Free line balancing calculator
Enter your work steps, their times and which steps must come first. The calculator finds the fewest stations for your takt time, proves that no balance needs fewer, and then evens out the load between them. It runs in your browser: nothing you type is sent anywhere.
Work it out from demand
Takt time = available time ÷ demand. See the takt time guide.
| # | Work step | Time (s) | After steps | Remove |
|---|
Paste steps from a spreadsheet
One step per line: name, time in seconds, then the numbers of the steps it comes after. Columns copied from Excel work, and so does a Scholl .IN2 benchmark file.
| Station | Steps | Time | Of takt |
|---|
This is one product, on paper. See what a full design adds: the floor, the machines and a simulation of the line.
A balance on paper is where a line design starts
The calculator answers one question well: how few stations one product needs at a takt time. A line that works on the floor needs more answers than that. Here is exactly what you get for free, and what the Machsizer app adds.
Free in the calculator
No sign-up, no limit, nothing stored.
- The fewest stations, proven. An exact solution of SALBP-1 for your steps and takt time, not a rule of thumb.
- The most even load. Of all balances with that many stations, the one whose busiest station has the most margin.
- The numbers that matter. Balance efficiency, idle time per unit, the busiest station and the capacity it allows.
- Textbook rules side by side. What ranked positional weight and largest candidate reach on the same steps.
- Steps longer than takt. Parallel stations for the step, how busy each one is, and how much shorter the step must be to save one.
- Your own data in, a link out. Paste from Excel or a Scholl benchmark file; share the balance as a link.
In the Machsizer app
Early access, 14-day trial, by request.
- Your product mix. Options and variants with their share of demand, balanced as the mix the line will really run. Products and variants
- Machines decide what goes where. Which machine can do which step and how long it takes, turns of the product between steps, and stations sized to the part. Machines and stations
- Subassemblies and feeder lines. Parts made on their own lines and joined to the main line where they are needed. Guide
- Laid out on the floor. Stations, buffers and material placed on your floor area; transport times come from the layout. Layout and space
- Simulated, not assumed. Varying times, breakdowns and repairs, full buffers, shared operators, shifts and breaks, over as many runs as you need. Breakdowns People
- Whether a parallel station pays off. On paper a step over takt needs one; on the floor, breakdowns elsewhere may hold the line back anyway. Improve simulates one station more or fewer and shows what each adds. Optimisation
- Why output falls short. The station that holds the line back, and whether it lacks capacity or waits on its neighbours. Explaining results
- Designs compared. Output, equipment, floor area, lead time, cost and operators, side by side. Improve
See everything Machsizer models, each capability shown by a test that builds and runs a factory.
Get early accessWhat the calculator does
Line balancing assigns work steps to stations in order, so that no station's work exceeds the takt time and no step lands before a step it depends on. Finding the fewest stations under those rules is the simple assembly line balancing problem, SALBP-1.
The calculator first solves it exactly. It searches station by station, fills each station as full as the order of steps allows, and stops a branch as soon as it cannot beat the best balance found so far. When the search finishes, the station count is proven: no balance of these steps needs fewer stations at this takt.
Many balances share that station count, and some put almost all the slack in one station. So the calculator searches again, for the smallest busiest station that still fits in that many stations. That gives every station some margin for the variation a real line has.
The search stops after a few seconds. On a very large step list it may not finish; it then shows the best balance it found and the least number of stations any balance needs, and says that the count is not proven.
Why rules of thumb leave stations on the table
Textbook methods such as ranked positional weight and largest candidate build one balance with a fixed priority. Each choice is made once and never revisited. They are quick to do by hand, but nothing tells you how far the result is from the best.
The Gunther example above, a 35-task instance from Scholl's published SALBP-1 data set, shows the gap. At a cycle time of 41, ranked positional weight needs 16 stations and largest candidate 15, as the calculator applies them. The optimum is 14. Each extra station is equipment, floor space and often an operator for the life of the line.
The calculator shows what both rules reach on your own steps, so you can see the difference for yourself.
What the numbers mean
Least possible stations = ⌈ total work ÷ takt ⌉
A quick bound: even perfectly packed stations cannot do better. The order of steps and long steps often force more.
Balance efficiency = total work ÷ (stations × takt)
The share of paid station time spent working. The rest is idle time, built into the line.
Balance delay = 1 − balance efficiency
The same idle time, as a share. Two balances with the same station count have the same balance delay, however unevenly the work is spread.
Smoothness index = √ Σ (busiest station time − station time)²
How evenly the work is spread. Zero means every station carries the same work. It is what tells two balances with the same station count apart.
Output = 3600 ÷ busiest station time (units per hour)
On paper, a line runs no faster than its busiest station. If that station sits at 100% of takt, any variation makes the line miss takt.
What a balance on paper leaves out
The calculator treats every step as taking exactly its time, every station as able to do every step, and one product as the only product. Real lines differ. Times vary and machines break down. Some steps need a particular machine, a turned product or a long station. Units travel between stations, and the floor has room for only so much.
A station at 100% of takt on paper is the first to fall behind on the floor. That is why a design is checked with a simulation of its actual layout before equipment is bought. Machsizer does that in the same step as the balance; see what it adds. Its balancer reaches every published optimum on the open benchmarks; see Validation.
Questions
Is the line balancing calculator free?
Yes. There is no sign-up and no limit on use, and nothing you enter is stored. It is a free part of Machsizer's website, not a trial.
How is this different from a balance efficiency calculator?
A balance efficiency calculator takes the total work, a number of stations and a cycle time, and returns a percentage. It assumes you already have a balance. This calculator makes the balance: it decides which step goes to which station, keeps every step after the steps it depends on, and finds the fewest stations that can do it. The efficiency, balance delay, idle time and smoothness index then come from that real assignment.
What does the Machsizer app add?
Everything a line needs beyond the balance: product mix and variants, which machine does what, feeder lines, the layout on your floor and a simulation with variation and breakdowns. See the comparison above, or everything Machsizer models.
Can I balance a mixed-model line here?
Not in this calculator: it balances one product. Enter the average times of your mix as a first estimate. Machsizer balances the mix itself and simulates it.
Is anything I enter stored?
No. The calculation runs in your browser. A copied link holds the steps after the # sign, which browsers never send to a server.
Where do the benchmark examples come from?
Jackson (1956) and Gunther (1983) are instances from A. Scholl's SALBP-1 data set, used across the line balancing literature. Their optimal station counts are published, and the calculator reaches them.