Run five plants like one estate, not five spreadsheets.
Multi-site industrial groups run three to twelve plants under one parent, on a fragmented ERP estate. SAP on plant A. Microsoft Dynamics F&O on plants B and C. NetSuite on plant D. Sometimes Sage X3 or Plex on a fifth. The leverage sits in consolidated supplier negotiation, inter-plant transfers, and freight consolidation across overlapping lanes. The risk sits in inconsistent approved-vendor lists. Polymr operates above the ERPs without imposing a single chart of accounts.
Built for the industrial shape.
Polymr parses multi-plant integration feeds (SAP IDocs, Oracle EBS exports, NetSuite SuiteTalk payloads, Infor BAAN extracts) into a single normalized operations context. Each plant keeps its own chart of accounts, item master, and routing master on the source side. The cross-plant view sits above the ledgers so the corporate operations VP does not wait on a quarter-end consolidation pack to see PLT-C drifting from the band.
Cross-plant vendor consolidation rolls up against category-level spend with the live RFQ window per vendor. Per-plant planner queues stay intact so a PLT-A planner does not lose authority over the local sequence. The central operations roll-up reads the same rows the planner just signed off on, so the conversation about freight, capacity reallocation, or category contracts opens on shared evidence.
Six failure modes a multi-plant operations team lives with weekly.
Each surfaces in the first half-hour of conversation with a central procurement lead or a corporate operations VP. None resolve by tightening any single plant's purchasing process. They live in the seams between plants where today no workflow runs.
- Vendor leverage leaks plant-by-plantPlant A pays $8.30 for PMR-4124; plant C pays $9.15 for the same SKU from a different vendor
Three plant purchasing groups maintain three approved-vendor lists for the same SKU. Plant A buys PMR-4124 from V-218 at $8.30/ea on a quarterly contract. Plant C buys the identical SKU from V-201 at $9.15/ea because nobody has shared the V-218 ceiling across sites. The 10 percent gap is a quarter-by-quarter steady-state leak.
- Cross-plant inventory invisiblePlant B has 1,280 units of PMR-4304 idle; plant D opens PO-87412 for 800 more
Plant B holds a buffer of PMR-4304 that drifted out of demand. Plant D needs 800 units, opens PO-87412 to V-244, pays freight, takes lead time. Inter-plant transfer from B to D was never proposed because the cross-plant on-hand view is a weekly export reconciled in a CFO spreadsheet rather than a live signal.
- Consolidated PO blocked by approval geometryFour-plant consolidated PO worth $42K shaves 4.2 percent, takes three weeks to clear
The consolidated draft routes to four plant controllers in serial because each plant signs its own commit. Each controller has their own queue and SLA. By the time signature four lands, the spot price has moved, the vendor needs requote, and the saving is partially reabsorbed. The 4.2 percent on the model becomes 2.1 percent in practice.
- Inter-plant transfer dragTransfer from plant B to plant D takes 11 days because the TMS is decoupled from the ERPs
Plant B ships an inter-plant transfer of PMR-4304 to plant D. The originating ERP (SAP) writes the transfer order; the destination ERP (Dynamics F&O) expects the inbound. The TMS in between is a third system the carrier-side updates by email. Eleven calendar days, not the four the lane should run.
- Supplier scorecard fragmentsV-244 sits at preferred in plant A and watch in plant C; both expedite the same week
Each plant maintains its own supplier scorecard against its own deliveries. V-244 has been hitting OTIF in plant A and missing in plant C for six weeks. Plant A and plant C both expedite the same week against the same vendor without seeing each other. The consolidated scorecard would have routed plant C's order to V-218.
- Freight consolidation unclaimed$840K per year in freight consolidation across overlapping lanes nobody can see
Plant A inbound from V-218 in Cleveland, plant C inbound from V-244 in Akron, plant D inbound from V-301 in Toledo. Three lanes feeding three plants from three suppliers within 90 minutes of each other. Nobody can see the full lane map at the level of detail needed to consolidate without rebuilding a master freight model from CSVs.
One screen, five plants, five metrics. The PLT-C row tells the story.
OTIF at 84 percent, OEE at 63 percent, cycle variance pulling up. PLT-C is below the band the rest of the estate holds. Nothing in any single plant ERP would put PLT-C beside PLT-A on the same screen because the source systems do not talk and the chart-of-accounts is not portable. Polymr normalizes the cross-plant view above the ledgers; each plant keeps its own COA on the source side.
| Plant | OTIF percent | OEE percent | Scrap percent | Cycle var pts | Inv days |
|---|---|---|---|---|---|
PLT-A Cleveland Bearing race SAP S/4 | 96.2 | 78 | 1.4 | +0.3 | 38 |
PLT-B Akron Stamped housing Dynamics F&O | 91.4 | 71 | 2.1 | +0.8 | 52 |
PLT-C Toledo Bearing race Dynamics F&O | 84.1 | 63 | 3.6 | +1.4 | 67 |
PLT-D Buffalo Sub-assembly NetSuite | 88.6 | 69 | 2.8 | +0.6 | 44 |
PLT-E Erie Stamped housing Sage X3 | 93.7 | 74 | 1.9 | +0.4 | 41 |
The press over 100 percent in PLT-C has slack three plants over.
The heatmap below stacks four plants against five shared work-center types. PLT-C press-22 is at 109 percent, the welding cell is at 88 percent, finishing is at 92 percent. PLT-D has slack across the full row.
An inter-plant transfer of 320 units off the PLT-C press-22 backlog onto PLT-D press-22 closes the gap. The freight cost is 14 percent of the expedite spend the alternative would burn. Polymr surfaces the transfer alternative on the consolidated approval queue before the week 23 schedule publishes.
| Plant | WC-MILL-3X | WC-LATHE-04 | WC-PRESS-22 | WC-WELD-01 | WC-FINISH-02 |
|---|---|---|---|---|---|
| PLT-A | 62% | 78% | 94% | 71% | 58% |
| PLT-B | 88% | 41% | 102% | 64% | 73% |
| PLT-C | 97% | 84% | 109% | 88% | 92% |
| PLT-D | 44% | 56% | 68% | 51% | 39% |
One central buyer signature replaces four serial plant controllers.
The queue below is the central buyer's. Three pending consolidated POs, each with the per-plant request roll-up, the consolidated vendor recommendation, the landed-cost delta against the five-plant-PO alternative, and the inter-plant transfer alternative where one exists. The central buyer signs once. Each plant controller receives a posted entry against their own COA on confirmation; the 3-week serial signature cycle compresses to a single day without changing any plant's signing authority.
What this looked like at a four-plant industrial components maker.
- Situation
- Four plants procured overlapping SKUs from partially overlapping vendor lists. Each site ran its own planner and its own approved-vendor file. Central operations saw a roll-up only at month-end.
- What was breaking
- PMR-4124 bearing race was bought by three plants from three vendors at unit costs spanning 18%. Plant-level POs were sent before central could batch them. Vendor consolidation was a slide deck, never an action.
- Planning + purchasing
- Quote-to-procure
- Margin and bottleneck analysis
Three workflow families do the heavy lifting across a multi-plant estate.
- Planning + purchasing
Cross-plant net requirements run against live on-hand from all plant ERPs. MOQ-break math runs against vendor tiers; consolidated draft POs land in the central approval queue with per-plant qty and dock split already resolved.
- Margin + bottleneck analysis
Per-SKU margin and the bottleneck signal per work-centre per site surface live, on the same rhythm as the cross-plant approvals queue. Vendor quotes score across all plants on a rolling baseline; savings publish as workflow outcomes rather than quarterly slides.
- Delay recovery
When V-218 slips on a multi-plant lane, the cascade across all five plants resolves once. Inter-plant transfer alternatives surface where they cover the gap. The expedite spend that today fragments across five plant buyers consolidates to a single decision packet.
Three siblings that share the same multi-site shape.
- High-volume
Plant capacity matters at multi-site scale. The bottleneck shifts between plants weekly. The capacity-planning patterns from the high-volume page extend across the cross-plant view here.
- Process
For multi-plant batch consolidation. When a process plant runs across four sites, the cross-plant batch genealogy and consolidated supplier negotiation patterns apply directly.
- Discrete
For cross-plant revision sync. When an engineering rev lands on plant A, the open POs in plants B, C, and D need to update on the same workflow rather than three plant emails.