Hybrid methods that combine constraint programming with mathematical programming make essential use of continuous relaxations for global constraints. We state a relaxation for the cumulative constraint. In particular we identify facet-defining inequalities for problems in which some jobs have the same duration, release time, and resource consumption rate. We also identify a much larger class of valid inequalities that exist in all problems.
|Name||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|
|Conference||8th International Conference on Principles and Practice of Constraint Programming, CP 2002|
|Period||9/09/02 → 13/09/02|
- Theoretical Computer Science
- Computer Science(all)