DRI Shaft Furnace Optimization for Hydrogen DRI

By James Smith on July 23, 2026

dri-shaft-furnace-optimization-ai

Direct reduced iron is supposed to be the bridge to lower-carbon steelmaking, but running a shaft furnace well is harder than the marketing material suggests. Metallization target, carbon content, and gas reformer efficiency all pull against each other, and small swings in reducing gas composition or burden distribution can knock productivity or product quality off target for hours before anyone notices. As more plants add hydrogen blending to existing Midrex or Energiron units, the control problem only gets harder, because the furnace behaves differently than it did on natural gas alone. iFactory's DRI optimization module gives shaft furnace operators the real-time visibility they need to keep pace with that shift.

IRONMAKING · DRI / HBI · 2026

Hold metallization and productivity steady while your gas mix keeps changing

iFactory reads shaft furnace process data in real time to optimize Midrex and Energiron operation, and prepares your control strategy for hydrogen-DRI as blending ratios increase.

3–6%
Typical productivity gain from optimized burden and gas control
±1.5%
Metallization variance achievable versus wider manual ranges
8–12 Wks
To pilot on a single shaft furnace line
H2-Ready
Model architecture supports increasing hydrogen blend ratios
WHERE SHAFT FURNACE CONTROL BREAKS DOWN

Three variables, all moving, all interacting

A Midrex or Energiron shaft furnace is a counter-current reduction process where gas composition, burden movement, and temperature profile all interact continuously. Operators typically manage these with static setpoints and periodic manual adjustment, which works until conditions shift faster than the adjustment cycle.

01

Metallization drifts with gas quality

Reformer performance and natural gas composition change gas quality feeding the furnace, and metallization can drift out of spec before lab sampling catches it.

02

Carbon content is hard to hold steady

Carbon deposition depends on gas composition and residence time together, and getting both dials right simultaneously is difficult with manual control alone.

03

Burden distribution affects gas flow evenly

Uneven burden distribution creates gas channeling inside the shaft, reducing reduction efficiency in ways that aren't visible from outside the vessel.

04

Hydrogen blending changes the rules

Hydrogen reduces faster than CO but carries heat differently through the burden, meaning control strategies tuned for natural gas don't transfer directly as blend ratios rise.

05

Reformer efficiency is easy to lose track of

Reformer performance degrades gradually with catalyst age and fouling, and by the time it shows up as a productivity drop, the root cause is hard to isolate.

06

Product quality feedback lags production

DRI or HBI quality metrics are often measured downstream of the furnace, so by the time an issue is confirmed, hours of off-spec production have already occurred.

Most DRI plants are leaving 3–6% productivity on the table simply because burden and gas control can't react as fast as conditions change. Book a walkthrough to see the gap on your own furnace.

HOW IFACTORY STABILIZES THE PROCESS

From static setpoints to continuous optimization

iFactory ingests reducing gas composition, top gas analysis, burden temperature profile, and reformer performance data in real time, modeling the interaction between them instead of managing each variable in isolation.

1

Read the full gas and burden profile

Continuous ingestion of reducing gas composition, temperature zones, and burden descent rate across the shaft.

2

Model metallization and carbon in real time

Predicts expected metallization and carbon content ahead of discharge, rather than waiting for periodic lab sampling.

3

Recommend gas and burden adjustments

Suggests reformer and burden distribution adjustments to hold target metallization as feed and gas conditions shift.

4

Track reformer health continuously

Flags gradual reformer efficiency decline before it becomes a production-limiting problem.

WHY DRI CONTROL IS UNDER MORE PRESSURE THAN EVER

The green steel transition runs directly through the shaft furnace

DRI's role in steelmaking is shifting from a scrap-substitute niche to a central pathway for lower-carbon production, and that shift is putting real pressure on plants to run shaft furnaces at higher utilization and tighter quality specification than the original design basis assumed. EAF operators buying DRI or HBI as scrap substitute increasingly specify metallization and gangue content tolerances that are tighter than what many furnaces were tuned to hold reliably under manual control alone.

At the same time, the hydrogen blending timeline is moving faster in some regions than DRI plant control systems are ready for. Plants that built their control practice entirely around natural-gas reduction chemistry are finding that even modest hydrogen blend ratios change furnace behavior enough to require real retuning, not just a minor setpoint adjustment. Getting ahead of that transition with a control system that already accounts for hydrogen reduction kinetics avoids a scramble later when blend ratios increase faster than expected.

There's also a capital allocation angle worth considering. Many plants evaluating a green hydrogen transition are weighing the cost of new furnace capacity against getting more consistent output from existing shaft furnaces first. Since the productivity and consistency gains available through better real-time control often cost a fraction of new capacity investment, tightening existing furnace performance is frequently the more immediate lever available to plant managers under budget pressure.

CAPABILITIES

Built for the shaft furnace control room

LIVE

Real-time metallization prediction

Continuously modeled metallization estimate with confidence band, replacing reliance on periodic lab sampling for control decisions.

LIVE

Carbon content optimization

Recommends gas composition and residence time adjustments to hold carbon content within target range for your product grade.

LIVE

Burden distribution diagnostics

Detects gas channeling patterns from temperature profile asymmetry, flagging burden distribution issues before they affect yield.

LIVE

Reformer performance tracking

Monitors reformer efficiency trend over time, flagging catalyst degradation or fouling before it limits gas production capacity.

LIVE

Hydrogen blend transition modeling

Supports control strategy adjustment as hydrogen blend ratio increases, accounting for the different heat and reduction kinetics of H2 versus CO.

LIVE

Productivity and yield reporting

Tracks productivity, metallization, and carbon consistency trends across shifts and campaigns for continuous improvement review.

MEASURABLE IMPACT

What DRI plants see within one quarter

Productivity gain
3–6%
From tighter burden and gas control across the campaign
Metallization variance
-38%
Narrower spread against target metallization
Off-spec production incidents
-45%
Fewer discharge batches outside carbon or metallization target
Reformer-related downtime
-27%
Earlier detection of reformer efficiency decline
DEPLOYMENT

What a shaft furnace pilot includes

Connects existing instrumentation

Uses gas analyzers, thermocouples, and reformer sensors already installed on Midrex or Energiron lines.

On-premise deployment

Runs on plant-network hardware with no cloud dependency for process data.

8–12 week pilot

Includes historical data calibration and shadow-mode validation before live operator use.

H2-blend ready architecture

Model design accounts for hydrogen reduction kinetics, supporting current and future blend ratios.

Works with DRI and HBI lines

Deployed across cold DRI, hot DRI, and HBI briquetting configurations.

24x7 managed monitoring

iFactory's operations team maintains model performance so your process engineers stay focused on the furnace.

QUESTIONS DRI PLANT MANAGERS ASK

Shaft furnace AI, explained plainly

Does this work on both Midrex and Energiron furnace designs?
Yes. The underlying model adapts to the counter-current reduction physics common to both designs, while calibration accounts for the specific gas flow and burden characteristics of your furnace type. Deployments are live on both Midrex shaft furnaces and Energiron zero-reformer configurations, with the model tuned to each plant's actual instrumentation and gas source during the pilot.
How does iFactory handle the transition to higher hydrogen blend ratios?
Hydrogen reduces iron oxide faster than carbon monoxide but carries less heat per unit volume through the burden, which changes the temperature profile operators need to target. iFactory's model is built to represent this difference directly rather than assuming natural-gas kinetics, so recommendations adjust automatically as your blend ratio increases. This is a core reason plants planning a hydrogen transition start with iFactory now rather than after the blend ratio changes.
What data do you need from our reformer to get started?
The pilot typically uses existing reformer outlet gas composition, temperature, and pressure data already collected for process control. If your reformer instrumentation is limited, the team will identify the minimum additional data needed during the scoping call, which is usually a small addition rather than a major sensor upgrade. Reach out through iFactory support to review your current instrumentation.
Can this help us qualify a new DRI grade faster?
Yes. Because the model tracks metallization and carbon content continuously rather than at periodic lab intervals, plants use it to shorten the trial-and-error cycle when qualifying new grades or adjusting for different ore sources. Tighter real-time visibility into how burden and gas changes affect product quality typically compresses qualification runs meaningfully.
How long until we see measurable productivity gains?
Most plants see the model producing reliable metallization and carbon predictions within the first month of shadow-mode operation. Once operators begin using live recommendations, typically by week 8–10, productivity and consistency gains usually become visible within the following month of production data. You can review a detailed pilot timeline when you book a demo.

Get ahead of the hydrogen-DRI control problem

iFactory optimizes your shaft furnace today and prepares your control strategy for tomorrow's blend ratios. Book a demo and we'll walk through it on your process data.


Share This Story, Choose Your Platform!