What Is Batch Scheduling?
A juice plant, a paint factory or a cosmetics maker does not produce one unit at a time, and its lines do not run one product nonstop all year. Instead, the plant mixes a defined quantity of material in a tank or mixer, processes it under the same settings, fills it, then cleans the line and sets up for the next product. That quantity, made in one cycle under the same conditions, is a batch, and every unit in it carries the same batch number.
Batch scheduling is the recurring decision that answers four questions. How much do we make per batch? In what order do we run different products? When does each batch start? On which line, tank or crew? A good answer balances goals that pull against each other: delivering customer orders on time, avoiding excess stock that may expire, minimizing hours lost to cleaning and changeovers, and protecting quality and compliance.
In many plants this schedule lives in the production manager's head or in a spreadsheet updated every morning. That works while the product range is small and the manager is in the building. As products, customers and constraints multiply, manual scheduling becomes fragile. This article explains the concept with worked numbers, shows how an ERP system supports it, and sets realistic expectations about the benefits.
Batch Production Compared With Other Modes
| Characteristic | Batch | Continuous | Discrete |
|---|---|---|---|
| Flow | Defined quantities move through stages together | Constant flow of one or a few products | Individual units assembled |
| Examples | Food, pharma, chemicals, paints | Cement, petrochemicals, refining | Furniture, appliances, equipment |
| Traceability unit | Batch or lot number | Time windows or tanks | Serial number |
| Changeover impact | Very high | Low | Medium |
| Scheduling focus | Size, sequence, shelf life | Run rate and maintenance | Bill of materials and capacity |
Many plants combine modes. A food factory may cook in batches and then fill on a near-continuous line. What matters is that the system knows where a batch starts and ends so that traceability and costing stay correct.
Decision One: Batch Size
Large batches mean fewer cleanings and setups, but they raise inventory, holding cost and expiry risk. Small batches give flexibility and lower stock, but consume more hours in changeovers. The classic tool for finding the balance is the economic order quantity, published by Ford W. Harris in 1913. In 1918 E. W. Taft extended it to production as the economic production quantity (EPQ), which accounts for the fact that stock builds up gradually during a run while part of it is being consumed.
The formula: optimal batch size = square root of (2 x annual demand x setup cost) / (holding cost per unit x (1 - daily demand rate / daily production rate)).
A hypothetical example
A plant sells 100,000 bottles of a drink per year, about 400 a day over 250 working days. When running this product, the line makes 1,000 bottles a day. Each setup (cleaning, sanitizing, test run, start-up scrap) costs SAR 500, and holding one bottle for a year costs SAR 2.
- Numerator: 2 x 100,000 x 500 = 100,000,000
- Denominator: 2 x (1 - 400 / 1,000) = 2 x 0.6 = 1.2
- Result: square root of 83,333,333 ≈ 9,130 bottles per batch, roughly 11 batches a year.
| Policy | Batches a year | Annual setup cost | Annual holding cost | Total |
|---|---|---|---|---|
| Weekly batch (about 1,920 bottles) | 52 | SAR 26,000 | about SAR 1,150 | about SAR 27,150 |
| EPQ (about 9,130 bottles) | 11 | about SAR 5,480 | about SAR 5,480 | about SAR 10,960 |
Holding cost here is based on average inventory, which is half the peak stock that builds during a run. The gap is roughly SAR 16,000 a year for one product. But the formula is only a starting point. In practice the system must respect constraints such as tank capacity, customer minimums, shelf life and cold-room space. If the product has a 60-day shelf life, a batch covering 23 days of demand is fine; a batch covering 90 days means writing some of it off. For the underlying model, see our guide to economic order quantity.
Decision Two: Batch Sequence
Changeover time usually depends on what ran before and what runs next. In a paint plant, going from white to beige needs a quick rinse; going from black to white needs a full clean. In food, moving from an allergen-free product to one containing peanuts is easy, while the reverse needs a documented clean and verification.
A changeover matrix
| From / To | White | Beige | Brown | Black |
|---|---|---|---|---|
| White | 0 | 20 min | 25 min | 30 min |
| Beige | 90 min | 0 | 20 min | 25 min |
| Brown | 100 min | 80 min | 0 | 20 min |
| Black | 120 min | 100 min | 80 min | 0 |
Running all four colors weekly as white, beige, brown, black costs 20 + 20 + 20 = 60 minutes within the week plus 120 minutes to return to white the following week: 180 minutes. A haphazard order such as black, white, brown, beige costs 120 + 25 + 80 = 225 minutes plus 25 minutes back to black next week: 250 minutes. That is 70 minutes a week, more than 60 line hours a year on a single line, and the gap grows with more colors and lines.
When the system stores a changeover matrix per line, it can propose the lowest-cost sequence and warn the planner when a rush order breaks it, so the cost of that decision is visible.
Decisions Three and Four: Timing and Resources
Timing is driven by the customer due date minus production, testing and packing time, with shelf life in mind. Scheduling can be backward, starting from the due date and working back to the latest start, or forward, starting as soon as possible and seeing when the batch finishes. Backward scheduling keeps inventory low; forward scheduling builds in a safety margin.
Resource allocation means choosing the right line or vessel, confirming materials will be available on the day, confirming a qualified crew is on shift, and avoiding clashes with planned maintenance. This is where the production calendar matters: working days, shifts, holidays and shorter Ramadan hours. A system that knows Muslim employees work six hours a day during Ramadan will not produce an impossible plan for that month. See Tranquil's production calendar.
A Worked Week in a Juice Plant
Imagine a small juice plant in Qassim with one mixing and filling line running two shifts, 16 hours a day, five days a week: 80 hours. It makes four flavors: apple, orange, mango and a mixed juice containing milk. Figures are hypothetical.
| Flavor | Weekly demand (bottles) | Production hours | Note |
|---|---|---|---|
| Apple | 24,000 | 16 | Lightest color and taste |
| Orange | 30,000 | 20 | Highest volume |
| Mango | 18,000 | 12 | Stronger color and taste |
| Mixed with milk | 12,000 | 8 | Contains an allergen; full clean afterward |
Production needs 56 of the 80 hours. Sequenced light to heavy, apple, orange, mango, then the milk product at the end of the week, the line needs only short rinses between the first three and one full clean after the milk run. If a full clean takes four hours and a rinse one hour, changeovers total about six hours, leaving 18 hours of buffer for maintenance and extra orders.
Insert the milk product midweek for a customer and the line needs an extra full clean before returning to the other flavors, at least four more hours. That is not a reason to refuse the order, but the planner should see the cost before agreeing, and may negotiate a date that fits the sequence.
Linking the schedule to purchasing
The weekly plan drives demand for concentrates, sugar, bottles, caps and cartons. When planning runs in the same system as purchasing, requirements appear automatically against stock and open orders. If the bottle supplier needs ten days, a plan approved less than two weeks out will run short. Many plants therefore freeze the next week, changeable only with approval, and keep the following two weeks flexible.
Seasons and Ramadan
Juice demand rises in Ramadan and summer, while working hours for Muslim staff fall to six a day during Ramadan. Good scheduling builds stock ahead of the season in slightly larger batches, within shelf-life and cold-storage limits, rather than trying to produce everything in the same month with less capacity.
How ERP Supports Batch Scheduling
- Recipes and bills of materials: define each batch's ingredients by ratio, so material needs are calculated for any batch size, including expected losses.
- Material availability checks: before a plan is released, the system checks stock and open supplier orders and flags batches at risk.
- Capacity view: daily load per line, with alerts when it exceeds capacity.
- Production orders with batch numbers: each order generates a batch number linking consumed materials to finished goods.
- Quality checks: in-process and final inspections, with the batch held until results are approved.
- Barcodes: material issue and finished-goods receipt by scan, cutting manual entry errors.
- Costing: material, labor and overhead collected per batch and compared with standard cost.
Tranquil's manufacturing pages cover production scheduling, multi-level BOM, material planning, quality checks and barcode integration.
Spreadsheet scheduling versus scheduling inside ERP
| Aspect | Spreadsheet | Inside ERP |
|---|---|---|
| Material availability | Phone the warehouse | Stock and open POs visible on the plan |
| Rush order | Manual reshuffle, unclear knock-on effect | Impact on other batches shown before release |
| Batch number | Typed by hand, may duplicate | Generated and linked to the production order |
| Cost | Estimated at month-end | Collected on the batch as work happens |
| Continuity | One person's file | Rules the whole team can see |
A spreadsheet is not wrong in itself; skilled planners run entire plants on one. The problem is that it is disconnected from everything else, so every decision needs a manual check with stores, purchasing and quality, and every delay in that check turns into line downtime or an inaccurate promise to a customer.
Benefits of a Structured Batch Schedule
Full batch traceability
When a complaint or quality issue appears, you can see which raw materials and suppliers went into the batch (backward trace) and which customers received it (forward trace). For food, the Implementing Regulations of the Saudi Food Law require factories and warehouses to keep a traceability guide, hold traceability records for at least 180 days, notify SFDA of suspected non-compliant food and recall it. For medicines, SFDA requires a GS1 DataMatrix code carrying product code, batch, expiry and serial number. More in our article on batch tracking.
More hours from the same lines
As the paint example shows, sequencing alone can free dozens of hours a year per line without buying equipment.
Lower stock and less expiry waste
Batch sizes tied to real demand and shelf life reduce write-offs and near-expiry discounting.
Better delivery promises
When sales can see line load and scheduled batches, they quote realistic dates instead of promises that need an apology call later.
Clearer product costs
Very small batches carry high setup cost per unit. IAS 2 requires fixed production overheads to be allocated on normal capacity; the per-unit share is not increased because of low output or idle plant, and unallocated overhead is expensed. Accurate changeover and downtime records help apply this correctly and price small orders properly.
Less dependence on one person
With rules and changeover matrices recorded in the system, another planner can step in when the manager is away.
Common Mistakes
- Fixing batch sizes for years: demand, setup and holding costs change; review sizes periodically.
- Ignoring shelf life: the EPQ formula knows nothing about expiry dates.
- Posting actual consumption at month-end: this weakens both traceability and costing. Issue materials to the batch as it happens.
- Accepting rush orders blind: every exception that breaks the sequence has a cost the decision maker should see.
- Separating quality from scheduling: a batch awaiting lab release is not ready to ship, even if it is packed.
KPIs for Schedule Quality
| KPI | Calculation |
|---|---|
| Schedule adherence | Batches completed on time / batches scheduled |
| Changeover ratio | Cleaning and setup hours / available hours |
| Batch yield | Actual good quantity / planned quantity |
| Remaining shelf life at dispatch | Average days per product |
| On-time in-full | Complete on-time customer orders / total orders |
Frequently Asked Questions
How is batch scheduling different from batch tracking?
Scheduling is a planning decision about size, sequence, timing and resources before production. Tracking is the record linking each batch to its materials and customers afterward. Good scheduling makes tracking easier because the batch number is created and linked to materials from the start.
Does EPQ work for every product?
It is a good starting point for products with fairly stable demand. Seasonal, short shelf-life or make-to-order products need adjustments or different rules.
How often should batch sizes be reviewed?
Quarterly suits most plants, with an immediate review when demand or setup cost changes significantly or a new line is added.
Can a small plant schedule batches without a dedicated system?
Up to a point. One line and ten products can be handled in a well-kept spreadsheet. With several lines, batch traceability and costing, linking production, inventory, purchasing and accounting in one system is what prevents errors.
How should rush orders be handled?
Reserve a flexible window in the weekly plan, have the system show the cost of inserting the order in extra changeover hours and delayed batches, then decide based on the order's value.
To discuss planning and scheduling for your plant, contact the Tranquil team.