An integrated steel plant running a large blast furnace had settled into a familiar pattern: strong output between stops, but availability stuck around 81% because unplanned trips kept eating into run time. Cooling stave leaks, tuyere burn-throughs, and charging equipment failures were being fixed fast once they happened, but nothing was catching them before they happened. Over 18 months of shifting to predictive maintenance and centralized asset tracking, availability climbed to 93%. Here's what that shift looked like. Sign up to see how the same approach could apply to your furnace.
81% → 93%
Blast furnace availability improvement over the program
12 Points
Availability gained, translating directly into extra production days
Fewer Trips
Unplanned stops dropped as failures were caught before they escalated
18 Months
Timeframe from starting condition monitoring to the new availability baseline
The Challenge
The furnace team was good at reacting. A stave leak got isolated fast, a tuyere burn-through got replaced fast, a charging crane failure got repaired fast. But every one of those events was still a surprise, discovered only once it had already stopped production, and that reactive pattern is what kept availability capped in the low 80s no matter how skilled the response.
The Approach
1
Condition Monitoring On Critical Cooling Circuits
Stave and tuyere cooling water flow and temperature were tracked continuously instead of checked on rounds
2
Predictive Maintenance Scheduling
A CMMS scheduled inspections and replacements based on wear trends per asset, not a fixed calendar
3
Root Cause Analysis On Repeat Failures
Recurring failure modes were logged and traced back to root causes instead of being treated as one-offs
4
Centralized Asset History
Charging cranes, staves, tuyeres, and tapping equipment were tracked as linked assets in one system
See What Predictive Maintenance Could Do For Your Furnace
Oxmaint brings condition monitoring, predictive scheduling, and centralized asset history together in one CMMS built for heavy process plants. Sign up for a free trial to explore it with your own equipment, or book a demo to walk through a plan for your furnace.
The Results
| Metric |
Before |
After |
| Furnace Availability |
81% |
93% |
| Unplanned Stops |
Frequent, reactive response |
Sharply reduced, most caught in advance |
| Failure Detection |
Found after the fact, during a stop |
Flagged early through trend monitoring |
| Maintenance Approach |
Fixed calendar, reactive repairs |
Condition-based, scheduled by wear data |
Why It Worked
The gain didn't come from a single fix, it came from replacing guesswork with data across every major furnace asset. Once staves, tuyeres, and cranes were all logged in one place with real wear trends, the team could act on a slow drift weeks before it would have become a stop, instead of finding out only when the furnace already had.
Frequently Asked Questions
Q
What is blast furnace availability and why does it matter?
Availability is the percentage of time a furnace is actually producing versus stopped, and even a few extra points translate directly into more tonnes produced without any change to the furnace itself.
Q
How does predictive maintenance improve furnace uptime?
Predictive maintenance uses ongoing condition data, like cooling flow or temperature trends, to flag a part that's wearing out before it fails, so the fix happens on a planned schedule instead of during an unplanned stop.
Q
Can a CMMS help replicate results like this at other plants?
Yes, a CMMS that tracks furnace assets individually and schedules work off real condition data gives any plant the same foundation this improvement was built on, regardless of furnace size or age.
Turn Reactive Maintenance Into Extra Production Days
Oxmaint brings condition monitoring, predictive scheduling, and centralized asset history together in one CMMS built for heavy process plants. Sign up for a free trial to explore it with your own equipment, or book a demo to walk through a plan for your furnace.
