Abstract

In this dissertation, I design and analyze tractable algorithms for sequential decision-making problems under uncertainty, with various applications in revenue management. More specifically, I focus on innovative business strategies that have emerged in recent years, with applications ranging from reusable product sales, opaque selling, and maritime shipping. For each application, I identify the unique operational requirements, highlight the complexity of the underlying decision problems, and develop tractable solution approaches. I also provide both theoretical analysis and numerical evaluation to evaluate the performance of the proposed policies. In the first chapter, I study the joint inventory and online assortment problem, wherein a decision maker (DM) must first select initial inventory levels for a collection of available products or resources, and then offer personalized assortments to customers who arrive over a finite selling horizon, and who make purchasing decisions according to a multinomial logit choice model. The goal across both sets of decisions is to maximize the expected revenue earned by the end of the selling horizon. We are the first to consider this joint optimization framework when the resources are reusable. That is, upon purchase or rental, each unit is consumed for a random duration, after which it returns to the DM for future use. Our cornerstone result when reusability is modeled in its classic form, is a constant factor approximation scheme when the usage duration distributions satisfy the increasing failure rate (IFR) property. In a nutshell, our approach exploits notions of submodularity within a fluid approximation of the original problem. This fluid problem approximates the IFR-based usage durations with appropriately defined geometric random variables. To show that this approximate approach is indeed valid requires establishing a novel link between the CDFs of geometric and IFR-distributed random variables, which may find broader applications beyond those considered in this paper. Next, we consider our joint optimization problem under an augmented version of basic reusability, wherein consumed resources can return to the DM as transformed versions of their original selves. The intent of this novel modeling feature is to capture reusability settings where the identity of a product can possibly change due to its consumption (a product purchased online and returned to the seller may become damaged during the try-on process) or through the very nature of a return (a bike rented at one dock may be returned to a different one). In this so-called network reusability setting, we propose a novel inventory refinement process that iteratively adjusts inventory decisions based on feedback from the online assortment stage. We ultimately establish a strong performance bound for our overall approach, which is network dependent. Through numerical experiments, we show that our approximation strategies perform nearly optimally across a wide range of reusability scenarios, demonstrating the robustness and practicality of our approach. In the second chapter, I study the algorithmic approach to mitigate the overbooking and no-show behavior in the maritime industry. In the airline industry, the practice of overbooking has been a celebrated operational tool that has led to revenue gains exceeding hundreds of millions of dollars when implemented correctly. By contrast, in the container shipping industry, the story of overbooking is filled with tales of chronic mistrust between shippers and carriers. Specifically, loose and unenforceable contracting practices have led to a failed market where shippers constantly renege on their agreement to produce containers as promised, and as a result, carriers overbook too frequently in an effort to hedge against this no-show behavior. The cost of such behaviors has been estimated to be in the range of \$30-40 billion annually, which highlights the glaring need for a remedy to this issue. In this paper, we propose and study a deposit-based booking system that draws inspiration from current practices that have been shown to be successful in mitigating no-show behavior and overbooking in the container shipping industry. Specifically, we consider a reservation system where inquiring shippers book cargo space using a customized deposit. The carrier, upon accepting the shipper’s booking request, matches the shipper’s deposit with a deposit of their own of equal size. If either party reneges on the agreement, the defaulting party loses their deposit to the more trustworthy party. However, if both parties uphold their side of the deal, the deposits are returned in full to both sides. Under this booking mechanism, we study the carrier’s sequential online booking problem, which gives rise to a new class of revenue management problems with overbooking and no-show behavior that share only superficial commonalities with existing frameworks. In the third chapter, I study the operational decisions involved in opaque selling—a practically motivated mechanism employed by platforms such as Priceline.com. Under this mechanism, customers are offered an opaque option, which provides them with one item from a predefined bundle without revealing which specific item they will receive at the time of purchase. In exchange for this uncertainty, customers receive a discounted price, allowing the retailer to gain greater flexibility in managing inventory levels. This work contributes to the growing literature on choice-based online matching by extending the classical model to incorporate this additional decision lever. Customers who choose the opaque product delegate the final allocation decision to the platform, introducing consumer-side uncertainty that must be accounted for in the underlying choice model. This trade-off between allocation flexibility and customer uncertainty presents new challenges in dynamic assortment and pricing. I rigorously formulate the problem and propose tractable algorithms that address these challenges, providing both theoretical performance guarantees and numerical evidence of the effectiveness and robustness of the proposed policies under realistic settings.

Committee Chair

Jacob Feldman

Committee Members

Deniz Akturk; Heng Zhang; Lingxiu Dong; Naveed Chehrazi

Degree

Doctor of Philosophy (PhD)

Author's Department

Supply Chain, Operations, and Technology

Author's School

Olin Business School

Document Type

Dissertation

Date of Award

5-22-2025

Language

English (en)

Download

Share

COinS