Lag Lab
When demand steps up, production overshoots, and planning calls it forecast error. Lag Lab shows that this is usually the wrong diagnosis, and by how much: it splits the overshoot into the part that is simple arithmetic, the part the delays make unavoidable, the part caused by decision rules, and the part that really is a forecasting problem.
What it does
- Runs the same demand step three waysYour plan as it is; the same plan with a perfect forecast; and a perfect forecast with the pipeline and vacancies fully accounted for and hiring on the capacity gap.
- Splits the overshootThe necessary refill of a bigger inventory and pipeline, the structure that remains, the policy share and the forecast share. They add up exactly, and each comes with its remedy.
- Lets you change the rulesStep size and timing, production, hiring and training delays, inventory and supply-line adjustment times, supply-line and vacancy weights, forecast smoothing, and whether hiring follows the inventory gap or the capacity gap.
- Shows the model itselfA stock-and-flow diagram of every stock, flow and decision rule. Click any variable for its equation, what it reads, what reads it and its value at any week, with the full list of equations and parameters.
- Shows the whole planInventory, backlog, the supply line, the workforce, trainees and vacancies, and utilisation week by week, with overshoot, settling time and oscillation for each, and CSV export.
How it is built
The model is a production and workforce chain in the style of Sterman's Business Dynamics (chapters 17 to 19): backlog, inventory, a third-order production pipeline, workforce, trainees, vacancies and smoothed expected demand, with anchor-and-adjust decision rules. Its tests check the delay and the forecast against their exact solutions, that material and people are conserved exactly, and the textbook results that ignoring the supply line makes inventory oscillate while full accounting does not. The decomposition is tested to add up exactly and to vanish when its cause is removed. Everything runs in your browser.
Author
Lag Lab was designed and built by Daniel Alberto Sepulveda Estay, PhD, founder of Estay Dynamics. He works on system dynamics and network optimisation for supply chain resilience. He holds a PhD from the Technical University of Denmark and an MSc from the Massachusetts Institute of Technology, and is a council member of the System Dynamics Society.
To cite: Sepulveda Estay, D. A. (2026). Lag Lab: capacity that lags demand (version 1.2.0) [Software]. https://kathuman.github.io/claude-projects/lag-lab/