Algorithm Optimisations

1. Pruning Conditions

• Equivalent Schedule checks
• Normalisation Schedule checks
• Identical task ordering checks (to be implemented)

2. Data Structures

• Converted all lists to corresponding HashSets/HashMaps
  • This improves time complexity as Sets and Maps have O(1) access
• Converted Object lists (i.e. Nodes/Edges object classes) to primitive array data structure types
  • Improves memory allocation
  • Improves execution time as object creation takes time

3. Algorithm Heuristic

• The heuristic is finding the max between Data Ready Time (DRT), Idle Time (IT) and Bottom Level (BL)
  • Estimates the cost of the partial schedule so that we only expand the partial schedule that has the least likely overall cost