A system dynamics simulation defines stocks, flows, auxiliaries, units, and initial values. Calibration, dimensional checks, and sensitivity analysis limit false precision.

1. 1

   Reference mode

   Reference mode

   

   Goal

   Produce reference mode so the core question “What feedback and accumulation structure generates the behaviour, and how do scenarios respond?” can be answered with stronger support.

   Inputs for this step

   • Reference mode and time series
   • causal hypothesis
   • parameter sources and units

   Actions

   • Define problem behaviour and horizon as data curves.
   • Record source, rationale, and open uncertainty directly in reference mode.

   Output

   reference mode

   reference mode · source · uncertainty · next decision

   Quality check

   Are measure, unit, and horizon clear?

   Pitfall

   A stock-flow model is calibrated without unit checks and creates false precision. Expose reference mode, equations, units, and independent validation; show uncertainty as a band.

   Timing: 10–20 minutes

2. 2

   Build stock-flow

   Build stock-flow

   

   Goal

   Produce simulation model so the core question “What feedback and accumulation structure generates the behaviour, and how do scenarios respond?” can be answered with stronger support.

   Inputs for this step

   • reference mode
   • causal hypothesis
   • parameter sources and units

   Actions

   • Model stocks, flows, feedback, and delays with equations.
   • Record source, rationale, and open uncertainty directly in simulation model.

   Output

   simulation model

   simulation model · source · uncertainty · next decision

   Quality check

   Does every equation pass dimensional analysis?

   Pitfall

   A stock-flow model is calibrated without unit checks and creates false precision. Expose reference mode, equations, units, and independent validation; show uncertainty as a band.

   Timing: 10–20 minutes

3. 3

   Validate model

   Validate model

   

   Goal

   Produce validation record so the core question “What feedback and accumulation structure generates the behaviour, and how do scenarios respond?” can be answered with stronger support.

   Inputs for this step

   • simulation model
   • causal hypothesis
   • parameter sources and units

   Actions

   • Test structure, extreme conditions, and historical fit.
   • Record source, rationale, and open uncertainty directly in validation record.

   Output

   validation record

   validation record · source · uncertainty · next decision

   Quality check

   Was the model tested against uncalibrated data?

   Pitfall

   A stock-flow model is calibrated without unit checks and creates false precision. Expose reference mode, equations, units, and independent validation; show uncertainty as a band.

   Timing: 10–20 minutes

4. 4

   Run scenarios

   Run scenarios

   

   Goal

   Produce scenario comparison so the core question “What feedback and accumulation structure generates the behaviour, and how do scenarios respond?” can be answered with stronger support.

   Inputs for this step

   • validation record
   • causal hypothesis
   • parameter sources and units

   Actions

   • Compare interventions and parameter sensitivity over time.
   • Record source, rationale, and open uncertainty directly in scenario comparison.

   Output

   scenario comparison

   scenario comparison · source · uncertainty · next decision

   Quality check

   Do assumptions and ranges remain visible?

   Pitfall

   A stock-flow model is calibrated without unit checks and creates false precision. Expose reference mode, equations, units, and independent validation; show uncertainty as a band.

   Timing: 10–20 minutes