Struct StochasticRanking 

pub struct StochasticRanking {
    pub p_f: f64,
    /* private fields */
}

Stochastic ranking for constrained evolutionary optimization.

A bubble-sort-based ranking that probabilistically compares solutions. With probability p_f, feasible-feasible comparisons are based on the objective; with probability 1 - p_f, the violation is used. Infeasible-infeasible comparisons always use the violation.

This balances exploration of infeasible regions (large p_f → more weight on objectives even for infeasible solutions) against driving toward feasibility (small p_f). p_f = 0.45 is a commonly recommended value.

References

Runarsson, T.P. & Yao, X. (2000). Stochastic ranking for constrained evolutionary optimization. IEEE Transactions on Evolutionary Computation, 4(3), 284–294.

Fields

p_f: f64

Probability of using objective comparison between feasible-feasible pairs.

Implementations

impl StochasticRanking

pub fn new(p_f: f64, seed: u64) -> OptimizeResult<Self>

Create a new stochastic ranking instance.

Arguments

• p_f — Probability of objective-based comparison for feasible pairs. Typical range: 0.3–0.5.
• seed — RNG seed.

pub fn rank( &self, objectives: &[f64], violations: &[ViolationSummary], n_passes: usize, ) -> OptimizeResult<Vec<usize>>

Rank a population by stochastic ranking (bubble-sort variant).

Arguments

• objectives — Objective values (one per individual).
• violations — Constraint violation summaries.
• n_passes — Number of bubble-sort passes to perform.

Returns

Ranked indices (index 0 = best-ranked individual).

Trait Implementations

impl Clone for StochasticRanking

fn clone(&self) -> StochasticRanking

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for StochasticRanking

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.