Resource Sharing with Local Differential Privacy

Utku Karaca

Efficient use of resources is one of the major objectives in every industry. Through collaboration, companies can coordinate their activities, increase their utilizations, and obtain significant savings. Although coordinating activities and sharing resources can provide significant benefits to companies, they also pose several challenges. These partnerships are mainly built around information exchange to coordinate the collective decision-making process. However, individual partners, though often working towards a common goal, can be competitors and may be unwilling or unable to fully disclose sensitive information about their operations. The data privacy concern in collaborations raises an important question: How can one mathematically guarantee data privacy while conducting optimization in a multi-party resource-sharing setting? In this work, we address the aforementioned issues and the question above for a multi-party resource-sharing model in a general linear optimization framework. We start with modeling the whole problem as a mathematical program, where all parties are required to exchange information to obtain the optimal objective function value. This information bears private data from each party in terms of coefficients used in the mathematical program. Moreover, the parties also consider the individual optimal solutions as private. In this setting, the concern for the parties is the privacy of their data and their optimal allocations. We propose a two-step approach to meet the privacy requirements of the parties. In the first step, we obtain a reformulated model that is amenable to a decomposition scheme. Although this scheme eliminates almost all data exchanges, it does not provide a formal privacy guarantee. In the second step, we provide this guarantee with a local differential privacy algorithm, which does not need a trusted aggregator, at the expense of deviating slightly from the optimality. We provide bounds on this deviation and discuss the consequences of these theoretical results. We also propose a novel modification to increase the efficiency of the algorithm in terms of reducing the theoretical optimality gap. The talk concludes with a numerical experiment on a planning problem that demonstrates an application of the proposed approach. As we work with a general linear optimization model, our analysis and discussion can be used in different application areas including production planning, logistics, and revenue management.

Short Bio

Utku Karaca is a Senior Data Scientist at Air Products. He earned his PhD in Econometrics from the Econometric Institute, Erasmus University Rotterdam, and MSc and BSc in Industrial Engineering from Bilkent University, Ankara, Turkey. Utku’s research interests include the intersection of data privacy and mathematical modeling, collaborative resource sharing, revenue management, assortment optimization and pricing. He presented his works at seminars and conferences in Ireland, Turkey and the Netherlands. His work has been published in European Journal of Operational Research. In his current job, Utku drives data science and optimization projects by applying his expertise in artificial intelligence and operations research.

Venue

Friday, October 24th, 2025, 4:00 pm

Online

The link for the seminar is: https://teams.microsoft.com/l/meetup-join/19%3ameeting_MzU0N2ZlY2MtODY4M...

Meeting ID: 352 571 542 876 3

Passcode: 7ci9pr3o