Synthetic Dataset for Ensemble Learning Approach to Predict the Best Scheduling Algorithm for Single-Machine Tardiness Minimization

This dataset accompanies the manuscript entitled: "Ensemble Learning Approach to Predict the Single-Machine Best Scheduling Algorithm for Minimizing a Tardiness Criterion". The dataset contains 7,200 synthetic scheduling instances generated for the experimental evaluation reported in the manuscript. The instances were generated using different combinations of:- number of jobs n ∈ {10, 15, 20, ..., 50}- due-date parameter α ∈ {0.2, 0.4, 0.6, 0.8} For each (n, α) combination, 200 instances were generated. Processing times were randomly generated in {1,...,20}, and due dates were uniformly generated in [0, α × Σpj]. The repository also contains the experimental results used in the study.