Decision making in the assignment process by using the Hungarian algorithm with OWA operators

Abstract

Assignment processes permit to coordinate two set of variables so each variable of the first set is connected to another variable of the second set. This paper develops a new assignment algorithm by using a wide range of aggregation operators in the Hungarian algorithm. A new process based on the use of the ordered weighted averaging distance (OWAD) operator and the induced OWAD (IOWAD) operator in the Hungarian algorithm is introduced. We refer to it as the Hungarian algorithm with the OWAD operator (HAOWAD) and the Hungarian algorithm with the IOWAD operator (HAIOWAD). The main advantage of this approach is that we can provide a parameterized family of aggregation operators between the minimum and the maximum. Thus, the information can be represented in a more complete way. Furthermore, we also present a general framework by using generalized and quasi-arithmetic means. Therefore, we can consider a wide range of particular cases including the Euclidean and the Minkowski distance. The paper ends with a practical application of the new approach in a financial decision making problem regarding the assignment of investments.

Keywords

• uncertainty modelling
• OWA operators
• Hungarian algorithm
• assignment theory
• distance measure