Result: A review of multiobjective linear programming methods. I

The paper is the first part of a critical insight into the existing methods of multiobjective linear programming (MOLP). On the background of a general outline of the problem, the methods are classified from the point of view of type and way of formulation of the global model of decision-maker's preferences (GMDP). Three main classes of the MOLP methods are distinguished: (1) methods with a priori synthesis of the GMDP, (2) methods with interactive synthesis of the GMDP, and (3) methods with a posteriori synthesis of the GMDP. These classes are generally characterized, and then a review of the methods of class (1) is given. Class (1) is divided into two subclasses of methods with (a) noncompensatory and (b) compensatory GMDP's. Further, methods of class (1) are categorized subject to the type of information given: (a,\(\alpha)\) hierarchy in the set of criteria, (a,\(\beta)\) thresholds for particular criteria, (b,\(\alpha)\) utility function, (b,\(\beta)\) reference point in the criterion space.