I've managed procurement for a mid-sized aggregate operation for the past six years—actually, seven now—and I've tracked every invoice, every work order, and every hour of downtime. We run Metso mining equipment: an HP200 cone crusher, slurry pumps, and a straight truck fleet for moving material. This comparison is based on my records, not on theory.
The two strategies I'll compare are reactive maintenance and predictive maintenance. Reactive means you run equipment until something fails, then fix it fast. Predictive means you monitor condition and replace parts before failure. I compare them on four dimensions: total cost, downtime, decision quality, and implementation difficulty. Keep in mind this is one buyer's experience, not a universal law.
At first glance, reactive maintenance looks cheaper. No sensors, no software, no training. You buy parts when you need them. But when we ran our HP200 cone reactively, the costs were never limited to the part. Emergency freight for Metso HP200 crusher parts added 30% to the invoice. Labor costs went up because failures tend to happen on weekends. I once paid $185 for an after-hours mechanic to swap a $47 wear insert. Including the hour he spent driving out and the paperwork, the total cost was closer to $250.
The other hidden cost was the purchasing process itself. A rush order means verification, approval, and expedited logistics. That adds hours of admin time. On a $3,500 emergency order, that admin time is real money.
Predictive maintenance had a different cost story. The initial investment in Metso IC70C automation for the cone crusher was substantial—roughly $40,000, including installation and staff training. But once the system was in place, spending leveled out. We scheduled wear part replacements, avoided most emergency freight, and reduced maintenance spend per ton by 17% in the second year according to our cost tracking system. It didn't remove all surprises. It just made them smaller.
Conclusion: predictive wins on cost structure, but only after you absorb the setup cost.
From the outside, reactive maintenance looks like it maximizes uptime because you never stop the machine for inspections. The reality is different: when a failure happens, you stop for days. Our worst case was an unplanned bearing failure in the HP200. The crusher was down for eleven days while we sourced parts. A scheduled inspection would have caught the bearing degradation in about an hour. A planned bearing replacement costs one shift. An unplanned one costs a week plus overtime and the risk of secondary damage.
Predictive maintenance creates controlled maintenance windows. Instead of scrambling, your team knows what to replace and when. This also improves parts inventory planning. We keep normal wear items for the crusher in stock and order high-dollar spares based on measurements rather than panic.
The straight truck in our fleet is a good example of the same logic. The fuel pump was starting to fail intermittently. In a reactive model, you wait until it dies on the highway. The better approach is to teach mechanics how to test fuel pump voltage and pressure on a schedule. We test every 500 engine hours. It takes roughly 20 minutes with a fuel pressure gauge from Tractor Supply. That simple check caught a voltage drop before the truck left the yard.
Conclusion: predictive wins on downtime by a wide margin.
Bad decisions happen under pressure. When a crusher is down and a contractor is waiting, you buy whatever fits. That's how we ended up with a non-genuine liner set for an HP200 that was cheaper by $900 but wore out 40% faster. In my opinion, the mistake wasn't buying non-genuine. It was buying without data. The $900 savings disappeared when we had to replace the liner twice in the same period.
Most buyers focus on part price and miss fitment and wear life. That's the blind spot. Genuine Metso HP200 crusher parts have traceability and engineering specifications behind them. Aftermarket parts can be perfectly good, but you need to verify them with measurements and test reports. Per FTC guidelines, performance claims should be substantiated—so ask for the data.
Predictive maintenance creates a different mindset. You plan the purchase, compare options, and test components before installation. The same discipline applies to fuel pumps. The question everyone asks is 'what's the cheapest pump?' The question they should ask is 'how do you know the pump was the problem?'
Conclusion: predictive wins on decision quality.
Now the part that surprised me. I assumed predictive maintenance would be harder to implement than it actually was. It was hard, just not for the reasons I expected. Our team pushed back on the IC70C automation. Older technicians trusted their ears more than the dashboard. For the first few months, the alerts felt like noise.
Training wasn't a one-time event. It took repeated sessions and a few mistakes before the crew trusted the data. But once they saw a bearing alert turn into a scheduled replacement instead of a failure, they came around.
It took me three years—and roughly 400 work orders—to understand that the best maintenance strategy is context-dependent. Reactive maintenance has advantages: no training, no software, no data infrastructure. If you run a single machine with standby capacity and tight cash flow, reactive can be the rational choice. But for Metso mining equipment like a primary crusher or a slurry pump loop, the downtime cost difference is too big to ignore.
Conclusion: reactive is easier to implement. Predictive is easy to justify once the numbers are on paper.
What would I recommend? If you run one crusher, have spare capacity, and your cash flow is tight, reactive maintenance can work—provided you stock critical spares and know how to test fuel pump issues on your own straight truck. Just don't convince yourself that emergency freight is a normal operating cost.
If you run multiple units, or if a single failure stops the whole site, build the business case for predictive maintenance. Start with the critical assets: the HP200 cone and the main haul truck. Put them on a condition monitoring schedule. Automate what you can with Metso IC70C if it's compatible. Then expand once the data proves the value.
You don't need the most advanced system on day one. Start with oil sampling, vibration readings, or simple voltage checks. The goal is to make the hidden visible before it becomes an emergency.
The industry is moving toward automation for a reason. From my perspective, efficiency is a competitive advantage. But it's not a religion. I still know how to test fuel pump output on our straight trucks myself sometimes. It's a good reminder that useful maintenance doesn't have to be high-tech. Weigh the numbers, test your assumptions, and keep the diagnostic habits—they pay off no matter which strategy you choose.
Our application engineers answer crusher and screen selection questions at no charge.
Ask an Expert