Production Scheduling for Soft-gel Capsule Supplements

Production planners and operations managers in soft-gel capsule supplements manufacturing can model their entire production pipeline in Schantt — from gelatin preparation through encapsulation, drying, inspection, and packaging — and let the scheduling algorithm optimise job sequences across parallel encapsulators, changeovers, and packaging lines. This guide shows how to configure each stage, machine, product class, and calendar to produce a workable schedule that respects batch-and-flow timing, drying dwell, and shift windows.

This guide follows a fictional composite company built from industry research on soft-gel capsule supplements; all names, parameters, and figures are illustrative.

Industry context

Soft-gel capsule manufacturing is a hybrid flowshop where liquid or semi-solid formulations are sealed inside a one-piece gelatin shell. The process begins with gelatin preparation — batch melting and blending of gelatin powder, plasticiser, and water in pressurised melt tanks at controlled temperatures — then moves to encapsulation on rotary die machines that form, fill, and seal each capsule from a continuous ribbon of gelatin. After encapsulation, capsules must rest in a climate-controlled drying room for 24 to 36 hours to harden the shells to the target gelatine strength before automated visual inspection and packaging into bottles or blister packs. The drying delay is the longest single interval in the production timeline — longer than any processing step — and it decouples encapsulation from the downstream inspection and packaging stages, making forward planning difficult when tracked by hand.

Batch operations accumulate wait time quickly when downstream stages are occupied. A 120-minute encapsulation cycle on any of four parallel machines means changeover time between product classes — ranging from 45 minutes for a simple colour swap to 120 minutes for a full allergen cleanout — must be factored into every job assignment. On the flow side, two automated inspection lines each process 60,000 capsules per hour, and packaging throughput spans two orders of magnitude: 36,000 capsules per hour on the blister line and 180,000 on the bottle line. The inspection and packaging stages run only during the 16-hour working day, but the drying dwell elapses continuously across nights and weekends — a capsule finished at 18:00 Friday emerges from drying at 06:00 Sunday, yet inspection cannot begin until Monday morning. A single sequencing decision on the encapsulation floor can therefore ripple across multiple calendar days of downstream operations.

A mid-market facility typically runs 30 or more SKUs across three fill categories — oil-based, suspension, and specialty formulations — each with a different processing route through the four stages. Some classes skip gelatin preparation and draw pre-prepared mass from a holding tank; others bypass packaging entirely and ship bulk capsules to a third-party partner. The schedule must reconcile these divergent paths alongside the 2,160-minute drying dwell that bridges encapsulation and inspection, all within a 16-hour working day.

AtlasGel Nutraceuticals runs approximately 95 people at a 4,500 m² facility, making 3 product classes across 4 production stages, scheduled by a 3-person planning team.

Process overview

flowchart LR
    GP["Gelatin Preparation<br/>(Batch, 3 tanks)"] -->|"OIL / SPEC<br/>15 min"| EN["Encapsulation<br/>(Batch, 4 machines)"]
    EN -->|"drying dwell<br/>36 hr"| IN["Inspection<br/>(Flow, 2 lines)"]
    IN -->|"OIL / SUSP<br/>10 min"| PK["Packaging<br/>(Flow, 2 lines)"]

Product flow through AtlasGel's four production stages. OIL (Fish Oil 1000 mg) follows the full route; SUSP (Vitamin D3 2000 IU) enters at Encapsulation, skipping Gelatin Preparation; SPEC (CoQ10 200 mg) ends after Inspection, skipping Packaging.

Skip routing: SUSP product class uses pre-prepared gelatin mass from the holding tank — no dedicated gelatin prep step. SPEC product class ships inspected capsules in bulk drums to a third-party packager, bypassing AtlasGel's packaging lines.

Scheduling challenges and how Schantt handles them

In this scenario the planner enters a fixed list of jobs with quantities and product mix by class — a typical demand profile for make-to-stock and contract manufacturing. Readers whose schedule is driven by customer order dates or pull signals can adapt this input approach. Schantt optimises for minimum total production time, scheduling forward from a chosen start date over a practical horizon of several weeks. Two scheduling modes are available: Auto mode explores both job sequence and machine assignment simultaneously; Semi-Auto mode keeps the planner's fixed job order but optimises which machines each job runs on within that sequence.

What Schantt handles well

• Sequential multi-stage production — Soft-gel manufacturing follows a fixed stage order (gelatin preparation, encapsulation, drying, inspection, packaging). Schantt models this as an ordered stage sequence with forward-only material flow and transfer times between stages, so every operation's start and end respects the production order.

• Multi-machine stages — Encapsulation runs on up to four rotary die machines in parallel; melt tanks run in parallel in gelatin preparation. Schantt assigns jobs across a stage's available machines, choosing the combination that minimises total production time.

• Mixed batch-and-flow pipelines — Gelatin preparation and encapsulation are batch operations; inspection and packaging are continuous-flow operations. Schantt models both production types in the same route, with batch-cycle duration for melt tanks and throughput-based timing for inspection and packaging lines.

• Sequence-dependent changeovers — Encapsulation machines require different setup durations depending on the product class change (fill formulation, gelatin colour, capsule size). Schantt models directional changeover times per machine per product-class pair, so the schedule accounts for the real time penalty of switching between products.

• Shift-aware availability — Production runs on defined shift patterns while drying (modeled as transfer time) elapses continuously. Schantt models working calendars per team so each stage's operations advance only during active working windows.

• Calendar exceptions and downtimes — Planned holidays, maintenance windows, and one-off schedule changes are modeled as calendar exceptions and machine downtimes that affect timing and render as shaded bands on the Gantt.

How Schantt handles each challenge

1. Parallel encapsulators with asymmetric changeovers.
- Four encapsulation machines share six directional changeover pairs each, and the duration varies by direction — 45 minutes for a suspension-to-specialty switch, 120 minutes for a specialty-to-oil allergen cleanout. A planner sequencing 15 to 20 weekly jobs manually must track both the machine each job occupies and the accumulated changeover time at every handoff, which leads to suboptimal grouping and 4 to 6 hours of avoidable changeover time per week.
- On each encapsulator, Schantt models the full per-direction changeover matrix — six pairs per machine, with durations from 45 to 120 minutes — and in Auto mode explores job sequences that group similar changeovers together. The schedule shows the exact changeover duration on each machine between every job pair, so the planner sees both the sequence the algorithm chose and the time each transition consumes.

2. Mixed batch-and-flow pipeline with a 36-hour drying dwell.
- Two batch stages (gelatin preparation at 85 minutes per 750 kg batch; encapsulation at 120 minutes per 350,000-capsule batch) feed two flow stages (inspection at 60,000 capsules per hour per line; packaging at 36,000 to 180,000 capsules per hour). Between encapsulation and inspection sits a 2,160-minute drying dwell that is longer than any operation in the route. A downstream inspector or packaging line that completes its current work before the next batch arrives from drying sits idle, wasting throughput.
- Schantt types each stage as batch or flow, so the scheduler applies the correct timing model per stage: batch stages repeat at cycle duration divided across batch multiples; flow stages run at a continuous throughput rate. The drying dwell is a fixed transfer time on the encapsulation-to-inspection edge — not a fifth stage — so the schedule advances from encapsulation completion, pauses for the 36-hour transfer, and then starts inspection. The Gantt shows the dwell as a delay between the two operations, and downstream stages start only after material arrives.

3. Divergent per-class routings requiring distinct stage paths.
- Oil-based products pass through all four stages. Suspension products skip gelatin preparation entirely. Specialty products end at inspection and skip packaging. When all three classes are scheduled together, the planner must manually ensure each job's route is followed and that encapsulation does not wait for a gelatin-prep batch that will never arrive.
- Each product class defines its own routing as a sequence of stages. The SUSP class's routing begins at encapsulation with no gelatin-prep leg — Schantt never inserts a gelatin-preparation step because the routing does not include it. Similarly, the SPEC class ends at inspection; packaging is absent from its routing. The scheduler chains only the stages each class needs, and on the Gantt the planner sees each job pass through exactly its class's stages in order.

4. Shift-limited working windows across a multi-day drying gap.
- The facility runs 16 hours per day, Monday through Friday. A batch of oil-based capsules that finishes encapsulation at 18:00 on Friday enters the 36-hour drying dwell, which elapses continuously over the weekend and completes at 06:00 on Sunday — but the inspection team does not start until Monday at 06:00. On the planner's manual spreadsheet this gap is easy to misalign, causing inspection lines to wait for work that arrived two days earlier.
- Schantt applies the production calendar (Monday–Friday, 06:00–22:00) to every stage operation, while transfer times, including drying dwell, elapse on continuous clock time. The schedule advances encapsulation to completion at 18:00 Friday, applies the 2,160-minute drying transfer, and then waits until Monday 06:00 before starting inspection — the Gantt shows the weekend gap as calendar non-working time following the drying transfer. The planner sees exactly when downstream stages resume, without manual offset calculation.

What to model in Schantt

The table below lists the top-level entities a planner creates to represent this production environment. Sub-configuration such as changeovers, transfer times, calendar exceptions, and downtimes are set on the detail pages of these entities.

Step-by-step setup

1. Create the stages in order. Add four stages — Gelatin Preparation (batch), Encapsulation (batch), Inspection (flow), and Packaging (flow) — in the sequence they appear on the production floor. On each stage's detail page, set the transfer time to the next stage:
* Gelatin Preparation → Encapsulation: 15 minutes
* Encapsulation → Inspection: 2,160 minutes (the 36-hour drying dwell)
* Inspection → Packaging: 10 minutes

2. Add machines to each stage. Place the three melt tanks in Gelatin Preparation, the four encapsulators in Encapsulation, the two inspection lines in Inspection, the bottle line and blister line in Packaging.

3. Create the product classes and define each class's routing. Add three classes — Oil-based, Suspension, Specialty. On each class's detail page, set the per-class routing:
* Oil-based: all four stages in order (gelatin prep → encapsulation → inspection → packaging)
* Suspension: encapsulation → inspection → packaging (skip gelatin prep)
* Specialty: gelatin prep → encapsulation → inspection (skip packaging)

No partial-transfer legs are needed — all batches move as complete lots between stages.

4. Add one product per class. Create Fish Oil 1000 mg (Oil-based), Vitamin D3 2000 IU (Suspension), and CoQ10 200 mg (Specialty), each assigned to its respective class.

5. Set machine capacity parameters and changeovers. On each machine's detail page, enter the capacity parameters and changeover times. These values depend on the product classes, so this step follows step 3.
* Melt tanks (all three): 85-minute batch cycle, 750 kg batch size, for Oil-based and Specialty classes
* Encapsulators (all four): 120-minute batch cycle, 350,000-capsule batch size, for all three classes
* Changeover matrix per encapsulator (6 directional pairs): OIL→SUSP (60 min), SUSP→OIL (75 min), OIL→SPEC (90 min), SPEC→OIL (120 min), SUSP→SPEC (45 min), SPEC→SUSP (45 min)
* Inspection lines (both): 60,000 capsules per hour throughput, for all three classes
* Bottle line: 180,000 capsules per hour throughput, for Oil-based class
* Blister line: 36,000 capsules per hour throughput, for Suspension class

6. Configure the calendar, exceptions, and downtimes. Set the production calendar to Monday–Friday, 06:00–22:00 (two shifts, 16 hours per day), and apply it as the default calendar. This calendar governs every stage operation — gelatin preparation, encapsulation, inspection, and packaging — while transfer times, including the 36-hour drying dwell, run on continuous clock time independent of the calendar.

Then add the calendar exceptions and machine downtimes that reflect the facility's planned non-production periods:
* Calendar exceptions (three): New Year's Day (January 1, non-working), International Workers' Day (May 1, non-working), year-end shutdown (December 21 onward, non-working)
* Machine downtimes (two): Encapsulator 1 annual overhaul (August 15–17, full 16-hour shifts each day — die roll bearings, pump seals, calibration) and a factory-wide shutdown (December 21–January 1, aligned with the year-end calendar exception)

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 for all encapsulation switches. Applying one changeover duration (for example, 60 minutes) to every product-class transition ignores the directional asymmetry — a specialty-to-oil changeover takes 120 minutes, while a suspension-to-specialty switch takes only 45. The schedule understates the time needed for allergen-related changeovers and overstates it for simpler ones.
Fix: Enter each direction as a separate changeover pair per machine — six pairs per encapsulator — with the real duration for that product-class transition. The dataset uses values from 45 to 120 minutes.

2. Grouping divergent routings into a single product class. Creating one product class for all soft-gel products forces every job through the same stage sequence, so suspension products appear to need gelatin preparation and specialty products appear to need packaging.
Fix: Split by routing — one class per unique stage path. Oil-based, Suspension, and Specialty each get their own class, and each class's routing includes only the stages its products actually pass through.

3. Modeling drying as a fifth stage with a machine assignment. Adding a "Drying" stage with fictional machine resources suggests the scheduler can assign drying capacity and that jobs wait for drying machines, when in reality drying is a fixed time delay that all capsules pass through.
Fix: Leave drying out of the stage list. Set the encapsulation-to-inspection transfer time to 2,160 minutes on the Encapsulation stage's detail page. The schedule applies the delay without creating a stage that would need calendars, machines, or capacity settings.

4. Overlooking the blister line's lower throughput when sequencing packaging. The bottle line runs at 180,000 capsules per hour, while the blister line runs at 36,000 — a five-to-one ratio. A schedule that alternates packaging tasks without considering this difference can bottleneck the blister line while the bottle line sits idle.
Fix: Model both packaging lines with their correct throughput values. Let the scheduler assign packaging-stage jobs to the machine whose speed matches the batch size and available time window. The algorithm's machine exploration will balance the load.

5. Forgetting a transfer time for a product class that skips a stage. When SUSP skips gelatin preparation and enters at encapsulation, the transfer-time chain from encapsulation onward still applies — but a planner who configures only the gelatin-prep-to-encapsulation transfer may leave the encapsulation-to-inspection drying dwell unset for SUSP. The schedule then shows inspection starting immediately after encapsulation for suspension products, missing the critical 36-hour dwell.
Fix: Transfer times are defined on the source stage's detail page, not per class. Set the encapsulation-to-inspection transfer to 2,160 minutes on the Encapsulation stage once, and every class that passes through encapsulation — OIL, SUSP, and SPEC — inherits the same drying dwell automatically.

What a good schedule looks like

Before Schantt, the planning team spends roughly two hours every Monday morning re-ordering the weekly job list to accommodate the previous week's carry-over work. The manual spreadsheet groups encapsulation jobs by rough similarity but misses changeover-minimising sequences, leaving 4 to 6 hours of avoidable changeover time spread across the four encapsulators each week.

Before (manual scheduling):
* Planner manually assigns each of ~15–20 weekly jobs to an encapsulator, estimating changeover duration from a mental lookup table
* 4–6 hours per week of avoidable changeover time across the four encapsulators due to suboptimal job grouping
* 2–3 gel mass degradation events per quarter when a gelatin-prep batch sits past its 4- to 8-hour hold window because the assigned encapsulator is still finishing a long changeover
* Inspection and packaging lines experience intermittent idle time because the drying dwell is tracked on a separate whiteboard, not linked to the encapsulation schedule

After (Schantt Auto mode):
* The algorithm sequences and assigns all jobs across the four encapsulators in a single run, grouping similar changeovers and reducing total changeover time compared to the manual grouping — the 4 to 6 hours of weekly changeover waste is cut significantly by sequencing transitions end to end
* Each encapsulator's Gantt shows the exact job sequence with the directional changeover duration between every pair — no mental lookup required, and the planner can drag a job to a different position and see the new changeover time update immediately
* Gelatin preparation batches are timed so they complete within the 4- to 8-hour window before their downstream encapsulation job begins; the planner confirms the gap between the two operations on the Gantt, and the schedule highlights any timing misalignment through the visible gap between the gelatin preparation and encapsulation bars
* The drying dwell appears as a scheduled transfer between encapsulation completion and inspection start; inspection and packaging tasks follow without manual offset calculation, and the planner sees exactly when each job is available downstream
* Total production time across the job list is minimised, and the planning team's weekly re-ordering effort drops from approximately two hours of manual spreadsheet work to a focused schedule review