Production Scheduling for Kombucha & Live-Culture Beverages

This guide is written for production planners, operations managers, and plant managers scheduling kombucha and live-culture beverages. It shows how to model parallel fermentation vessels, directional flavour changeovers, seasonal capacity shifts, and divergent product routings in Schantt so your schedule reflects how a real kombucha brewery runs.

This guide follows a fictional composite company built from industry research on kombucha & live-culture beverages; all names, parameters, and figures are illustrative.

Industry context

Kombucha and live-culture beverages go through a multi-stage production pipeline that starts with brewing sweetened tea, then moves into a long fermentation period where a SCOBY (symbiotic culture of bacteria and yeast) converts the sugar into a tangy, lightly effervescent drink. The base kombucha can be blended with fruit flavours, carbonated, and packaged as a non-alcoholic beverage, or fermented longer and then pasteurised to produce hard kombucha with a higher alcohol content. Production runs across both batch stages (tea brewing, fermentation, blending, carbonation) and flow stages (pasteurisation, packaging) in a single sequential route, with at least three distinct product classes that each follow different paths through the pipeline. What makes kombucha scheduling distinct from conventional beverage bottling is the biological cycle time of First Fermentation — measured in days rather than minutes — which forces planners to think in terms of staggered vessel starts and maturation cadence rather than hourly line throughput. The downstream packaging line operates at a much faster pace, so the handoff between the slow fermentation stage and the rest of the plant is the critical coordination point in every schedule.

The fictional Wild Cultures facility occupies 2,500 square metres with 80 employees. The plant operates 6 fermentation vessels of 2,000 litres each with per-class cycles of 10 days for Classic, 12 days for Fruit, and 14 days for Hard Kombucha — the longest of any stage, making First Fermentation the binding capacity constraint. Two brew kettles (1,000 litres, 60-minute cycle) feed the vessels, followed by two blend tanks (1,000 litres, 30-minute cycle) and two brite tanks for carbonation (2,000 litres, 60-minute cycle). A plate pasteuriser at 6,000 litres per hour handles hard kombucha only, and a canning line at 4,000 cans per hour packages all three classes. First Fermentation runs on a continuous 24/7 calendar because biological cultures cannot be paused, while downstream stages follow the plant shift calendar — single shift, Monday to Friday, 08:00 to 17:00 in winter, expanding to two shifts, Monday to Saturday, 06:00 to 22:00 during the summer peak months of May through August. Scheduling is handled by a 5-person planning team, covering 3 product classes across 6 production stages.

Process overview

flowchart LR
    TB["Tea Brewing<br/>Batch · 2 kettles"]
    F1["First Fermentation<br/>Batch · 6 vessels<br/>10–14 day cycle"]
    FB["Flavor Blending<br/>Batch · 2 tanks"]
    CB["Carbonation<br/>Batch · 2 brite tanks"]
    PT["Pasteurisation<br/>Flow · Hard class only<br/>6,000 L/hr"]
    PK["Packaging<br/>Flow · Canning line<br/>4,000 cans/hr"]

    TB --> F1
    F1 --> FB
    FB --> CB
    CB --> PT
    PT --> PK
    CB --> PK

Six-stage production pipeline from tea brewing through packaging. First Fermentation runs on a continuous 24/7 calendar; downstream stages follow the plant shift calendar. Pasteurisation applies only to hard kombucha.

Fruit-flavoured kombucha skips the Flavor Blending stage — fruit concentrate is dosed directly into the First Fermentation vessel. Hard kombucha adds Pasteurisation between Carbonation and Packaging. Both variations use per-class routing to include or exclude stages.

Scheduling challenges and how Schantt handles them

This guide assumes demand is driven by seasonal consumption patterns — the summer peak from May through August, a winter trough, and transitional months between them — with a 12-week rolling planning horizon. If your brewery runs to firm customer orders or co-pack commitments rather than seasonal shifts, you can still apply the same configurations; the schedule calendar periods let you align capacity with whatever date ranges your demand profile follows.

Schantt's scheduling algorithm minimises total production time — the overall completion of all jobs in the schedule — and schedules forward from a user-chosen start date. The practical horizon for this scenario is 12 to 16 weeks, matching a typical kombucha planning cycle from vessel inoculate to packaged pallet. Two optimisation modes are available: Auto mode, where the algorithm decides job sequence, machine assignments, and timing from a list of products and quantities; and Semi-Auto mode, where the planner fixes the production order and the algorithm optimises machine assignments within that sequence, respecting any per-job earliest-start times at the first required stage.

What Schantt handles well

• Sequential multi-stage production with mixed batch and flow stages — kombucha's pipeline mixes batch stages that measure cycles in minutes or days (tea brewing at 60 minutes, First Fermentation at 10 to 14 days) with flow stages that run at continuous rates (6,000 litres per hour or 4,000 cans per hour). Schantt detects material starvation where a downstream stage outruns its upstream supply and renders those intervals as wait gaps on the Gantt, so the planner sees exactly where the pipeline constricts.

• Multi-machine stages with parallel vessels — the First Fermentation vessel bank of 6 vessels is a single stage with parallel machines. Auto and Semi-Auto modes explore assignment across vessels to minimise total production time, and the planner can examine the resulting per-vessel load on the Gantt to confirm no single vessel is over- or under-utilised.

• Sequence-dependent changeovers by flavour clan — a directional changeover-time matrix on each vessel per flavour-class pair (illustrative range: 15 to 90 minutes — a starting point that each plant calibrates from its own cleaning experience). The scheduling algorithm favours sequences that cluster similar classes back-to-back, reducing the total time spent on cleaning across the vessel bank without manual guesswork.

• Multi-product routing with stage skipping — Fruit class skips Flavor Blending; Hard class adds Pasteurisation. Per-class routing omits unneeded stages and bridges transfer time across skips so the schedule does not insert idle operations for skipped steps.

• Seasonal calendar shifts and exceptions — summer peak (2 shifts, 6 days) versus winter off-peak (1 shift, 5 days) via schedule calendar periods, plus calendar exceptions for public holidays and downtime entries for the year-end shutdown and annual deep cleaning.

How Schantt handles each challenge

1. Staggering fermentation vessel starts for steady downstream flow.

• When all First Fermentation vessels start on the same day, six batches mature simultaneously, flooding downstream stages roughly two weeks later and then starving them for days until the next maturation wave. Planners need one to two vessels maturing per day for a consistent downstream feed.

• Schantt's scheduling algorithm sequences vessel starts across the available calendar by distributing job starts over the planning horizon. The Gantt reveals the resulting maturation cadence — showing each vessel's cycle, its completion date, and when its batch reaches each downstream stage — allowing the planner to adjust the production order or quantity mix until the downstream load levels out.

2. Managing directional flavour changeovers by class pair.

• Switching a vessel from Classic to Fruit takes 30 minutes of cleaning, but switching from Fruit to Classic takes 60 minutes and from Hard to Classic takes 90 minutes. These asymmetric times add up across a multi-vessel bank. With six vessels and three classes cycling through them, the cumulative changeover time can consume the equivalent of a full shift per week across the bank if sequences are chosen without regard for transition direction.

• Schantt models these directional changeover times on each rotating vessel as a per-pair matrix — the planner enters each from-class-to-class duration once on the machine detail page. In Auto mode, the algorithm favours job sequences that cluster similar classes back-to-back, reducing the total changeover time across all vessels. In Semi-Auto mode, the planner fixes the production order and the algorithm assigns vessels to minimise the time spent on changeovers within that sequence. The resulting changeover intervals appear as labelled segments on each operation's Gantt bar.

3. Maintaining the First-Fermentation-to-packaging handoff window.

• Fermented kombucha has a limited window — ideally 48 to 72 hours — between maturation and packaging. A cold crash at 1 to 4 degrees Celsius can extend this hold by 24 to 48 hours, but exceeding the window risks over-carbonation and alcohol creep.

• Schantt chains transfer times from First Fermentation through each downstream stage to Packaging. The Gantt visualises the elapsed time between First Fermentation completion and each downstream operation as a continuous timeline, enabling the planner to verify that every job stays within the handoff window before finalising the schedule. Because cold crash is entered as a fixed transfer time rather than a separate stage, the planner sees it as a single delay interval rather than a binding step.

4. Adjusting capacity for seasonal demand swings.

• Summer demand can reach roughly twice the winter trough, requiring extended shifts (2 shifts, Monday to Saturday, 06:00 to 22:00) whereas winter runs a single shift (Monday to Friday, 08:00 to 17:00). Manually adjusting the weekly shift pattern for each schedule against the plant calendar is error-prone — planners routinely forget to flip the May switch or double-book a Saturday shift that no longer exists in October.

• The planner assigns schedule calendar periods to the relevant date ranges — the summer calendar from May through August, the winter calendar from October through March, with transitional months on whichever period applies. First Fermentation vessels maintain a separate 24/7 continuous calendar regardless of seasonal shifts. The schedule then respects the correct working windows for each date range, and the Gantt overlay renders non-working time in the appropriate period — the planner sees the shift boundary change on the Gantt without manual recalibration.

5. Handling vessel downtime from culture failures and deep cleaning.

• A SCOBY failure can sideline a vessel for 2 to 3 weeks. Scheduled annual deep cleaning also takes a vessel offline for a full week. Without explicit downtime entries, the schedule may assign jobs to unavailable vessels, creating delays when the plan reaches the floor.

• The planner enters downtime windows on each affected machine — a factory-wide year-end shutdown and an annual deep CIP for one vessel. The schedule respects these unavailability windows and reassigns jobs to eligible vessels during those periods.

What to model in Schantt

The following five entities are the first-class objects a planner creates in Schantt to set up this scenario:

Step-by-step setup

1. Create the stages in production order and set transfer times. Create six stages — Tea Brewing (batch), First Fermentation (batch), Flavor Blending (batch), Carbonation (batch), Pasteurisation (flow), Packaging (flow) — each with its position number. On each stage's detail page, set the forward transfer times to the next stage:

Stage-to-stage transfer times:
- Tea Brewing to First Fermentation: 30 minutes
- First Fermentation to Flavor Blending: 45 minutes
- First Fermentation to Carbonation (Fruit skip bridge): 45 minutes
- Flavor Blending to Carbonation: 30 minutes
- Carbonation to Pasteurisation: 30 minutes
- Pasteurisation to Packaging: 15 minutes
- Carbonation to Packaging (cold crash): 2,880 minutes

2. Add the machines to each stage. Assign 2 brew kettles to Tea Brewing, 6 vessels to First Fermentation, 2 blend tanks to Flavor Blending, 2 brite tanks to Carbonation, 1 plate pasteuriser to Pasteurisation, and 1 canning line to Packaging. Configure vessel eligibility on each First Fermentation machine so the scheduling algorithm knows which classes each vessel handles:

• Vessels 01 and 02: Classic and Fruit only
• Vessel 04: Hard only
• Vessels 03, 05, and 06: all three classes

3. Create the product classes and define per-class routing. Create three product classes — Classic / Original, Fruit, Hard Kombucha. For each class, define its routing by selecting the required stages in order:

• Classic: Tea Brewing → First Fermentation → Flavor Blending → Carbonation → Packaging
• Fruit: Tea Brewing → First Fermentation → Carbonation → Packaging (skips Flavor Blending)
• Hard: Tea Brewing → First Fermentation → Flavor Blending → Carbonation → Pasteurisation → Packaging

Enable partial transfer on the First Fermentation stage for Classic and Fruit with a quantity of 1,500 litres. This lets the downstream stage start processing the first portion before the vessel's full batch completes, reducing the idle time between the long fermentation and subsequent steps.

4. Add one representative product per class. Create Original Kombucha (assigned to the Classic class), Berry Kombucha (Fruit), and Hard Original (Hard). Each inherits its class's routing, machine eligibility, and changeover configuration automatically.

5. Set machine capacity parameters and changeovers. On each machine's detail page, configure the batch cycle duration and batch size for batch-stage machines, and the throughput rate for flow-stage machines. Then enter the directional changeover-time matrix on each First Fermentation vessel that handles two or more classes:

Batch cycle duration and batch size:
- Brew kettles: 60 minutes, 1,000 litres (all classes)
- First Fermentation vessels (Classic): 14,400 minutes (10 days), 2,000 litres
- First Fermentation vessels (Fruit): 17,280 minutes (12 days), 2,000 litres
- First Fermentation vessels (Hard): 20,160 minutes (14 days), 2,000 litres
- Blend tanks: 30 minutes, 1,000 litres (all classes)
- Brite tanks: 60 minutes, 2,000 litres (all classes)

Throughput for flow stages:
- Plate pasteuriser: 6,000 litres per hour (Hard only)
- Canning line: 4,000 cans per hour (all classes)

Directional changeover times on rotating vessels — examples:
- Classic to Fruit: 30 minutes
- Fruit to Classic: 60 minutes
- Classic to Hard: 60 minutes
- Fruit to Hard: 75 minutes
- Hard to Classic: 90 minutes
- Hard to Fruit: 75 minutes

Configure the full symmetric matrix for each vessel's eligible class pairs. Vessels 03, 05, and 06 need all six pairs; Vessels 01 and 02 need the Classic–Fruit pairs only.

6. Configure calendars, exceptions, and downtimes (optional, last). Set Plant Shift as the default calendar (Monday to Friday, 08:00–17:00). Create a second calendar named Fermentation 24/7 (24/7, all days) and assign it to each of the six fermentation vessels on their machine detail page so their cycles run uninterrupted regardless of plant shifts. Add calendar exceptions for New Year's Day (1 January) and International Workers' Day (1 May). Enter a factory-wide year-end shutdown from 24 to 31 December and an annual deep CIP downtime for one First Fermentation vessel during the third week of January (19 to 25 January).

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

Common mistakes

1. Using a single blanket changeover time instead of per-pair directional values. The schedule then treats every flavour transition the same, ignoring the threefold difference between a Hard-to-Classic cleanout at 90 minutes and a Classic-to-Fruit cleanout at 30 minutes. The algorithm cannot favour low-changeover sequences, and the Gantt shows the same average time on every transition — removing the incentive to cluster similar classes. Fix: Enter each directional from-class-to-class pair on every rotating vessel that handles two or more product classes. Configure the full symmetric matrix so the algorithm has accurate transition times for every ordering it evaluates.

2. Creating one product class for all kombucha varieties. All products then share the same routing, so Fruit cannot skip Flavor Blending and Hard cannot add Pasteurisation. Every product runs every stage, inflating processing time on stages it does not need. Fix: Create a separate product class for each divergent routing — Classic, Fruit, and Hard — each with its own stage sequence.

3. Leaving all six First Fermentation vessels eligible for all product classes. The algorithm may then assign a Classic batch to a dedicated hard-only vessel, creating a scheduling conflict when a Hard job needs that vessel. The schedule must be manually repaired before it reaches the floor. Fix: Restrict each vessel's eligibility to the classes it physically handles. Leave only the intended class–machine rate entries configured for dedicated vessels.

4. Forgetting to assign the 24/7 calendar to the fermentation vessels. The vessels then inherit the default Plant Shift calendar, adding 15 hours of non-working time per weekday and full weekends to every 10-to-14-day cycle. The resulting completion dates drift well past the real maturation window. Fix: Create the Fermentation 24/7 calendar and assign it to each of the six vessels individually on their machine detail page.

5. Using a single uniform cold-crash duration for all product classes. The carbonation-to-packaging transfer time then misrepresents the total First-Fermentation-to-packaging elapsed time for classes whose actual hold differs from the configured value. The schedule's handoff window check becomes unreliable. Fix: If classes have different cold-crash requirements, configure the transfer time on each class's routing to match its actual hold duration.

What a good schedule looks like

A well-configured schedule for this scenario replaces guesswork about vessel availability and downstream timing with a reliable, visual plan.

Before (manual spreadsheet scheduling): The planning team builds the 12-week plan by hand, relying on tribal knowledge to stagger vessel starts and guess changeover sequences.

• Vessel starts cluster at the beginning of the month — the team loads the plan when orders come in — creating a downstream flood of 4 to 6 mature batches arriving at once two weeks later, followed by a 5-to-7-day starvation gap. Packaging stands idle for most of the second week of every cycle.
• Changeover sequences are chosen arbitrarily because the team cannot evaluate the 3-by-3 directional matrix across six vessels by eye. The result is 8 to 12 hours of avoidable cleaning time per vessel per month.
• Seasonal calendar changes are applied manually; the first winter schedule after summer routinely schedules work on a Saturday that no longer has a shift, and the transition to summer shifts is always a week late because someone forgets the start date.
• Past due dates are common on co-pack orders because the team can only approximate the elapsed time from vessel inoculate to packaged pallet. Without a clear view of the handoff window, planners pad every estimate by 2 to 3 days, which compresses the downstream schedule further.

After (Schantt Auto mode): The same product mix and quantities produce a schedule the algorithm builds from the modeled constraints in a few minutes — no manual trial-and-error required.

• Vessel starts are distributed across the planning horizon, yielding a steady maturation cadence of one to two vessels per day. Downstream stages run predictably with no starvation gaps — packaging sees a consistent daily flow instead of a feast-or-famine cycle.
• The algorithm clusters similar flavour classes back-to-back on shared vessels, reducing total monthly changeover time across the bank. The 40-to-50-minute average transition replaces the old pattern where every switch was treated as a worst-case cleanout.
• Schedule calendar periods switch summer and winter shift patterns automatically at the configured dates. The planner sets the period boundaries once per year; the May and October transitions happen without manual intervention, and the Gantt shows the correct working windows from day one of each period.
• Every job's elapsed time from First Fermentation through to packaging is continuously visible on the Gantt, so handoff windows are verified before the schedule is released. Past due dates drop to isolated exceptions rather than recurring events because the planner sees the real timeline through the entire pipeline at a glance.