Reproducibility Data and Code for “Adaptive Sparse Computation and Threshold Optimization for a BMMAP/PH/1 Retrial Queue with Markovian Priorities and Server Unreliability”

Description

This dataset contains the source code, model configurations, numerical outputs, reported-result tables, validation records, and automated tests supporting the manuscript “Adaptive Sparse Computation and Threshold Optimization for a BMMAP/PH/1 Retrial Queue with Markovian Priorities and Server Unreliability.” The package reproduces the sparse continuous-time Markov chain calculations, independent discrete-event simulation results, adaptive orbit-cutoff diagnostics, joint capacity-threshold regret analysis, combined batch-composition and marked-BMMAP stress tests, saturated-system drift audit, solver-ordering benchmarks, and structural sparsity results reported in the manuscript. Fixed model configurations, simulation seeds, validation logs, and SHA-256 checksums are included to support independent verification. The canonical numerical configuration uses a retrial rate of 0.35 and buffer-impatience rates of (0.01, 0.03, 0.04, 0.05). The analytical and simulation implementations are separately constructed, and the retained validation results include 28 CTMC–DES comparisons, automated numerical checks, and reproducibility tests. No personal, confidential, human-subject, or third-party proprietary data are included.

Institutions

• Sangji University

  Gangwon-do, Wŏnju
• National Institute of Technology Raipur

  Chhattisgarh, Raipur

Categories

Funders

• National Research Foundation of Korea

  Ministry of Science and ICT

  Daejeon, Daejeon

  Grant ID: NRF-2022R1I1A3067543

Licence