top of page
Company Type

Columbia Pipe, a commercial pipe distributor (250K active SKUs, 18 DCs); rescue of a failed JDE EnterpriseOne implementation; the SIOP framework was architected inside the recovery.

What Drove the Need
  • JDE ranks ABC by sales dollars, margin, or cost — never by unit volume — so high-volume/low-cost pipe items were systematically mis-stocked

  • Procurement planning and purchasing execution for 250K SKUs across 18 DCs were manual and error-prone

Business Impact
  • ~99% of manual procurement planning and purchasing execution eliminated (Case 21)

  • Monthly ABC re-ranking cut from 16+ hours to ~9 minutes with a single custom index

Systems/Data Integration
  • Writes ABC rankings, lead times, safety stock, and planning codes directly to JDE item/branch and planning tables under pessimistic locking

  • Companion to the SIOP algorithm suite (SE-12) and DSI barcoding scripts

Design
  • Build order: user-defined object types → functions → procedures → main scripts, so every procedure composes tested primitives

  • Seven user-defined object and table types; a core package holding all hard-coded values and shared functions; Oracle analytic CUME_DIST() over the sales ledger by branch and item for the cumulative demand distribution

  • Purchasing lead time derived from receipt history (greatest of ceiling median and ceiling average + 2 days, floored at 3); monthly forecast spread to weeks by workday share via a two-pass calendar algorithm

  • BULK COLLECT … LIMIT batching with per-batch commits; pessimistic locking on every JDE update; every written value validated against its UDC table

  • Developed in DV; logic and test-scenario review with Central Planning power users; DBA code review; promotion to CRP with power-user execution rights; then production

Technology Stack
  • Oracle PL/SQL — user-defined objects, packages, procedures, functions

  • JDE EnterpriseOne Oracle schema (F42199 sales ledger, F4102 item/branch, F43090, F41001, F0007)

  • Custom index F42199_CPSCPME01

High-Volume/Low-Cost Inventory ABC Analysis Algorithm

PL/SQL analytic engine ranking every item/branch by unit volume with monthly recategorization — A/B items forecasted, C items on reorder points — feeding planning codes automatically into JDE. Ran hourly through monthly with full change-audit and run logs.

bottom of page