A totally non-compensatory multi-criteria method for evaluating and improving level of satisfaction (LoS): proposal and application on Airport Terminal of Passengers

TL;DR

This paper introduces a novel non-compensatory multi-criteria decision model for evaluating passenger satisfaction at airports, allowing individual criteria selection and providing tailored improvement insights based on extensive survey data.

Contribution

It presents the first application of a fully non-compensatory MCDA model using ELECTRE TRI ME for airport service evaluation, accommodating diverse evaluator criteria.

Findings

Identified key criteria for passenger satisfaction at airports.

Clustered airport terminals based on satisfaction levels.

Provided targeted improvement recommendations for each cluster.

Abstract

To evaluate and assign a service according customer's level of satisfaction (LoS) is a relevant issue in operations management. This is a typical situation in which the evaluators, have passed by heterogeneous experiences along their life which implies they could consider different variables when evaluating a product. Despite it, the models for measuring Los usually consider a homogeneous set of criteria when facing LoS evaluation. This study applies a totally non-compensatory modeling that allows each customer to select the criteria, from a whole set of aspects, the customer wants to use for evaluating LoS. The proposal was tested in evaluating LoS regarding the services provided by Airport Terminal of Passengers (ATPs) in Brazil, with data collected in a survey involving 19,240 passengers, interviewed at 15 Brazilian international airports. The data collected was imputed into ELECTRE TRI ME algorithm to obtain the a credibility degree of sorting the instances. The values of credibility degree were them used to obtain groups of ATPs. Finally, the statistical modes of the evaluations in each group were analyzed and compared. The proposal allowed a full non-compensatory approach to obtain the credibility degree even when considering perceptions from several evaluators that could use different criteria. As a result, it was identified, for each cluster of ATP, the criteria sets to be improved and even those to be prioritized. The pioneer modeling proposed in this article for evaluating LoS plus its instancing in ATPs terminals represents an original advance in the establishment of a multi-criteria decision aid (MCDA) model to assess the quality of services and fills a relevant gap for a full non-compensatory approach able to classify the LoS in the airport context, considering perceptions of multiple evaluators even if they use different criteria in their evaluations.