Detailed Guide on ERP in Production Planning

What Production Planning Actually Has to Solve

Every factory answers the same four questions each week: what do we make, how many, when, and with which machines, people and materials? Production planning is the discipline of answering them before the line starts running, so the right product comes off the line in the right quantity on the date you promised, without raw material piling up in the warehouse or a cell standing idle while it waits for one missing component.

Most small manufacturers start with a spreadsheet the production manager rebuilds every Sunday. That works while the product range is narrow and demand is steady. It breaks as soon as you add more SKUs, customers with different delivery terms, and suppliers with lead times that range from three days to six weeks. One changed order means recalculating everything by hand, the stock figures in the sheet stop matching the warehouse, and purchasing finds out about shortages when it is already too late to fix them cheaply.

An ERP system changes the starting point. Instead of a separate planning file, sales orders, inventory, purchasing, production and costing share one database, so the plan is built on live numbers rather than last week's export. This guide walks through the planning levels, shows how material requirements are calculated with a full worked example, explains capacity checks and scheduling, lists the KPIs worth tracking, and finishes with a practical rollout sequence for a mid-sized plant.

The Planning Hierarchy: From Annual Volume to a Work Order

A common mistake is to treat planning and scheduling as the same thing. They are different layers, each with its own time horizon and its own question:

Each layer feeds the next. If the aggregate plan is unrealistic, no amount of clever shift-level scheduling will rescue it. Good planning works top down, then gets corrected bottom up whenever scheduling reveals a constraint nobody anticipated.

Master Data: What the System Needs Before It Can Plan

ERP planning is only as good as the master data underneath it. Most planning projects that stall do so here, long before anyone argues about algorithms. Before the first planning run, make sure you have:

• Bills of materials: every component and its quantity per unit, across all levels. A finished product is made of sub-assemblies; sub-assemblies are made of raw materials. An error at any level is multiplied in every work order.
• Routings: the sequence of operations, the work center for each, and setup and run time per unit.
• Work center capacity: number of machines, shift hours, working days, planned maintenance, and a realistic efficiency factor rather than the brochure figure.
• Lead times: the real time from issuing a purchase order to having the material received and inspected, for each item and major supplier.
• Stocking policies: safety stock, minimum order quantities and pack multiples imposed by suppliers.
• Accurate stock balances: available, allocated and on quality hold, by warehouse.

If your BOMs still live in scattered files, start there. Our guide to the bill of materials explains how to structure them so planning can rely on them.

How ERP Calculates Material Requirements: A Worked Example

Take a hypothetical office furniture plant in Dammam that builds an office chair. The simplified BOM per chair:

• Steel base: 1 (made in-house from steel tube)
• Upholstered seat: 1 (bought from a local supplier)
• Casters: 5
• Screws: 12

Each steel base consumes 2.4 meters of steel tube.

The master schedule calls for 800 chairs in week 5. The current position looks like this:

The logic on every row is the same: net requirement equals gross requirement, minus on-hand, minus scheduled receipts, plus safety stock. Notice that steel tube is not calculated from 800 chairs. It comes from the net requirement for bases: 700 bases × 2.4 m = 1,680 m. That is a BOM explosion. The system works down one level at a time and nets each level before passing demand to the next.

Then lead times are applied. Chairs are needed in week 5 and bases take a week to make, so the base work order must start in week 4. Steel tube takes two weeks to arrive, so its purchase order must go out in week 2. Casters take four weeks, which means the purchase order has to be released in week 1, which is now. If someone spotted the caster shortage by hand in week 3, the whole order would ship at least two weeks late.

For one product this is a page of arithmetic. For a plant with several hundred active items it is thousands of lines, recalculated every time an order changes. That is what an MRP engine does on every run. The output is a list of planned orders: purchase orders for bought items and work orders for made items, each with a release date and a need date. The planner reviews and firms them up rather than building them from scratch.

Lot sizing changes the answer

The example orders exactly the net requirement. In practice, the caster supplier may sell in cartons of 500, pushing the order from 2,600 to 3,000. You might also combine three weeks of demand into one order to save on freight. These rules sit on the item record and the system applies them automatically. For the trade-off between ordering cost and holding cost, see our article on economic order quantity.

Capacity Planning: Can the Shop Floor Deliver This Plan?

MRP assumes capacity is available. Capacity planning tests that assumption. Continuing the example: welding a base takes 6 minutes, with one hour of setup per batch. The plant has two welding stations on one 8-hour shift, six days a week, at a realistic 85% efficiency.

• Available capacity per week: 2 stations × 8 hours × 6 days × 0.85 = 81.6 hours.
• Load for 700 bases: 700 × 6 minutes = 70 hours, plus 1 hour setup = 71 hours.
• Utilization: 71 ÷ 81.6 = 87%.

The plan is feasible, but tight. If another order needing 20 welding hours lands in the same week, the load goes past 100% and the planner has to choose: overtime, a second shift, pulling part of the work forward into week 3, or renegotiating a delivery date. The system does not make that decision. It surfaces the problem two weeks out instead of on the day.

When overtime is the answer, cost it properly. Under the Saudi Labor Law, normal hours are capped at 8 per day or 48 per week (6 and 36 during Ramadan for Muslim workers), and an overtime hour is paid at the hourly wage plus 50% of the basic wage, with annual overtime limited to 720 hours unless the worker agrees otherwise. A plan that depends on permanent overtime is really a capacity shortfall with a higher price tag, and that cost should be visible when the decision is made.

Detailed Scheduling: Sequencing Work on the Line

Once total capacity is confirmed, scheduling decides which order goes first and on which machine. There are two basic approaches:

• Forward scheduling starts each order as early as possible and calculates a finish date. It answers the sales question: when can we deliver?
• Backward scheduling starts from the due date and works back to the latest possible start for each operation. It keeps work in progress low but leaves no buffer for disruption.

Sequence on each machine matters too. In plastics, moving from a light color to a dark one is easier than the reverse, and every mold change costs time. Grouping similar colors can save hours of changeover a week. In food plants, cleaning cycles and allergen rules also dictate the order. A good schedule balances three goals that often pull against each other: meeting due dates, minimizing setups, and keeping work in progress low.

Production scheduling works alongside a production calendar of working days, holidays and shifts, so no order is scheduled on a day the plant is closed.

Execution and Feedback: Closing the Loop

A plan that is never compared with what actually happened loses its value within a week. ERP ties each work order to events on the floor:

1. Release: the system allocates the required materials and produces a pick list for the warehouse.
2. Material issue: actual issues are recorded, ideally by barcode scan, and stock balances drop immediately.
3. Progress reporting: quantities completed at each operation, actual time, and scrap with reasons.
4. Quality checks: inspection points at defined stages, with the order held until results are approved.
5. Finished goods receipt: output enters stock and its actual cost is built from materials, labor and overhead.
6. Variance analysis: actual versus standard consumption and time, feeding corrections into master data before the next run.

That last step is the one most plants skip. If real tube usage is 2.6 m per base rather than 2.4, every future MRP run will under-order by about 8%, and shortages will keep appearing with nobody sure why. Correcting BOMs and routings from actual data is part of planning, not an optional extra.

KPIs That Show Whether Planning Is Working

Feeling that planning has improved is not enough. These measures come straight out of system data:

Pick three or four to start and review them weekly in the planning meeting. A KPI nobody reviews changes nobody's behavior.

Common Problems and How to Handle Them

Last-minute demand changes

A key customer pulls an order forward by a week. The answer is not to replan every hour. Set a frozen zone in the master schedule, for example two weeks, that changes only with the production manager's approval, and show the impact on other orders before any change is accepted.

Late suppliers

When a supplier slips, the planner needs to know immediately which work orders are affected. Pegging purchase orders to the work orders that need them makes the impact visible and lets you re-sequence or source elsewhere. Our article on lead time management goes deeper.

Unreliable stock balances

MRP trusts whatever balance the system holds. If the balance is wrong, so are the planned orders. Regular cycle counts on critical items and real-time recording of every issue are non-negotiable.

Planner resistance

A planner attached to a personal spreadsheet will keep running it in parallel unless the system earns trust. Run both methods side by side for a month, compare results openly, and fix whatever data caused each difference. Trust comes from numbers, not memos.

A Practical Rollout Sequence

1. Pick a pilot scope: one line or one product family that represents a meaningful share of revenue.
2. Clean master data: BOMs, routings, lead times, and stock balances after a physical count.
3. Set policies: safety stock, lot sizes, frozen zone and MRP run frequency.
4. Train planners and supervisors: on the logic, not just the screens, so they understand why the system proposed what it did.
5. Run in parallel: at least four weeks, reviewing differences weekly.
6. Scale up: once KPIs stabilize, add lines and product families in stages.

The Saudi National Industrial Strategy, launched in October 2022, aims to triple industrial output and reach 36,000 factories by 2035. More factories means tougher competition on price and delivery, and a plant that plans from live data has a real edge in tenders and supply contracts.

How Tranquil Supports Production Planning

Tranquil's manufacturing solutions cover multi-level BOMs, material planning, production scheduling, quality checks, barcode integration and a production calendar. They sit in the same cloud ERP as inventory, purchasing and accounting, so material requirements reach purchasing, issues update stock balances, and work order costs flow into the books. Tranquil's team also customizes screens, reports and approval levels to match how your plant actually runs.

Frequently Asked Questions

What is the difference between MRP and ERP?

MRP is a specific function that calculates which materials are needed and when, based on the production plan, BOMs and stock. ERP is the wider system that includes MRP alongside sales, purchasing, inventory, accounting and HR, all working from the same data.

How often should MRP run?

It depends on how fast demand moves. Most mid-sized plants run it weekly; some run it daily when orders are volatile. What matters is a fixed rhythm everyone knows.

Is ERP-based production planning worth it for a small plant?

Yes. Small plants often feel the benefit first, because a single missing material can stop everything. You can start with BOMs and material planning, then add capacity planning and scheduling later.

What is the single biggest success factor?

Master data accuracy, especially BOMs, stock balances and lead times. The best planning engine produces wrong answers from wrong inputs.

How do I get started with Tranquil?

Contact the Tranquil team to review how your plant plans today and get a quote within 24 hours based on the modules and users you need.