Crate u_nesting_core 

U-Nesting Core

Core traits and abstractions for the U-Nesting spatial optimization engine.

This crate provides the foundational types and traits that are shared between the 2D nesting and 3D bin packing modules.

Core Components

• Geometry traits: Geometry, Geometry2DExt, Geometry3DExt
• Boundary traits: Boundary, Boundary2DExt, Boundary3DExt
• Solver trait: Solver - Common interface for all optimization algorithms
• GA framework: GaRunner, GaProblem - Genetic algorithm infrastructure
• BRKGA framework: BrkgaRunner, BrkgaProblem - Biased Random-Key GA
• SA framework: SaRunner, SaProblem - Simulated Annealing
• Transform types: Transform2D, Transform3D, AABB2D, AABB3D

Optimization Strategies

The Strategy enum defines available optimization algorithms:

Strategy	Speed	Quality	Description
BottomLeftFill	Fast	Basic	Greedy bottom-left placement
NfpGuided	Medium	Good	NFP-based optimal positioning (2D)
GeneticAlgorithm	Slow	High	GA with permutation encoding
Brkga	Medium	High	Biased Random-Key GA
SimulatedAnnealing	Medium	High	Temperature-based optimization
ExtremePoint	Fast	Good	EP heuristic (3D only)

Configuration

Use Config to configure solver behavior:

use u_nesting_core::{Config, Strategy};

let config = Config::new()
    .with_strategy(Strategy::GeneticAlgorithm)
    .with_spacing(2.0)
    .with_margin(5.0)
    .with_time_limit(30000);

Feature Flags

• serde: Enable serialization/deserialization support

Re-exports

pub use alns::AlnsConfig;

pub use alns::AlnsProblem;

pub use alns::AlnsProgress;

pub use alns::AlnsResult;

pub use alns::AlnsRunner;

pub use alns::AlnsSolution;

pub use alns::DestroyOperatorId;

pub use alns::DestroyResult;

pub use alns::OperatorStats;

pub use alns::RepairOperatorId;

pub use alns::RepairResult;

pub use brkga::BrkgaConfig;

pub use brkga::BrkgaProblem;

pub use brkga::BrkgaProgress;

pub use brkga::BrkgaResult;

pub use brkga::BrkgaRunner;

pub use brkga::RandomKeyChromosome;

pub use error::Error;

pub use error::Result;

pub use exact::ExactConfig;

pub use exact::ExactResult;

pub use exact::SolutionStatus;

pub use ga::GaConfig;

pub use ga::GaProblem;

pub use ga::GaProgress;

pub use ga::GaResult;

pub use ga::GaRunner;

pub use ga::Individual;

pub use gdrr::GdrrConfig;

pub use gdrr::GdrrProblem;

pub use gdrr::GdrrProgress;

pub use gdrr::GdrrResult;

pub use gdrr::GdrrRunner;

pub use gdrr::GdrrSolution;

pub use gdrr::RecreateResult;

pub use gdrr::RuinResult;

pub use gdrr::RuinType;

pub use gdrr::RuinedItem;

pub use geometry::check_quantity;

pub use geometry::Boundary;

pub use geometry::Boundary2DExt;

pub use geometry::Boundary3DExt;

pub use geometry::Geometry;

pub use geometry::Geometry2DExt;

pub use geometry::Geometry3DExt;

pub use geometry::GeometryId;

pub use geometry::RotationConstraint;

pub use placement::Placement;

pub use result::SolveResult;

pub use result::SolveSummary;

pub use sa::CoolingSchedule;

pub use sa::NeighborhoodOperator;

pub use sa::PermutationSolution;

pub use sa::SaConfig;

pub use sa::SaProblem;

pub use sa::SaProgress;

pub use sa::SaResult;

pub use sa::SaRunner;

pub use sa::SaSolution;

pub use solver::Config;

pub use solver::ProgressCallback;

pub use solver::ProgressInfo;

pub use solver::Solver;

pub use solver::Strategy;

pub use transform::Transform2D;

pub use transform::Transform3D;

pub use transform::AABB2D;

pub use transform::AABB3D;

Modules

alns

Adaptive Large Neighborhood Search (ALNS) Framework

brkga

Biased Random-Key Genetic Algorithm (BRKGA) framework.

error

Error types for U-Nesting.

exact

Exact solver configuration and result types.

ga

Genetic Algorithm framework for optimization.

gdrr

Goal-Driven Ruin and Recreate (GDRR) Framework

geom

Re-exports from u-geometry for direct access to generic computational geometry.

geometry

Core geometry traits and types.

memory

Memory optimization utilities.

metaheur

Re-exports from u-metaheur for direct access to generic optimization frameworks.

placement

Placement representation for positioned geometries.

result

Solve result representation.

robust

Robust geometric predicates for numerical stability.

sa

Simulated Annealing framework for optimization.

solver

Solver traits and configuration.

timing

WASM-compatible timing abstraction.

transform

Transform types for 2D and 3D coordinate transformations.

Enums

StopReason

Why a solve stopped – the same vocabulary every runner in u-metaheur reports, so a nesting run and any other search say it in one set of words. Each strategy here sets it where it stops; Completed is a fill that tried every piece, or a search that spent the budget it is given internally. Why a run stopped: the run’s own account of the solution it returns.