Golden Arrow Bus Services, Cape Town — Turning a Demand-Charge Constraint into an Operating Strategy

The Problem

Golden Arrow Bus Services (GABS) is one of the few transit operators anywhere running battery-electric buses at real, metered scale. Their challenge wasn't hypothetical: with a growing fleet of 60 buses on 30 shared chargers, every kilowatt of peak demand shows up directly on a monthly utility bill through a demand charge — a cost structure where the peak, not just the energy, drives the price.

GABS's operating team made a deliberate choice: cap each charger at 30 kW rather than running them at full power, deliberately trading charging speed for a lower, more predictable demand-charge bill. The question a depot planner — or a fleet operator studying GABS as a model to replicate — needs answered is: does that trade-off actually work, and what does it cost per month?

How the Depot Digital Twin Was Used

Step 1 — Model the real, capped charging strategy. The 60-bus fleet was built against GABS's actual operating setup: 30 shared chargers, deliberately capped at 30 kW each — not the charger's rated maximum, but the real, deliberately-throttled operating point the depot uses to manage demand charges.

Step 2 — Run the worst-plausible-day simulation. With every bus present and the depot's realistic overnight charging window modeled, the engine computed the actual coincident peak the depot's electrical service would see.

Step 3 — Read the peak straight off the results ledger:


Service capacity (nameplate)	1,000 kW
Modeled charging peak (worst day)	900 kW — exactly 30 chargers × 30 kW
Spare after charging	100 kW

The peak landing exactly on the charger-capped ceiling is the simulation confirming, mathematically, that the operator's real-world strategy is doing precisely what it was designed to do: the charger cap — not the service connection — is what's actually setting the depot's peak.

Step 4 — Quantify the demand-charge exposure directly. The Demand-Charge Exposure module turns that 900 kW peak into a dollar figure against an entered utility rate: $7,200/month, directional — giving the depot's finance team a clean, auditable number for exactly what the capped-charging strategy is costing on the demand-charge line of their utility bill, with the rate and the math both shown, not hidden behind a black-box total.

Step 5 — See the full 24-hour shape of the strategy. The load-profile chart shows the depot's night-shift charging ramping to a flat, capped plateau at 900 kW rather than a sharp, uncapped spike — visual proof that the charger cap is smoothing the depot's draw exactly as intended, night after night.

GABS 24-hour load profile: off-peak charging window, flat 900 kW plateau at the charger-cap ceiling

The Result

The Depot Digital Twin's modeled peak of 900 kW matched GABS's real operating point precisely — the direct, physical consequence of the depot's 30-charger, 30-kW-cap strategy, validated against the same real fleet documented in the published operational study of this depot (George-Kayode, Stratford & Booysen, 2025, World Electric Vehicle Journal), where the actual utility engaged with GABS to design exactly this kind of capped-charging approach to manage demand-charge exposure on a growing fleet.

Why It Matters

"When the modeled peak lands exactly where the real depot's charger cap says it should, that's not a coincidence — that's the twin proving the operating strategy is sound before the next fleet expansion bets more money on it."

For an operator scaling past 60 buses, this is the tool that answers "does our demand-charge strategy still hold at 100 buses, or 150?" — before the utility bill arrives and tells them the hard way.