Privacy Attacks on Schedule-Driven Data

Schedules define how resources process jobs in diverse domains, reaching from healthcare to transportation, and, therefore, denote a valuable starting point for analysis of the underlying system. However, publishing schedule may disclose private information on considered jobs. In this paper, we provide first threat model published schedules, thereby defining completely new class data privacy problems. We then propose distance-based measures assess loss incurred by schedule, and show their theoretical properties an uninformed adversary, which can be used as benchmark informed attacks. attack phrased inverse scheduling problem. instantiate idea formulating well-studied single-machine problem, namely minimizing total weighted completion times. An empirical evaluation synthetic problems shows effectiveness attacks compares results bounds