Production line simulation for lean manufacturing teams
Subscribe to keep running simulations and building better production lines.
Loading…
Welcome back!
Press any key or click anywhere to dismiss
Thanks for subscribing to Flowcell. Your production line simulation is now fully unlocked.
Now go find that bottleneck, slash your takt time, and make your line sing. Your team is waiting.
Complete guide to simulating and optimising your production line
Flowcell is a discrete-event production line simulator for lean manufacturing teams. Build a virtual version of your factory floor, throw your real production queue at it, and watch what happens — in seconds rather than days.
Flowcell simulates the movement of products through a configurable sequence of workstations, buffers, and sub-assembly stations. It models operator productivity, JIT material feeding, trolley-based replenishment, and kanban stocking — giving you a live view of throughput, bottlenecks, lead times, and idle time.
Physical line trials are expensive. Moving machinery, redeploying operators, and disrupting live production to test a new layout can cost a day of output. Flowcell lets you run that same test in 10 seconds, try 20 variations, and walk onto the floor already knowing the answer.
Get your first simulation running in under 3 minutes by following these steps in order.
In the Process Times panel, you'll find a default list of operations. Edit the names and times to match your real process steps. Each process has a time in seconds per unit. If a process is stocked at point-of-use (screws, fixings), tick Kan to mark it as kanban — it won't need feeder trips.
In Product Types, create entries for each variant you manufacture. For each type, set how many of each process it needs. A "5 Drawer Cabinet" might need 10 slides and 5 drawers but no doors — enter 0 for anything not required.
Use the Line Layout panel to create a sequence of stations. Click + Work to add a workstation, + Buffer for a queue point between stations, and + Pack for your final QC/packing station. Assign which processes happen at each workstation using the checkboxes.
In Operatives, enter your operators' names and productivity percentages. 100% is standard pace. Assign each operative to a station in the Line Layout panel.
In the Production Queue at the bottom, select a product type from the Batch dropdown, set a quantity, and click Add Batch. This populates your shift schedule. You can also add individual rows manually with + Add Row.
Press ▶ Start Shift in the top bar. Watch your line run. Adjust the speed control to slow it down for analysis or speed it up to 500× warp to get results instantly. Press ↺ Reset to clear the floor and reconfigure.
Processes are the individual operations performed on a product. Each process has a name, a time in seconds per unit, and a kanban flag.
Click the + Add button to create a new process. Click the name field to rename it. Set the time in seconds — this is the standard time to complete that operation on one unit at 100% operator productivity.
Click the ✕ button to the right of any process to remove it. You must have at least one process defined. Removing a process also removes it from all product types and line stations.
Drag the ⠿ grip handle on the left to reorder processes. The order here determines the column order in the Production Queue table.
Tick Kan to mark a process as kanban-stocked. Kanban processes are assumed to be available on demand at the workstation — they don't require feeder trips to replenish. Use this for low-cost consumables like fixings, screws, or small brackets that are stocked in unlimited lineside bins.
Product Types define the process requirements for each variant you manufacture. They are the building blocks of your production queue.
Click + New Type to add a product type. Give it a name (e.g. "4 Shelf Cupboard"), then fill in the quantity of each process it needs. Leave a process at 0 if that product doesn't require it.
When a unit enters the line, the simulation reads its product type requirements to calculate the work content at each station. A station with two doors to fit will take twice as long as one with one door. Feeder quantities are also derived from these requirements.
All fields are editable inline. Click the red ✕ to delete a type — this won't affect units already in the queue that use it.
The Line Layout panel defines the physical sequence of stations on your production line. Products flow left-to-right through the sequence you build here.
Each workstation card shows:
The simulation flows products through stations in the order they appear in the layout. Place your buffers between workstations to prevent upstream stations from starving or being blocked.
The Pack station has two optional time fields: Manual time (seconds of operator work) and Auto time (machine cycle time that runs without the operator). If both are set, the operative can move away during the auto cycle.
Move times control how long it takes products and feeders to travel between locations on your line.
The time in seconds to physically move a unit from one station to the next — including lifting, pushing the trolley, or conveying. This applies every time a unit transfers to a new station.
The time in seconds for an operative to walk one station along the line. This is used to calculate travel between workstations. A product moving from WS1 to WS3 passes two intermediate stations — that's 2 hops. Buffer stations are free pass-throughs and don't add hops.
These settings control the JIT replenishment system:
Operatives are the workers on your line. Each has a name and a productivity rating that scales their effective work speed.
Click + Add Operative at the bottom of the Move Times panel to add a new team member. By default the sim includes Pete, Nathan, and Matt — rename or delete these to match your real team.
Set productivity as a percentage of standard pace. 100% means the operative works at the standard rate used to define process times. 80% means they take 25% longer for each operation. 110% means they're 10% faster than standard.
In the Line Layout panel, each workstation has an operative selector. Assign your team members to stations here. Multiple stations can't share the same operative — each station needs its own person.
The production queue is your shift's work order — the ordered list of units to run through the line. The simulation processes them in sequence, first in first out.
Select a product type from the Batch dropdown, set a quantity, and click Add Batch. This adds that many rows of the selected type to the bottom of the queue. Use this to quickly build realistic production mixes.
Click + Add Row to add a single unit. You can then change its name and manually set quantities for each process — useful for one-off specials or non-standard configurations.
Each row shows the unit name and process quantities. Click any quantity field to edit it. Use the up/down arrow buttons to reorder rows. The 📋 button duplicates a row. The ✕ button removes it.
Click 🔀 Shuffle to randomise the queue order. This is useful for testing high product-mix scenarios where units arrive in unpredictable order — it stress-tests your feeder system and buffer sizing.
Click Clear all to remove all queue items and start fresh.
Once your line is configured and queue is loaded, you're ready to run.
Press ▶ Start Shift to begin. The simulation clock starts counting up in the top bar. Press ⏸ Pause to freeze the simulation mid-run for analysis. Press ▶ Resume to continue. Press ↺ Reset to clear the floor and return to ready state.
The speed dropdown lets you choose how fast simulated time passes:
When all units in the queue have passed through the Pack station, the shift complete report appears automatically, showing KPIs and operator utilisation.
Flowcell gives you two types of output: live analytics during the run, and a detailed shift report at the end.
Three panels at the bottom of the floor view update in real time:
At end of run, the report modal shows:
The operator with the highest utilisation bar is working at or near capacity — this is your constraint. To improve throughput, either reduce their work content (move a process to another station), give them a faster operative, or add a second station in parallel.
Flowcell simulates a JIT (Just-In-Time) replenishment system where dedicated feeder operatives deliver materials from a supermarket to lineside stations on demand.
Each workstation that has non-kanban processes assigned to it requires physical materials — doors, shelves, drawers, etc. These are delivered in trolleys. The simulation manages a two-trolley system per station:
When the active trolley is consumed, it swaps with the stocked trolley — the station keeps working. Meanwhile a feeder is dispatched to pick a new trolley at the supermarket and return it to the station before the stocked trolley runs out.
In the Move Times panel, set the number of feeder operatives (1–3), the round-trip travel time in seconds, and the fetch time per part. These three values determine how long each replenishment cycle takes.
Kanban processes represent materials that are stocked in unlimited bins at the point of use — they never need a feeder trip to replenish.
Mark a process as kanban if the materials it consumes are:
Don't mark a process as kanban if:
Sub-assembly stations run in parallel to the main line, producing components that feed into a specific workstation. They model kitting operations, pre-assembly benches, or offline preparation areas.
Click + Sub-Assy in the Line Layout buttons. A sub-assembly station card appears. Configure:
Sub-assembly stations produce components independently. When a unit arrives at the target workstation needing a process that's fulfilled by sub-assembly, the station waits if no components are ready in the buffer. The sub-assembly operator works continuously to keep the buffer stocked, stopping only when the max buffer is reached (to avoid overproduction).
If you add or rename workstations after creating a sub-assembly station, click the ↻ refresh button on the sub-assy card to update the target dropdown with the latest station names.
Trolley capacity settings control how many parts of each type can fit on a single feeder trolley. This directly affects how often feeders need to make replenishment trips.
For each non-kanban process, you can set a maximum number of parts per trolley. For example, if doors have a trolley capacity of 20, a trolley can carry at most 20 doors. The simulation calculates how many units the trolley can serve before needing a refill.
In the Trolley Capacity section of the Move Times panel, you'll see an entry for each non-kanban process. Set the max parts to a realistic trolley capacity based on your physical trolleys. Set to 0 for no limit — the trolley is assumed to carry enough for the entire run (useful for small, light items).
Smaller trolleys mean more frequent feeder trips. If your trolleys are small relative to batch sizes, you'll need more feeders or faster round-trip times to keep stations supplied. Use the trolley capacity settings to find the right balance.
Save your line configurations to quickly switch between different layouts, shift patterns, or what-if scenarios. Save, Load, Export and Import are available to subscribers.
Give your setup a name in the text field at the top of the Setups panel, then click Save. The current configuration — line layout, processes, product types, move times, operatives, and queue — is stored. Up to 20 setups can be saved.
Click Load next to any saved setup to restore it. The current configuration will be replaced. The simulation must be in reset state to load.
Click Export to download your current setup as a JSON file. This lets you share configurations with colleagues, back them up, or version-control them alongside your lean documentation.
Click Import to load a previously exported JSON file. This replaces the current configuration with the imported setup. Useful for sharing baseline models across a team.
Key terms used in Flowcell and lean manufacturing.
Techniques for getting the most out of Flowcell.
Before testing improvements, build your line exactly as it is. This is your baseline. Every subsequent test should be compared against it. Don't guess — simulate the reality, then improve it.
If you run a variety of product types, shuffle the queue to simulate random arrival order. Some configurations that look good on paper fall apart under high mix. Shuffle is particularly brutal on feeder systems with tightly-sized trolleys.
Set productivity to 80% for one operative and run vs 100%. The productivity impact at the bottleneck is amplified — a 20% slower bottleneck operator can cut line output by more than 20%.
Start with 1 feeder and run. If stations are starving, add a second. Repeat. The minimum number of feeders that prevents stockouts at your trolley size is your target operating model.
Flowcell buffers are unlimited in size — they model the space as infinite. In reality, you have physical constraints. If the buffer fill during your run regularly exceeds your physical space, you have a flow problem upstream.
Before changing anything, save your current setup with a descriptive name like "Baseline - 3 ops - 2 feeders". Then make one change, run, compare, save the new version. Build a library of scenarios you can return to.
When you're iterating quickly — trying many configurations — use 500× warp to get results in seconds. Only slow down to watch the floor animation when something unexpected happens and you need to understand why.
Ready to run simulation