Learning the thresholds in the ORESTE method from historical preference information

Abstract

The ORESTE (organísation, rangement et Synthèse de données relarionnelles, in French, meaning organization, storage and synthesis of relevant data in English) method is an effective multi-criteria decision-making method which takes into account the importance of both criteria and alternatives. Nevertheless, how to objectively determine the thresholds used in the ORESTE method is still an unsolved issue. In this regard, this study proposes a model that combines the threshold learning method based on historical data and the traditional MCDM (multi-criteria decision-making) method, to learn the thresholds used in the ORESTE method from the preference information provided by stakeholders. We implement a simulation analysis to verify the feasibility of this model and reflect the relation between the weights balancing the importance of alternatives and criteria and the thresholds, and the relation among different thresholds. Finally, an illustration concerning the selection of lung cancer screening methods is given to show the applicability of the threshold learning model. Sensitivity analysis, comparative analysis, and managerial insights are also provided.

Keywords:

• Multiple criteria analysis
• threshold learning
• ORESTE
• preference learning
• lung cancer screening

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

The work was supported by the National Natural Science Foundation of China (Nos. 71771156, 71971145, 72171158).