Comet's™ high level of abstraction simplifies constraint programming
Your ability to solve difficult business process problems can be limited by tools and technologies. Constraint programming techniques within Comet™ create significant efficiencies because their execution is based upon looking for certain characteristics of a solution. Constraint programming over finite domains implements a novel problem-solving paradigm for optimization in which the focus is on finding a feasible solution that is high quality in a short period of time. Constraints are used to reduce the search space by pruning values from the variable domains which cannot appear in any solution. The end result is a powerful environment that can drastically reduce your time to a high quality solution.
Key Constraint Programming Capabilities in Comet
- Arithmetic, logical meta-constraints (reification), table constraints, and global constraints
- Advanced global constraints including cardinality, sequence, knapsack, packing, assignment, lexicographic ordering, deviation, and spread
- Scheduling abstractions and algorithms, including a variety of resource constraints and their underlying algorithms
- Ability to specify the search procedure using high-level nondeterministic constructs
- Ability to specify the search controllers that determine how the search tree is explored
- Transparent parallelization using a work-stealing model
