How to Reduce Maximum Demand Charges in a Factory

Measure first — the prerequisite every other tactic depends on

Before any demand-reduction measure can be sized, targeted, or verified, you need to know exactly when your peak demand occurs, how high it goes, and which loads are responsible. This sounds obvious, but a surprising number of factories manage demand reactively — they discover the peak on the monthly bill, weeks after it happened, with no information about which equipment caused it or what the trigger was.

Real-time demand monitoring closes this gap. The Titan energy meter (Class 0.5S per IEC 62053-22) computes block and sliding-window demand continuously using the same interval method your DISCOM meter uses. It tracks how much headroom remains before your contract demand limit and sends configurable alerts when demand is on track to exceed that limit before the current interval closes.

That alert — even two or three minutes before the interval ends — gives your operator or energy manager time to defer a compressor start, reduce chiller capacity, or pause a non-critical process. Titan provides the monitoring and the advance warning; the load curtailment itself is an operator or EMS/PLC action. With sub-metering at the feeder level, you can also see which loads are active during each peak interval — the foundation for every targeted intervention that follows.

Soft-starters and VFDs: what they do and don’t do

A soft-starter or variable frequency drive (VFD) limits inrush current at startup, so the motor ramps up instead of drawing several times rated current for a few seconds. That reduces voltage dips and mechanical stress on couplings and driven equipment — but because inrush contributes so little to a 15-minute average, a soft-starter on its own does not meaningfully cut a maximum demand charge.

A VFD is different where the load itself varies. On cooling fans, pumps on variable-flow systems and similar loads, running the motor slower when full output is not needed reduces the running power drawn — and lower running power during peak periods is what lowers maximum demand. Whether that is worth doing depends on your load profile, so size it from measured data.

Prioritise variable-load motors that run during peak demand periods, are large relative to your contract demand, or show up in your metering data as contributors to peak intervals.

Use predictive demand alerts — the real-time safety net

Even with staggered startups and load-shift schedules in place, demand spikes can still occur from unplanned events — an unscheduled batch run, an equipment failure that forces a restart sequence, or HVAC and production loads coinciding during an unusually hot afternoon. Predictive demand alerting provides a real-time safety net.

Titan tracks accumulated demand within the current billing interval and projects the end-of-interval value. When the projection crosses a configurable threshold — for example, 88% of contract demand — it sends an alert to the energy manager’s phone or a display in the control room. The alert arrives while there are still several minutes left in the interval: enough time for an operator to hold off a non-urgent load, reduce chiller output, or call the production manager to pause a batch.

This real-time layer is the difference between managing demand proactively and discovering the penalty on the bill. It is also the enabler that makes it practical to set contract demand closer to actual operating demand — the tighter the safety margin, the more important the alert.

Right-size contract demand — both ways

Contract demand is not a set-and-forget figure. It should be reviewed periodically against actual operating patterns and revised when there is a sustained mismatch — in either direction.

If maximum demand consistently stays well below contract demand, the facility may be paying minimum demand charges on capacity it never uses. Reducing contract demand through a formal DISCOM application lowers the minimum billing floor. The risk is that if operating demand later increases — a new production line, a capacity expansion — you may face exceedances again. Use at least six to twelve months of demand data before applying for a downward revision, and factor in seasonal variation and planned expansions.

If maximum demand regularly approaches or crosses contract demand, and operational measures cannot reliably keep it under the limit, a formal upward revision may be more cost-effective than paying repeated exceedance penalties. Compare the extra demand charge on the higher contract demand against the excess-demand penalties you have actually been paying to decide.

DISCOM procedures for demand revision, any restrictions on revision frequency, and applicable fees vary by state. Keep power factor in view too: where demand is billed in kVA, poor PF raises kVA demand for the same kW (see power factor penalties). The demand monitoring data from Titan gives you the historical evidence base to make either case objectively.