Determine the evaluation procedure used for competing algorithms in the reported results

Determine whether the performance statistics reported for HPSO, SSGA-I, SSGA-II, IOM, and ASIA in the comparison table were computed using the same procedure as the CR-EA-C statistics, including reevaluation of final solutions and exclusion of failed runs.

Background

The paper reports comparative results for CR-EA-C and five other algorithms on three benchmark problems. For CR-EA-C, the authors explicitly state that final solutions were reevaluated with 300,000 samples and that best, worst, mean, standard deviation, and confidence-interval statistics were computed using only successful runs. However, the authors cannot establish whether the previously published results for the other algorithms used the same evaluation and reporting procedure. This uncertainty affects the direct comparability of the reported performance statistics.

References

However, it is unclear whether the results reported for the remaining algorithms in Table~\ref{tab:results_table} were computed using the same procedure in .

— A Confidence-Driven Evolutionary Algorithm for Noisy Optimization with Joint Chance Constraints  (2609.21318 - Halim et al., 18 Sep 2026) in Section 5, Results and Discussion, subsection “Overview”