Run sensitivity analysis

Varies one objective's weight for a stakeholder group and returns how satisfaction scores change. Identifies tipping points where the top-ranked alternative changes. Use this to test whether the ranking is robust or fragile.

Usage

sdm_run_sensitivity(
  scenario,
  group,
  objective,
  weight_seq = seq(0, 1, by = 0.05),
  dry_run = FALSE,
  verbosity = 0
)

Arguments

scenario

An sdm_scenario() object.

group

Character. Generic key of the stakeholder group (e.g., "group_1").

objective

Character. Generic key of the objective to vary (e.g., "obj_1").

weight_seq

Numeric vector. Weight values to evaluate, typically seq(0, 1, by = 0.05).

dry_run

Logical. If TRUE, return the request without sending.

verbosity

Integer. Verbosity level (0 = silent).

Value

A data frame with columns:

weight

The weight value being tested.

One column per alternative

Satisfaction scores (0–100) at each weight, named after the alternative.

top_alternative

Name of the highest-scoring alternative at that weight.

tipping_point

Logical. TRUE if the top alternative changed from the previous weight step. Tipping points close to the current weight indicate a fragile ranking.

Details

When the focal objective's weight is set to w, the remaining weights are rescaled proportionally so the total still sums to 1.0. The relative proportions among the other objectives stay the same.

Examples

if (FALSE) { # \dontrun{
sens <- sdm_run_sensitivity(
  scenario = scn,
  group = "group_1",
  objective = "obj_1",
  weight_seq = seq(0, 1, by = 0.05)
)

# Find tipping points
sens[sens$tipping_point, ]
} # }