01 — Notebook Information & Scope
“The objective of all manufacturing control systems is to have the right material at the right place at the right time in the right quantity, without carrying excessive inventory.” — George Plossl.
- Domain: Operations & Supply Chain Systems
- Subject: Enterprise Resource Planning
- Pedagogical Lead: Prof. Rahul Altekar
- Core Reference Model: Plossl’s Manufacturing Flow, Dependent vs Independent Demand, and Multi-Level BOM Explosion
02 — Learning Map
[Independent Customer Demand] ──> [Master Production Schedule (MPS)]
│
▼
[Inventory On-Hand / On-Order] ──> [MRP Calculation Engine] <── [Bill of Materials (BOM)]
│
▼
[Planned Order Releases]
│ │
▼ ▼
[Purchase Orders] [Shop Work Orders]
(Procurement) (Manufacturing)
03 — Core Concepts: Dependent vs. Independent Demand
The modern ERP manufacturing engine is grounded in Joseph Orlicky’s fundamental demand classification:
- Independent Demand:
- Demand for finished goods and service spare parts that originates directly from external market customers.
- Inherently uncertain, stochastic, and must be forecasted using statistical techniques (moving averages, exponential smoothing, Holt-Winters).
- Dependent Demand:
- Demand for sub-assemblies, fabricated components, and raw materials that directly depends on the production schedule of parent items.
- Perfectly deterministic once the Master Production Schedule (MPS) is committed. It must be calculated, never forecasted.
Anti-Pattern in Manufacturing Planning
Forecasting raw material demand using statistical time-series models while ignoring the Bill of Materials produces catastrophic Bullwhip distortion and inventory gluts. Dependent demand must always be calculated via BOM explosion.
04 — The Bill of Materials (BOM) & Indented Explosion
A Bill of Materials (BOM) is an engineering master record that defines the hierarchical recipe of components, sub-assemblies, and raw materials required to produce one unit of a finished product.
Multi-Level Indented Tree Example: Industrial Pump (Model P-100)
Level 0: Industrial Pump P-100 (1 unit)
├── Level 1: Pump Housing Assembly (1 unit, Lead Time: 2 weeks)
│ ├── Level 2: Cast Iron Casing (1 unit, Lead Time: 4 weeks) [Purchased]
│ └── Level 2: Flange Bolts (8 units, Lead Time: 1 week) [Purchased]
├── Level 1: Impeller Sub-Assembly (1 unit, Lead Time: 1 week)
│ ├── Level 2: Machined Stainless Impeller (1 unit, Lead Time: 3 weeks) [Manufactured]
│ └── Level 2: Drive Shaft (1 unit, Lead Time: 2 weeks) [Purchased]
└── Level 1: Electric Motor Drive (1 unit, Lead Time: 6 weeks) [Purchased]
Low-Level Coding
When an identical raw component (e.g., standard M8 bolts) appears at multiple different tiers of a complex BOM, MRP engines assign it the lowest level code (LLC) across all parent paths. The calculation engine postpones calculating requirements for that item until all higher-level parent explosions have completed, avoiding fragmented purchasing runs.
05 — The Master Production Schedule (MPS) Interface
The Master Production Schedule (MPS) translates disaggregated sales forecasts and firm customer sales orders into an operational build schedule stating exactly what end items will be assembled in weekly time-buckets over a planning horizon.
- Rough-Cut Capacity Planning (RCCP): Verifies whether bottleneck work centers, machine hours, and labor hours can support the proposed MPS prior to running full MRP.
- Planning Time Fence (PTF):
- Frozen Zone (0 to 4 weeks): No modifications permitted without executive authorization.
- Slushy Zone (4 to 8 weeks): Product mix adjustments permitted; total volume fixed.
- Liquid Zone (> 8 weeks): Open to algorithmic scheduling and demand revisions.
06 — Deterministic MRP Matrix Arithmetic Walkthrough
The MRP matrix computes requirements per time bucket based on the fundamental replenishment formula:
Net Requirements Balance Equation
FORMULAWhen Gross Demand exceeds on-hand available stock and firm incoming receipts, a Net Requirement is triggered. This immediately dictates a Planned Order Receipt, which is offset backward by lead time into a Planned Order Release.
$NR_t$Net Requirements in time bucket t (actual deficit to fulfill)$GR_t$Gross Requirements in time bucket t (derived from parent planned orders)$PA_{t-1}$Projected Available inventory balance carried over from bucket t-1$SR_t$Scheduled Receipts arriving in bucket t (firm open purchase or shop orders)$SS$Safety Stock threshold buffer maintained across all periodsWorked Numerical Problem
- Part: Sub-Assembly B-200
- Lead Time ($L$): 2 Weeks
- Lot Sizing Rule: Fixed Order Quantity ($FOQ = 100$ units)
- Initial On-Hand Inventory ($PA_0$): 40 units
- Safety Stock: 0 units
| Bucket (Week) | W1 | W2 | W3 | W4 | W5 | W6 | W7 | W8 |
|---|---|---|---|---|---|---|---|---|
| Gross Requirements ($GR$) | 30 | 20 | 50 | 40 | 60 | 30 | 80 | 20 |
| Scheduled Receipts ($SR$) | 0 | 50 | 0 | 0 | 0 | 0 | 0 | 0 |
| Projected Available ($PA$) | 10 | 40 | 90 | 50 | 90 | 60 | 80 | 60 |
| Net Requirements ($NR$) | 0 | 0 | 10 | 0 | 10 | 0 | 20 | 0 |
| Planned Order Receipts ($PORc$) | 0 | 0 | 100 | 0 | 100 | 0 | 100 | 0 |
| Planned Order Releases ($PORl$) | 100 | 0 | 100 | 0 | 100 | 0 | 0 | 0 |
Step-by-Step Computational Trace:
- Week 1: Available $40 - 30 = 10$. No deficit ($NR = 0$).
- Week 2: Available $10 + 50 (SR) - 20 = 40$. No deficit ($NR = 0$).
- Week 3: Demand is 50. Carried stock is 40. Deficit is $50 - 40 = 10$ ($NR = 10$). Under $FOQ = 100$, we trigger $PORc = 100$. Ending stock $PA_3 = 40 - 50 + 100 = 90$.
- Lead Time Offset: Because lead time is 2 weeks, the Week 3 receipt must be released in Week 1 ($PORl_1 = 100$).
- Week 4: Available $90 - 40 = 50$. No deficit ($NR = 0$).
- Week 5: Demand is 60. Carried stock is 50. Deficit is $60 - 50 = 10$ ($NR = 10$). $PORc_5 = 100$, offset back to Week 3 ($PORl_3 = 100$). Ending $PA_5 = 90$.
07 — Closed-Loop MRP vs. MRP II vs. ERP Evolution
The Evolutionary Shift: MRP I to Closed-Loop MRP to MRP II
COMPARISONCalculates material requirements, then evaluates plant capacity via Capacity Requirements Planning (CRP). If shop-floor work centers are overloaded, the master scheduler loops back to adjust the MPS manually.
Integrates material flows directly with financial ledgers, labor scheduling, machine maintenance, tool cribs, and business planning. Cost of goods sold (COGS) and inventory valuation update synchronously with physical movements.
08 — Manufacturing Master Data Governance
The reliability of an ERP manufacturing run depends on the precision of four foundational master data tables:
- Material Master: Defines unit of measure, procurement type (Make vs Buy), planning strategy (Make-to-Stock vs Make-to-Order), lead times, and lot-sizing rules.
- Bill of Materials (BOM): Defines component validity dates, scrap factors (percentage waste expected during assembly), and alternate component substitutions.
- Work Center Master: Defines machine capabilities, queue time, setup time, labor headcount, and standard hourly cost rates.
- Routing Master: Specifies the exact sequential operational path (e.g., Op 10: Laser Cutting $\rightarrow$ Op 20: CNC Bending $\rightarrow$ Op 30: Robotic Welding $\rightarrow$ Op 40: Powder Coating) and standard setup/run hours per batch.
If engineering changes a component but fails to update the ERP BOM revision code, the factory floor builds obsolete revisions while procurement issues purchase orders for incorrect components. Master Data Governance is not administrative; it is operational survival.
09 — Mini-Case: Precision Hydraulics Assembly Overhaul
- Company: Precision Hydraulics Ltd. (Medium-duty hydraulic cylinders)
- Pre-ERP Situation: Factory planners utilized individual Excel spreadsheets for component calculations. Lead times were assumed to be a uniform 3 weeks across all 140 parts. Stockouts of critical O-ring seals ($0.50 unit cost) halted final testing of $2,400 cylinders, leading to 28 days average delivery delay and $1.2M in buffer stock.
- ERP Intervention:
- Implemented SAP Production Planning (PP) with low-level coding.
- Stratified lead times: local rubber seals (2 days), forged chrome shafts (6 weeks).
- Tied vendor purchase order releases directly to MRP planned order releases.
- Results:
- WIP inventory reduced by 34%.
- On-time cylinder shipments increased from 61% to 94.5%.
- Total warehouse carrying costs reduced by $380,000 annually.