Masking primal and dual models for data privacy in network revenue management

• DOI: 10.1016/j.ejor.2022.11.025
• Publication Date: 2022-11-24T15:09:50Z
• Authors: Utku Karaca; Ş. İlker Birbil; Nurşen Aydın; Gizem Mullaoğlu
• Keywords: Private information retrieval

We study a collaborative revenue management problem where multiple decentralized parties agree to share some of their capacities. This collaboration is performed by constructing large mathematical programming model that available all parties. The then use the solution this in own capacity control systems. In setting, however, major concern for privacy input data, along with individual optimal solutions. first reformulate general linear can be used wide range network problems. Then we address data reformulated and propose an approach based on solving equivalent data-private constructed masking via random transformations. Our main result shows that, after model, each party safely access only its allocation decisions. also discuss security transformed considered multi-party setting. Simulation experiments are conducted support our results evaluate computational efficiency proposed model. work provides analytical insights how manage shared resources while ensuring privacy. Constructing requires information exchange between may not possible practice. Including problems sharing new literature relevant