A constraint-satisfaction problem (CSP) is a computational paradigm in which a solution is found by satisfying a finite set of constraints over a finite set of variables. A CSP has three parameters: (1) a set of variables, each with a domain of allowable values, (2) a set of constraints specifying allowable combinations of values for subsets of variables, and (3) an assignment that satisfies all constraints simultaneously. The persistence mechanism is algorithmic: constraint propagation and backtracking search enforce the constraints across iterations until a valid assignment is found or the search space is exhausted. The concept persists through formal education in computer science and operations research, and through its instantiation in solvers, planning systems, scheduling software, and configuration tools.
[formal: constraint_satisfactio | substrate: mind | horizon: hours | explicit: yes | epoch: 0.01]