decision_making_core 0.2.0

Pure Dart multi-criteria decision-making (MCDM) core library implementing AHP, SAW, and TOPSIS algorithms.

Changelog

All notable changes to decision_making_core will be documented in this file.

0.2.0

Added

• Objective criteria weighting (Shannon Entropy & CRITIC): Added ObjectiveWeightCalculator.calculateEntropy, ObjectiveWeightCalculator.calculateCritic, ObjectiveWeightCalculator.applyWeightsToCriteria, ObjectiveWeightCalculator.applyWeightsToMatrix, and ObjectiveWeightCalculator.applyWeightsToTopsisMatrix, returning EntropyWeightResult (proportionMatrix, entropy, divergence, weights, toJson()) and CriticWeightResult (normalizedMatrix, standardDeviations, correlationMatrix, informationContent, weights, toJson()).

0.1.0

Initial Release

• Pure Dart multi-criteria decision-making (MCDM) computational library (SDK >=3.0.0 <4.0.0).
• Zero Flutter dependencies: runs seamlessly in CLI tools, backend servers, Dart frog/Shelf, and frontend frameworks.
• Synchronous calculation engines:
  • SAW (Simple Additive Weighting): SawCalculator.normalize(), SawCalculator.calculateScores(), SawCalculator.calculate(), SawCalculator.calculateDetailed(), and SawCalculator.analyzeSensitivity() returning SensitivityAnalysisResult with systematic weight perturbations and rank reversal detection.
  • TOPSIS (Technique for Order Preference by Similarity to Ideal Solution): TopsisCalculator.normalizeEuclidean(), TopsisCalculator.weightMatrix(), TopsisCalculator.calculateIdealSolutions(), TopsisCalculator.calculateSeparationDistances(), TopsisCalculator.calculateRelativeCloseness(), TopsisCalculator.calculate(), TopsisCalculator.calculateDetailed(), and TopsisCalculator.analyzeSensitivity() returning SensitivityAnalysisResult with systematic weight perturbations and rank reversal detection.
  • AHP (Analytic Hierarchy Process): AhpCalculator.identification(), AhpCalculator.generateHierarchy(), AhpCalculator.generatePairwiseCriteria(), AhpCalculator.generatePairwiseAlternative(), AhpCalculator.generateResultPairwiseMatrixCriteria(), AhpCalculator.generateResultPairwiseMatrixAlternative(), AhpCalculator.calculateEigenVector(), AhpCalculator.checkConsistencyRatio(), AhpCalculator.calculateFinalScore(), AhpCalculator.calculate(), and AhpCalculator.calculateDetailed() returning AhpDetailedResult with complete comparison matrices, normalized matrices, weights (w_j), consistency metrics (lambda_max, CI, RI, CR), scores per criteria, and toJson().
• Low-level AHP helpers (getRI, riTable, normalizeComparisonMatrix) are marked @internal.
• Pure domain models and validation utilities (MatrixValidator).
• Comprehensive unit and golden test suite verified with tolerance 1e-4.