Skip to content
All apps

Flow lab: throughput, WIP, cycle time

About this app

A four-station line with yield loss in 3D: utilisation u = λ · t per station, and the bottleneck moves from station 2 to station 4 as yield loss falls. Two live cards show cycle time over utilisation (Kingman/VUT) and Little’s law from a discrete-event simulation, right up to an unstable system.

With explain mode: an AI tutor walks you through the app, points out what matters in the scene and answers your questions.

Factory physics in brief

In a production line, the utilisation of a station is u = λ · t: arrival rate times processing time. The station with the highest utilisation is the bottleneck, not necessarily the slowest one. Scrap reduces the flow reaching downstream stations. Cycle time grows disproportionately with utilisation, which Kingman’s approximation (the VUT equation) captures: CT = t · (1 + (c²ₐ + c²ₑ)/2 · u/(1 − u)). Little’s Law links inventory, throughput and cycle time: WIP = TH × CT. Once u reaches 100 %, the queue grows without bound.

What you see

A line of four stations with 0.7, 0.8, 0.9 and 1.0 minutes of processing time runs as a discrete-event simulation in 3D. A ring around each machine shows the calculated utilisation, a marker on it the measured one. Two live cards follow the run: cycle time against utilisation with Kingman curves and the measured point, and WIP compared with TH × CT over a moving window. You can set the input rate λ, the scrap rate, the station after which scrap occurs and the variability c².

Experiments

The lecture example has 50 % scrap after station 2. Station 2 then runs at 80 % and is the bottleneck, even though station 4 takes longest. Lower the scrap below 20 % and the bottleneck moves to station 4. Set variability to 0 and cycle time stays close to the pure processing time. Then raise λ until one station reaches 100 % and watch inventory and cycle time run away.

Frequently asked questions

What does Little’s Law say?
Little’s Law states that average work in process equals throughput times average cycle time: WIP = TH × CT. It holds for long-run averages of a stable system. Short measurement windows or an overloaded system will deviate.
What is Kingman’s formula used for?
Kingman’s approximation estimates the waiting time at a station from variability, utilisation and processing time (VUT). The factor u/(1 − u) explains why waiting times rise sharply at high utilisation. More variation in arrivals or processing makes the effect stronger.
How do you find the bottleneck of a production line?
Compute the utilisation u = λ · t for each station, using the rate that actually arrives there. The station with the highest utilisation is the bottleneck. Scrap or branching changes the arrival rates and can shift the bottleneck.
Does the app calculate or simulate?
Both. The line runs as a discrete-event simulation, and the measured values appear next to the formula values. You can see where Kingman and Little’s Law fit and where short windows or overload cause deviations.

Subject: Technical logistics & intralogistics

More from Technical logistics & intralogistics

A VisuApp by heyprof: interactive, free in your browser, no sign-up. What is a VisuApp? · All apps