 kW
HP peak maintained — vs up to 1,400 kW grid connection limit
vs 1,400 kW limit
~€18K
Annual reduction in ORES capacity term
Peak hours · Mon–Fri 07:00–22:00
12 / 12
Months with PS threshold respected — zero overruns
Full year