MRP — Material Requirements Planning

Material Requirements Planning (MRP) is the classic procurement and production-planning routine inside an ERP system. From the sales plan and the bill of materials, MRP derives net requirements per item: which materials need to be ordered or produced, in what quantities, and by when. The newer extension MRP II adds capacity planning; APS (Advanced Planning & Scheduling) goes beyond infinite-capacity assumptions and resolves bottlenecks finitely.

MRP logic in four steps

Gross requirements: derived from the production plan plus forecast or open sales orders
Net requirements: gross minus available stock and minus existing open purchase or production orders
Lot sizing: per the configured policy (lot-for-lot, fixed order quantity, period-of-supply, EOQ)
Order proposals: scheduled-back by lead time, output as purchase or production order suggestions

Master data prerequisites

MRP only works as well as its master data. Critical fields per item: lead time, safety stock, lot-size policy, BOM structure with valid component allocations, and routing for in-house production. Many failed MRP projects trace back to inconsistent or out-of-date master data rather than ERP-tool shortcomings.

Practical example: the nightly MRP run

A machinery manufacturer runs its ERP nightly MRP automated: the system walks every open order, explodes each bill of materials to component level, calculates gross demand per material, subtracts available stock and determines net requirement including replenishment lead time. Each morning, 80-150 purchase suggestions await approval; the buyer reviews, consolidates where useful and releases to suppliers. Before MRP automation, the buyer needed 4-5 hours daily for the same task; now 30-60 minutes of validation.

Common pitfalls

MRP runs daily or weekly and is computationally heavy — for large product catalogs with several BOM levels, run times of several hours are not unusual. Variant-rich production benefits from a separate APS layer that resolves capacity bottlenecks.

Selection criteria for US mid-market buyers

Master-data ergonomics: how easy is it to maintain lead times, lot-size policies and safety stocks per item across multiple plants? Bulk-edit and mass-update tooling save days of work each quarter.
BOM depth handling: confirm the system runs your real BOM depth (5-10 levels is common in machinery and industrial equipment) within an overnight maintenance window. Ask for benchmarks against a comparable customer.
Exception handling: useful MRP outputs are not the 10,000 order proposals, but the 50 exceptions that require human judgment. The quality of exception monitors and workbenches separates strong from weak implementations.
Multi-plant and multi-company logic: US mid-market manufacturers often operate 2-5 plants with stock transfers and contract manufacturing. The MRP engine must net requirements across the network, not plant by plant.
Integration with finite scheduling: clarify whether MRP II capacity checks or an embedded APS layer cover your bottleneck resources, or whether you will need a separate scheduling tool (see APS).

Integration patterns

MRP rarely runs in isolation. Typical interface patterns in US mid-market landscapes: ERP ↔ PLM for BOM and engineering-change propagation (see engineering change management), with effectivity dates triggering MRP recalculation. ERP ↔ MES for shop-floor feedback — actual scrap, rework and downtime adjust supply availability inputs to the next MRP run (see MES). ERP ↔ supplier portals (EDI, web-EDI, iPaaS) for releasing order proposals as releases against blanket purchase orders. ERP ↔ demand-planning tools (SAP IBP, Kinaxis, o9, Blue Yonder, Anaplan, RELEX) feed a consensus forecast into MRP rather than relying on ERP-native statistical forecasting. ERP ↔ warehouse management reconciles physical stock with the system-of-record stock that MRP uses (see WMS). Each integration adds value but also a failure surface — latency, master-data inconsistency or mapping errors quickly degrade MRP output quality.

Real-world use cases in the US

Case 1: Midwest special-machinery builder, ~180 employees. Variant-rich one-off machines, BOMs typically 4-6 levels deep with 600-1,200 parts each. MRP runs weekly in batch; project-coordinator review of order proposals takes two hours each Monday morning. Critical success factor: clean engineering change management feeds correct BOMs into MRP; without that discipline the proposals would be unusable.

Case 2: Southeastern consumer-products manufacturer, ~450 employees. Mid-volume, mid-variety assembly with strong seasonality. MRP runs nightly with a consensus forecast from S&OP; lot-sizing combines period-of-supply with minimum-order quantities from contract manufacturers. The buyer-planner team uses exception workbenches rather than reviewing every proposal — about 5% of items each day require intervention.

Comparable terms

MRP sits within a family of planning disciplines. MRP II extends MRP with finite-capacity and financial integration. APS resolves bottleneck constraints beyond infinite-capacity MRP. Demand-to-supply describes the end-to-end process MRP enables. Plan-to-produce covers the manufacturing-execution side downstream of MRP. Just-in-time and kanban are pull-based alternatives that complement or partially replace MRP in stable, high-frequency environments. For accounting integration after MRP-driven procurement, see ERP accounting integration.