Preference modelling in sorting problems: Multiple criteria decision aid and statistical learning perspectives

Abstract

Many decision problems in a variety of fields such as marketing, quality prediction, and economics correspond to the sorting decision problematic where an ordinal scale is used to express a preference of objects. Both Multiple Criteria Decision Aid and Statistical Learning fields offer methodologies to represent the preference of the decision maker facing the sorting problem, however, there are differences in terminology, objectives, key assumptions, and solution philosophies. In this context, this paper aims to explain these differences as well as similarities and connections between these two fields by reviewing exemplary methodologies in sorting problems.As we discuss, there are significant research opportunities for developing new methodologies by exploiting the strong aspects of these two fields.

A robust multi-objective model for healthcare resource management and location planning during pandemics

“…Instead of a scalarization approach, interactive approaches where the DM interacts with the method for searching the most preferred solution or a posteriori approaches where a Pareto frontier is presented to the DM can be used to solve the problem. When the DM does not have a clear idea of the problem’s mathematical structure, then the problem consists not only of finding the best method to apply to DM’s problem but also establishing a representation of the problem in a constructive way (Erişkin 2021 ). In these cases, the interactive approaches are particularly useful.…”

Special issue on multiple criteria sorting methods

“…Another survey by Erişkin (2021) has just been published in the present journal. It reverts to the positioning of multiple criteria sorting methods w.r.t.…”

Data analytics during pandemics: a transportation and location planning perspective

“…In general, studies that employ surveys and questionnaires for transportation-related data analytics deal with ranking and choice decision-making problematics. In ranking decision-problematic, alternatives are rank-ordered from the most preferred to the least preferred while within the choice decision-making problematic, the best or limited set of best alternatives are identified (Erişkin 2021 ). Within the context of choice decision-making problematic, the researchers aimed to explain the relationship between people’s travel mode choices and other explanatory factors such as demographics (i.e., monthly income, age, gender), distance to public transport services, number of confirmed cases, and vehicle ownership with statistical models.…”