Production Scheduling for Carbonated Soft Drink Bottling

This guide shows production planners and operations managers at carbonated soft drink bottling facilities how to model their lines in Schantt — from syrup mixing and inline carbonation through filling, labelling, and palletising — and set up a schedule that respects batch-and-flow routing, dietary changeover penalties, and seasonal shift patterns. By capturing these production constraints in a single model, planners can reduce changeover-driven downtime, eliminate palletiser contention, and automate seasonal capacity transitions without manual spreadsheet rework. The result is a repeatable schedule that keeps both can and PET lines running at target throughput through demand swings and shift changes.

This guide follows a fictional composite company built from industry research on carbonated soft drink bottling; all names, parameters, and figures are illustrative.

Industry context

Carbonated soft drink bottling is a multi-stage process where syrup is batched in mix tanks, carbonated inline, filled into cans or PET bottles, sealed, labelled (for PET), packed into multi-packs, and palletised. The production environment is defined by several structural characteristics: batch syrup mixing feeds continuous-flow downstream stages, requiring careful synchronisation between discrete and rate-based operations; dietary changeovers (diet to sugar, or vice versa) impose cleaning penalties of 60 to 90 minutes on fillers, while same-sweetener flavour changes take 15 to 30 minutes; seasonal demand can swing 60 percent above winter troughs, forcing a switch between single-shift and extended-shift calendars; and filler throughput differs by format — can lines run at 36,000 units per hour, while PET lines run at 18,000 units per hour. This guide assumes make-to-fill syrup coupling — syrup is mixed on the same day it is consumed, tightly coupled to filling. Plants operating make-to-stock with day-tank buffers can treat syrup mixing as a simpler decoupled step.

The filling and sealing stages each have two parallel machines — one for the can line and one for the PET line — while a single shared palletiser handles both lines at the end of the process. Carbonation is inline and flow-matched to the downstream filler, so no holding delay occurs. CO₂ supply for inline carbonation is assumed adequate for concurrent line operation — bulk storage and vaporisation capacity at SMB/mid-market scale are sized to handle peak demand from both lines running simultaneously. Quality checks — brix, CO₂ volume, and sensory panels — are completed in minutes on in-line instruments and do not hold the schedule. Water treatment runs continuously during production hours and is assumed always available at adequate capacity; at SMB/mid-market scale it is not a binding scheduling constraint. The site runs fourteen machines across seven stages, with three product classes and one representative product per class. Two seasonal calendars govern working hours: a single-shift winter pattern running 40 hours per week and a double-shift summer pattern running 96 hours per week, with the switch triggered by demand that rises 60 percent above the winter baseline.

Summit Springs Bottling Co. runs approximately 105 people at a 10,500 m² facility, making three product classes across seven production stages, scheduled by a three-person planning team.

Process overview

flowchart LR
    SM["Syrup Mixing<br/>(Batch)"] --> CA["Carbonation<br/>(Flow)"]
    CA --> FI["Filling<br/>(Flow)"]
    FI --> CS["Capping/Seaming<br/>(Flow)"]
    CS --> LA["Labelling<br/>(Flow)"]
    CS --> MP["Multi-pack<br/>(Flow)"]
    LA --> MP
    MP --> PA["Palletising<br/>(Flow)"]

7-stage process flow for Summit Springs Bottling Co. Carbonation is inline (flow, matched to filler throughput). Can classes skip Labelling via bridging transfer from Capping/Seaming to Multi-pack.

Can classes (Cans-Cola-Sugar) use pre-printed cans and skip the Labelling stage. A bridging transfer time from Capping/Seaming to Multi-pack applies the handoff delay across the skipped span.

Scheduling challenges and how Schantt handles them

In this scenario, the schedule is driven by a demand plan — a list of products and quantities to produce over a planning horizon. Readers whose facility operates on make-to-order or contract-fill demand will follow the same workflow but with shorter horizons and smaller job batches. Schantt optimises the schedule to minimise total production time — the overall completion time across all jobs — scheduling forward from a chosen start date. Retailer cut-off dates and due-date pressure are managed through sequencing and calendar buffers — Schantt does not enforce hard due-date constraints. The optimizer minimizes total production time from the start date rather than penalising late completion against fixed deadlines. This guide assumes a planning horizon of one to two weeks, which is long enough to show seasonal calendar switching and changeover-sensitive sequencing. Shorter or longer horizons follow the same setup.

Schantt offers two optimisation modes. In Auto mode, the user enters products and quantities; the algorithm decides the job sequence, machine assignments, and exact timing. In Semi-Auto mode, the user fixes the production order and may set earliest-start constraints per job; the algorithm optimises machine assignments within that fixed sequence.

What Schantt handles well

• Sequential multi-stage production with transfer times — CSD bottling moves products through an ordered series of stages. Schantt models the full sequence with material handoff delays between each stage, so each downstream step starts only when the prior step has finished and its product has arrived.

• Multi-machine stages (parallel filling lines) — Can and PET filling run in parallel on separate lines. Schantt treats each line as a machine within the same stage and explores which line runs which product in Auto and Semi-Auto modes, choosing the assignment that minimises total production time.

• Mixed batch-and-flow pipelines — Syrup mixing is batch (fixed-size loads on a cycle timer); filling, capping, and packaging are continuous flow (rate-based throughput). Schantt supports both production types in a single route, inserting wait-material pauses automatically when a flow stage outruns its batch supply.

• Multi-product routing with stage skipping — Pre-printed cans skip the labelling stage entirely. Schantt models this through per-class routing — each product class defines exactly which stages it uses — and applies a bridging transfer time across the skipped span so handoff delay is still accounted for.

• Sequence-dependent changeover times — The cleaning penalty between diet and sugar products (60 to 90 minutes) differs from same-sweetener flavour changes (15 to 30 minutes). Schantt captures each from-to pair as a directional duration on each machine and favours sequences that cluster similar products to reduce total changeover time.

• Shift-aware availability with seasonal calendar switching — Summit Springs expands from a single winter shift to extended summer shifts. Schantt assigns different calendars to different date ranges, respecting working hours per period, clamping starts into working windows, and splitting operations across shift boundaries automatically.

How Schantt handles each challenge

1. Diet-to-sugar changeover penalty on the PET line.
- Switching between diet and sugar products on the PET filler incurs an asymmetric cleaning penalty — 75 minutes from diet to herbal-sugar, 60 minutes from herbal-sugar to diet — because the intensive CIP protocol differs by direction. A sequence that alternates sweetener types multiple times per cycle can lose over two hours of productive filler time each week.
- Schantt captures each directional changeover as a from-to pair on the PET filler. When sequencing jobs in Auto mode, the algorithm favours runs that group diet products together and sugar products together, minimising the number of sweetener transitions. The planner can also lock in a fixed order using Semi-Auto mode and let the system optimise machine timing within that sequence. The changeover duration the planner enters includes the full cleaning cycle — pre-rinse, caustic wash, intermediate rinse, acid wash, final rinse — as a single figure.

2. Seasonal demand swings and capacity switching.
- Summer volume rises 60 percent above the winter baseline. The facility responds by moving from a single shift (Monday to Friday, 08:00 to 17:00, 40 hours per week) to an extended summer schedule (double shift Monday to Friday plus a Saturday early shift, totalling 96 hours per week). Manually adjusting the plan for each seasonal transition can take one to two days of re-optimisation.
- The planner sets up two named calendars — Standard (Winter) and Summer Peak — each with its own weekly shift pattern, then assigns each calendar to its applicable date range on the schedule. Schantt respects the active calendar when computing operation timing, so a schedule that spans the transition date automatically applies the correct working hours to each operation split across the boundary. The system does not decide when or by how much to expand capacity; the planner designates the switch date and the shift hours for each period.

3. Shared palletiser contention between can and PET lines.
- Both the can line and the PET line feed a single Layer Palletiser. The palletiser processes one line at a time at 14,400 units per hour for cans and 7,200 units per hour for PET bottles. When both lines finish packing at roughly the same time, the palletiser creates a backlog that delays the next round of palletising for the second line by 20 to 30 minutes.
- Schantt models the palletiser as a machine shared by both product classes. The algorithm staggers the completion times of upstream stages — multi-packing on each line — so the palletiser receives products from one line, then the other, rather than simultaneously. The planner configures the palletiser's throughput per product class, and the optimiser uses those rates to shift timing where it shortens total production time. CO₂ supply is also a shared resource across lines, but this guide assumes adequate capacity — Schantt does not model CO₂ inventory or delivery constraints.

4. Maintaining flow continuity between batch syrup mixing and inline filling.
- Four syrup mix tanks supply two filling lines. Each tank runs a 1,500 kg batch with a 45-minute cycle. The PET line feeding 500 mL bottles consumes syrup at roughly 900 kg per hour — faster than one mixer can supply if the line runs continuously. A flow stage that outruns its batch supply pauses and waits for material, breaking the downstream flow.
- Schantt simulates the interaction between batch and flow stages automatically. When the syrup batching schedule cannot keep pace with filler demand, the simulation inserts wait-material pauses on the affected flow operations — visible on the Gantt as idle intervals downstream. The planner can increase the number of tanks assigned to a product class, adjust batch sizes where feasible, or start syrup mixing earlier in the schedule to build a buffer before the filler starts. All four syrup mixers are identical, so the algorithm can assign any tank to any product class.

5. Stage-skipping routing for pre-printed cans.
- Classic Cola Can uses pre-printed aluminium cans and therefore skips Labelling entirely. The can route moves directly from Capping/Seaming to Multi-pack. Without an explicit bridging handoff, the schedule might incorrectly assume zero delay between those two stages or fail to chain them correctly.
- Schantt models this skip through per-class routing: the Cans-Cola-Sugar product class does not include Labelling in its stage sequence. A bridging transfer time from Capping/Seaming to Multi-pack (5 minutes) accounts for the crossover conveyor delay across the skipped span. The two PET classes include Labelling as an intermediate stage with a 3-minute transfer from Capping/Seaming. All routes converge at Multi-pack and continue to the shared Palletising stage. The bridging entry is configured on the Capping/Seaming stage's detail page alongside the other transfer times.

What to model in Schantt

The entity-count table below lists the five top-level objects a planner creates when setting up this scenario.

At plants where labelling throughput never constrains filler output, the labelling step can be bundled into the filling and capping stage as a combined downstream flow step. This guide models labelling separately because the PET route uses a dedicated machine whose availability can be scheduled independently.

Step-by-step setup

1. Create the stages in order. Define seven stages matching the plant layout: Syrup Mixing (batch, position 10), Carbonation (flow, position 20), Filling (flow, 30), Capping/Seaming (flow, 40), Labelling (flow, 50), Multi-pack (flow, 60), and Palletising (flow, 70). On each stage's detail page, enter the transfer times to its downstream successor. For the can class's skip-Labelling route, add a bridging transfer time from Capping/Seaming (stage 40) to Multi-pack (stage 60) with a duration of 5 minutes — this ensures the handoff delay is accounted for even though the intermediate labelling step is absent from the can routing.

2. Add machines to each stage. Assign the fourteen machines to their respective stages:
- Syrup Mixing — 4 identical Syrup Mix Tanks
- Carbonation — Inline Carbonator (Can) and Inline Carbonator (PET)
- Filling — Rotary Can Filler and Rotary PET Filler
- Capping/Seaming — Rotary Can Seamer and Rotary PET Capper
- Labelling — Shrink-sleeve Labeller (PET route only)
- Multi-pack — Drop Case Packer (Can) and Side-load Case Packer (PET)
- Palletising — Layer Palletiser (shared by both lines)

3. Create product classes and define per-class routing. Create three product classes — Cans-Cola-Sugar, PET-Colas-Diet, and PET-Herbal-Sugar — each with an appropriate unit (can or bottle). On each class's detail page, define its stage sequence. Cans-Cola-Sugar routes through all stages except Labelling (6 stages). Both PET classes route through all seven stages. No partial-transfer legs are needed for this scenario — each stage completes fully before the next stage starts.

4. Add one product per class. Create three products — Classic Cola Can (under Cans-Cola-Sugar), Diet Cola PET (under PET-Colas-Diet), and Herbal Root PET (under PET-Herbal-Sugar). Assign a display colour to each for Gantt visualisation.

5. Set machine capacity parameters and changeovers. On each machine's detail page, enter the processing parameters per product class. For the batch Syrup Mixing stage, enter a batch size of 1,500 kg and a cycle duration of 45 minutes per batch — these values apply to all four tanks and all three product classes. For each flow stage, enter the throughput in units per hour:

• Can line: Inline Carbonator (Can) 36,000; Rotary Can Filler 36,000; Rotary Can Seamer 36,000; Drop Case Packer (Can) 57,600
• PET line: Inline Carbonator (PET) 18,000; Rotary PET Filler 18,000; Rotary PET Capper 18,000; Shrink-sleeve Labeller 18,000; Side-load Case Packer (PET) 21,600
• Layer Palletiser (shared): 14,400 for can route; 7,200 for PET route

Then add the directional changeover durations for each from-to product-class pair on machines shared by multiple classes. The four syrup mix tanks need bidirectional changeovers for all three class pairs (25 minutes for same-sweetener flavour changes, 35 minutes for diet-to-anything transitions). The PET filler needs the asymmetric diet transitions (75 minutes for diet to herbal-sugar, 60 minutes for herbal-sugar to diet). The PET carbonator, PET capper, labeller, and PET case packer each need bidirectional entries for the two PET classes (5 to 10 minutes each). The shared palletiser needs six entries covering all three class pairs (5 minutes each direction).

6. Configure calendars, exceptions, and downtimes. Create the Standard (Winter) calendar: single shift, Monday to Friday, 08:00 to 17:00. Create the Summer Peak calendar: double shift Monday to Friday (06:00 to 22:00) plus a single Saturday early shift (06:00 to 14:00). Mark the Standard calendar as the default. Add two team-wide calendar exceptions for non-working public holidays — New Year's Day (January 1) and International Workers' Day (May 1). Add three machine downtime entries — a year-end factory-wide shutdown (December 24 to 31), and staggered annual deep-CIP days for the can filler (June 15) and PET filler (June 16) so at least one line remains available.

For step-by-step instructions on configuring each of these in Schantt, see the Schantt documentation.

Common mistakes

1. Using one blanket changeover duration across all product pairs. A single changeover time ignores the material difference between a same-sweetener flavour rinse (25 minutes on syrup mixers) and a diet-to-sugar intensive CIP (75 minutes on the PET filler). Fix: Enter each directional from-to pair separately on each machine's detail page so the algorithm can distinguish low-penalty from high-penalty transitions and sequence jobs accordingly.

2. Omitting the bridging transfer time for the can skip-Labelling route. Without a transfer time from Capping/Seaming to Multi-pack, the schedule treats those two stages as disconnected — no handoff delay is applied, and the alignment between them is lost. Fix: On the Capping/Seaming stage detail page, add a bridging transfer time to Multi-pack (5 minutes) for the can-class routing.

3. Modelling the shared palletiser with a single throughput value. The palletiser processes can cases and PET cases at different rates (14,400 versus 7,200 units per hour). A single blanket throughput assigns the same timing to both, masking the contention that actually occurs. Fix: Enter the palletiser's throughput per product class so the schedule reflects the real rate difference.

4. Creating a separate product class for every individual SKU. If each brand or recipe variation is its own product class, the planner must duplicate routing, changeover entries, and machine parameters for each one. The facility runs three distinct routings — any additional products that share one of these three patterns can be added as products within the same class. Fix: Define product classes by routing and changeover profile, not by individual SKU numbers. Add variants as products under the applicable class.

5. Forgetting to assign the correct calendar to each date range on the schedule. If the summer season starts and the schedule is still using the winter calendar, operations are only planned for 8 hours per day instead of the available 16. Fix: Use schedule calendar periods to assign the Summer Peak calendar to the April-through-September range and the Standard calendar to the rest of the year.

What a good schedule looks like

Before seasonal transition planning and manual line assignment, the schedule typically shows three unresolved friction points: diet and sugar products alternate randomly on the PET line, incurring two or more sweetener transitions per cycle and losing over two hours of filler time to changeovers each week; the palletiser backlog builds up when both lines finish packing simultaneously, adding 20 to 30 minutes of idle waiting per occurrence; and the shift from winter to summer capacity requires one to two days of manual re-optimisation before the schedule stabilises.

Before (spreadsheet-based scheduling):
- Diet-to-sugar transitions scattered through the week, consuming 120-plus minutes of unproductive changeover time on the PET filler
- Palletiser contention visible as repeated 20-to-30-minute backlogs when completion times collide
- Seasonal calendar switch triggers a multi-day manual rescheduling effort before the new pattern settles
- Matching syrup batch completion to filler start times relies on planner experience, with gaps or waits showing up during execution

After (Schantt Auto mode): The algorithm groups products by sweetener type to reduce changeover transitions to a single diet-to-sugar handoff per cycle, reclaiming over an hour of filler time that was previously lost to scattered cleaning windows. It staggers completion times across the can and PET lines so the shared palletiser receives steady work without building the 20-to-30-minute backlogs that occur when both lines finish simultaneously. The seasonal calendar switch is configured ahead of time — working hours extend automatically on the transition date, and operations that span the boundary are split correctly across shift windows. Syrup batching and filler timing are computed together, so wait-material pauses are minimised and the downstream rate is sustained through the production run.