MPS, MRP, CRP for Paper Mills: How Production Planning Actually Works
The three-layer planning stack every paper mill needs — explained in plain language with paper-specific examples.
Every paper mill runs production planning — but most run it manually in Excel and call it "scheduling." A real production planning stack has three layers: MPS, MRP, CRP. Done right, they form an integrated planning system that catches bottlenecks before they happen.
MPS — Master Production Schedule
The top layer. Answers: what will we produce, on which machine, when?
Inputs:
- •Sales orders (firm + forecast)
- •Machine availability and PM schedules
- •Customer priorities and credit status
- •Grade transition costs and constraints
Output: A time-phased plan typically covering 4–12 weeks. "PM1 makes 100 GSM kraft from Mon to Wed, then transitions to 120 GSM Thu–Fri."
For Indian paper mills: MPS is where you decide which orders make the cut, which get delayed, which need overtime.
MRP — Material Requirements Planning
The middle layer. Answers: what raw materials do we need, and when?
Inputs:
- •MPS (what we're making, when)
- •BOMs (recipe per grade)
- •Existing inventory (pulp, chemicals, packing)
- •Supplier lead times
Output: Time-phased material requirements. "Need 80 tons of OCC by Wednesday. Need 5 tons of starch by Tuesday. Issue PR for resin by Friday for next week's run."
For Indian paper mills: MRP prevents the all-too-common scenario of starting a customer order Tuesday morning only to discover Monday night that the right chemical isn't in stock.
CRP — Capacity Requirements Planning
The validation layer. Answers: does our planned production fit our capacity?
Inputs:
- •MPS proposed schedule
- •Machine capacity (effective hours considering downtime targets)
- •Labour availability
- •Bottleneck operations
Output: Capacity violation alerts. "PM1 is overloaded by 8 hours next week. Either delay an order or run overtime."
For Indian paper mills: CRP catches over-commitment early. Without it, sales teams promise dates that operations can't meet, then the mill scrambles at month-end.
How they connect
Sales Orders → MPS → MRP → Purchase Orders
↓
CRP → Schedule Adjustments
It's a closed loop. Sales orders feed MPS, MRP issues purchase orders, CRP validates capacity. Changes anywhere ripple through.
Paper-specific nuances
Grade transitions cost real money
Switching from 100 GSM to 80 GSM on PM1 takes 2 hours of off-spec production = 4 tons of broke. MPS must account for transition costs when sequencing grades.
Furnish constraints across PMs
A 200 TPD mill with PM1 (kraft) and PM2 (writing) shares pulp prep capacity. If PM2 needs hardwood furnish at 10 AM and PM1 needs softwood at 11 AM, stock prep is the bottleneck. CRP must catch this.
Customer specs aren't just GSM
Customer A wants 100 GSM kraft with BF ≥ 18. Customer B wants the same paper but in 800 mm reels. The deckle plan (downstream of MPS) determines which customers can be combined into one run.
Trim waste is part of the plan
A standalone MPS might say "make 50 tons of 100 GSM kraft." A paper-mill-aware MPS knows that without integrating deckle optimization, you'll waste 8% to trim. So you must produce 54 tons to ship 50.
How Indian mills typically run today
- •Small mills (<20 TPD): Whiteboard scheduling, no formal MPS/MRP/CRP. Owner intuition.
- •Mid mills (20–100 TPD): Excel for MPS, manual MRP, no CRP. Frequent firefights.
- •Large mills (>100 TPD): ERP-driven MPS/MRP, but often with stale data and manual workarounds.
The gap between Excel-based planning and real integrated MPS/MRP/CRP is where most operational improvement opportunity sits.
What modern integrated planning looks like
In Papyrus BPApp:
- •Sales orders flow into MPS automatically — no re-entry
- •MPS auto-sequences orders using customer priority, deckle compatibility, grade transitions
- •MRP fires daily, issuing PRs for materials going below safety stock
- •CRP runs every shift, flagging capacity violations 2–4 weeks ahead
- •Deckle Optimizer is part of the plan, not a downstream tool
Result: planners spend hours on exception handling, not data entry. Mill output increases 8–15% without buying new equipment.
See how Papyrus BPApp solves this
Book a demo tailored to your mill — we'll show you exactly the workflows discussed in this article.
Request a Demo