key takeaways
If you only read 30 seconds of this article:
- ERP planning modules were built for demand-driven finished goods; spare parts fail at random and have no demand curve, so ERP formulas mis-handle them.
- The zero-demand trap is the core failure: an intermittent critical part returns near-zero recommended stock, so the ERP tells you to run out of the part that stops the line.
- Multi-ERP fragmentation compounds it: the same part under different numbers across systems means no ERP sees true network on-hand.
- The fix is a layer over the ERPs, not a replacement: a Fortune 500 CPG manufacturer on SAP identified $63M and verified $60M across 41 sites, based on Verusen customer results.
MRO ERP limitations
Short answer: MRO inventory optimization fails in ERP-driven environments because ERPs were designed to plan finished goods from demand history, and spare parts have no usable demand history, they fail at random. Standard ERP safety-stock formulas return near-zero for intermittent critical parts, and multi-ERP fragmentation hides true on-hand across systems. The answer is not to replace the ERP but to add an AI optimization layer over it that stocks by criticality and reads every system as-is.
ERP-driven MRO: Managing spare parts through the ERP's own inventory and planning modules, which optimize on demand history and treat each system's records in isolation.
Why ERPs can't optimize spare parts
ERP planning modules, and demand tools like SAP IBP, were built for finished goods that sell on a forecastable schedule. They optimize by projecting demand and buffering its variability. Spare parts do not sell; they fail, at random, so the demand curve the ERP needs simply does not exist. Feeding a maintenance part into a demand-planning engine asks a sales question about an item that answers to failure. That mismatch, not a bad setup, is why the numbers come out wrong. Purpose-built AI-powered MRO inventory optimization exists precisely because the ERP category was never designed for this.
The failure is structural, so no amount of ERP configuration fixes it; you need a different decision rule.
The zero-demand trap
The clearest symptom is the zero-demand trap. A bearing that fails twice in five years has almost no demand history, so an ERP safety-stock formula returns near-zero and the system recommends stocking none, on the exact part whose failure stops the line. Then it fails, and the line is down for three weeks waiting on a part the ERP told you not to hold.
| ERP assumption | MRO reality | Result |
|---|---|---|
| Demand is frequent and forecastable | Failures are rare and random | Formula returns ~0 for critical parts |
| History predicts the future | Intermittent parts have no history | Under-stocks line-stop spares |
| Each system is complete | Parts span many ERPs | No true network on-hand |
The replacement rule is criticality, covered in the companion MRO criticality analysis guide; the demand-side formula that does still apply to fast movers is in this safety stock formula reference.
MRO ERP limitations zero demand trap
What to do instead: a layer over the ERP
The fix is not to rip out the ERP; it is to leave it as the system of record and add an optimization layer above it. That layer ingests materials, consumption, and lead times from every ERP, EAM, and P2P system as-is, resolves duplicates across them, and stocks by criticality and lead time rather than demand. Because it reads the systems you already run, it delivers in weeks without a migration.
The proof is verified savings inside ERP-driven environments. A Fortune 500 CPG manufacturer on SAP across 41 sites identified $63M and verified $60M while cutting material review time from over 20 minutes to 4 minutes; Domtar, running six ERP instances, identified $42M and verified $11M; a major US energy company on Maximo reviewed 45,000 materials and verified $29.7M with 100% audit capability, based on Verusen customer results.
| CPG on SAP: identified / verified, 41 sites | $63M / $60M |
| Domtar: identified / verified across 6 ERPs | $42M / $11M |
| working solution from data connection | < 45 days |
MRO ERP limitations what to do instead
How to start without leaving your ERP
You keep the ERP and add the decision layer it was never built to provide.
- Connect every ERP, EAM, and P2P system as-is; no migration, no cleanse first.
- Resolve duplicate materials so network on-hand and demand are true.
- Stock by criticality and lead time, starting with line-stop parts the ERP under-stocks.
- Let the ERP remain the system of record; the layer feeds it the right levels.
- Review quarterly as failure rates and lead times drift.
Most customers reach a working solution in under 45 days and unlock $20M in working capital on average, no cleanse first, based on Verusen customer results. Then talk to an MRO expert to see it run against your own ERP data.
Further reading: MRO spares inventory optimization guide, spare parts inventory management guide, and safety stock formula methods.
Frequently asked questions
Because ERP planning modules were built for finished goods that sell on a forecastable schedule, and spare parts fail at random with no demand curve. Standard ERP safety-stock formulas return near-zero for intermittent critical parts, so the system under-stocks the parts that stop production. The mismatch is structural, not a configuration error.
A part that fails twice in five years has almost no demand history, so an ERP formula returns near-zero and recommends stocking none, on the exact part whose failure stops the line. The replacement is criticality-based stocking, which sizes by consequence of failure and lead time.
No. The fix is an AI optimization layer over the ERP that reads every system as-is, resolves duplicates, and stocks by criticality, while the ERP stays the system of record. It delivers in weeks without a migration.
Because the same part carries different numbers across systems, so no single ERP sees true network on-hand. You overstock in one place and run short in another, and no ERP report can total the real position. An optimization layer that spans systems resolves it.
Yes, and the proof is verified savings there: a CPG manufacturer on SAP verified $60M across 41 sites, and a US energy company on Maximo verified $29.7M across 45,000 materials, based on Verusen customer results, all without replacing the ERP.
PN
- Paul Noble
- Founder & CEO, Verusen
Paul founded Verusen to bring AI-native systems of record to industrial materials. He has spent 15+ years working alongside F&B, oil & gas, and manufacturing operators on the MRO data problem.