Description: Constraint Programming site
programming (6711) global (4139) solver (93) sequence (87) rostering (30) constraint (17) cardinality (5) alldiff (4) ac algorithm (4) arc consistency (4)
A constraint network (CN) consists of a set of variables; domains of possible values associated with each of these variables; and a set of constraints that link up the variables and define the set of combinations of values that are allowed. The search for an instantiation of all variables that satisfies all the constraints is called a Constraint Satisfaction Problem (CSP), and such an instantiation is called a solution of a CSP.
A lot of problems can be easily coded in terms of CSP. For instance, CSP has already been used to solve problems of scene analysis, placement, resource allocation, crew scheduling, time tabling, scheduling, frequency allocation, car sequencing, and so on.
Unfortunately, a CSP is an NP-Complete problem. Thus, much work has been carried out in order to try to reduce the time needed to solve a CSP. Constraint Programming (CP) is one of these techniques.