Relief-based Anesthesiologist Scheduling with Stochastic Surgery Durations

We present a two-stage stochastic programming model for scheduling anesthesiologists to operating rooms under uncertainty in surgery durations. The proposed model takes a relief order to balance anesthesiologists’ workload as input and captures the trade-offs between anesthesiologist relief times, handoffs and under-staffing. To address the computational challenges of solving the proposed model, we derive supervalid … Read more

Covering for Set-Valued Mappings in the Absence of Metric Regularity

Covering properties build the foundation of stability and sensitivity analysis of solutions to a generalized equation and more specific optimization-related stationarity and equilibrium problems. It has been well-understood that metric regularity of the mapping defining the generalized equation is a key to furnish Lipschitzian stability of the solution of interest. With this work, we want … Read more

Extrapolation-based Direct Search for Nonsmooth Stochastic Zeroth-Order Optimization

We propose and analyze a stochastic direct-search method for unconstrained zeroth-order minimization of locally Lipschitz, possibly nonsmooth, objectives. The method combines random polling directions with a stochastic extrapolating line search based on a sufficient-decrease test of order \(p\). Under conditional accuracy assumptions on the stochastic estimates, which can be verified for mean-zero finite-higher-moment oracle noise … Read more

Stochastic Bilevel Optimization for the Network Design of Multimodal Transit Systems with Heterogeneous Rider Preferences under Uncertain Travel Times and Demand

Designing efficient and user-friendly multimodal transit networks is critical for modern urban mobility. We study a novel stochastic multimodal transit network design problem that integrates fixed-route services with on-demand shuttles, explicitly accounting for heterogeneous rider preferences, uncertain travel times, and passenger demand. The hierarchical decision-making process is modeled using a two-stage stochastic bilevel optimization problem, … Read more

Nested Benders Decomposition for Large-Scale Multi-Follower Bilevel Optimization

We propose a scalable nested Benders decomposition (BD) framework for single-leader, multi-follower bilevel optimization problems. The proposed framework is applicable to bilevel optimization problems in which each follower solves a linear program and is particularly well suited for instances involving a large number of followers. By identifying the upper-level decisions as complicating variables, the method … Read more

Stochastic convergence of parallel asynchronous adaptive first-order methods

A new class of asynchronous adaptive first-order optimization methods is introduced, comprising asynchronous variants of several popular algorithms. Versions of these methods using momentum and/or inexact normalization are also considered. The convergence of methods in the class on non-convex functions is analyzed in a fully stochastic setting, and is shown to be (up to logarithmic … Read more