arXiv subjects

Beste Basciftci

Publications and source records attributed to Beste Basciftci.

15 recordsLinked to original sources

arXiv2026

Learning Optimization Proxies for Sequential Contextual Stochastic Programs: An Order Fulfillment Application

Sequential contextual stochastic programs model real-time decision systems in which each time epoch commits to an action under uncertainty whose consequences propagate into future decisions. In many practical contexts, these programs require obtaining solutions rapidly as new information becomes available. These problems can be represented through scenario approximations to be solved by off-the-shelf optimization solvers, which achieve high decision quality offline but typically run in seconds to minutes per instance, falling short of the sub-second responses that peak periods of planning require. This paper develops a learning-based optimization proxy: a scenario-embedded neural network trained offline on solver-generated labels, paired online with a decoder that enforces feasibility, replacing the per-epoch solve with a single forward pass. The framework is specialized to omnichannel order fulfillment, where each arriving order requires a sub-second assignment of products to distribution centers and carrier services under stochastic delivery times and future demand. A two-stage contextual stochastic program is introduced to formulate this problem, and its contextual sample average approximation (C-SAA) supplies the offline labels, while a composite training loss combines label imitation, a constraint-violation penalty, and self-supervised cost alignment. In a calibrated simulator built from JD.com transactional records, a detailed computational study is provided. The proxy reduces decision latency by roughly 2800x relative to the online finite-sample C-SAA reference and improves over it by 3.3% in realized fulfillment cost. Relative to established fulfillment policies, the proxy lowers total realized cost by at least 10.7% and roughly halves the late-delivery rate.

math.OC↗

arXiv2026

Transit Network Design with Two-Level Demand Uncertainties: A Machine Learning and Contextual Stochastic Optimization Framework

Transit Network Design is a well-studied problem in the field of transportation, typically addressed by solving optimization models under fixed demand assumptions. Considering the limitations of these assumptions, this paper proposes a new framework, namely the Two-Level Rider Choice Transit Network Design (2LRC-TND), that leverages machine learning and contextual stochastic optimization (CSO) through constraint programming (CP) to incorporate two layers of demand uncertainties into the network design process. The first level identifies travelers who rely on public transit (core demand), while the second level captures the conditional adoption behavior of those who do not (latent demand), based on the availability and quality of transit services. To capture these two types of uncertainties, 2LRC-TND relies on two travel mode choice models, that use multiple machine learning models. To design a network, 2LRC-TND integrates the resulting choice models into a CSO that is solved using a CP-SAT solver. 2LRC-TND is evaluated through a case study involving over 6,600 travel arcs and more than 38,000 trips in the Atlanta metropolitan area. The computational results demonstrate the effectiveness of the 2LRC-TND in designing transit networks that account for demand uncertainties and contextual information, offering a more realistic alternative to fixed-demand models.

cs.LG↗

arXiv2025

Hydrogen Network Expansion Planning considering the Chicken-and-egg Dilemma and Market Uncertainty

Green hydrogen is thought to be a game changer for reaching sustainability targets. However, the transition to a green hydrogen economy faces a critical challenge known as the `chicken-and-egg dilemma', wherein establishing a hydrogen supply network relies on demand, while demand only grows with reliable supply. In addition, as the hydrogen market is in the early stage, predicting demand distributions is challenging due to lack of data availability. This paper addresses these complex issues through a risk-averse framework with the introduction of a distributionally robust hydrogen network expansion planning problem under decision-dependent demand ambiguity. The problem optimizes location and production capacity decisions of the suppliers considering the moments of the stochastic hydrogen demand as a function of these investment decisions. To obtain tractable representations of this problem, we derive two different reformulations that consider continuous and discrete hydrogen demand support sets under different forms of decision dependencies. To efficiently solve the reformulations, we develop a tailored algorithm based on the column-and-constraint generation approach, and enhance the computational performance through solving the master problems to a relative optimality gap, decomposing the subproblems, and integrating pre-generated columns and constraints. To validate the effectiveness of our approach, we investigate a real case study leveraging data from the "Hydrogen Energy Applications in Valley Environments for Northern Netherlands (HEAVENN)" project. The results reveal that considering the chicken-and-egg dilemma under uncertain hydrogen market conditions leads to earlier and more diverse investments, providing critical insights for policymakers based on the degree of decision dependency.

math.OC↗

arXiv2024

Dual Sourcing of Green Hydrogen: Balancing Local Production with Stochastic Capacity and Import with Random Yield

Green hydrogen is a critical component for achieving the European Union's 2050 net-zero emissions goal. However, ensuring a reliable and stable supply is challenging, particularly when local production of green hydrogen is subject to high variability due to fluctuating renewable energy output. Although importing from regions with stable renewable resources may offer greater reliability, they introduce longer lead times and potential energy losses during transportation and conversion. To address this issue, we develop optimal dual sourcing policies for green hydrogen through modeling and solving a Markov Decision Process that integrates general lead times, stochastic local supply capacity, and random yield from import. Alongside optimal dual sourcing policies, we propose heuristic policies that offer both flexibility and stability, enabling practical implementation while achieving near-optimal performance. We test our framework on a case study for the Netherlands and obtain the following insights: (i) based on the results across different countries and cost settings, our dual sourcing model demonstrates an average cost benefit of 8 percent compared to models that ignore stochastic supply capacity and random yield, (ii) the proposed heuristic policies can perform comparably to optimal policies under varying country-specific conditions and cost settings, offering insights for shaping hydrogen trade agreements between importing and exporting countries, (iii) sensitivity analyses on production and storage costs, as well as variability in supply capacity, demand, and random yield, reveal the conditions needed to achieve specific local production levels. These results thereby support feasibility of the Netherlands' climate scenarios and ambitions while guiding green hydrogen investment planning.

math.OC↗

arXiv2024

Distributionally Robust Optimization with Multimodal Decision-Dependent Ambiguity Sets

We consider a two-stage distributionally robust optimization (DRO) model with multimodal uncertainty, where both the mode probabilities and uncertainty distributions could be affected by the first-stage decisions. To address this setting, we propose a generic framework by introducing a $\phi$-divergence based ambiguity set to characterize the decision-dependent mode probabilities and further consider both moment-based and Wasserstein distance-based ambiguity sets to characterize the uncertainty distribution under each mode. We identify two special $\phi$-divergence examples (variation distance and $\chi^2$-distance) and provide specific forms of decision dependence relationships under which we can derive tractable reformulations. Furthermore, we investigate the benefits of considering multimodality in a DRO model compared to a single-modal counterpart through an analytical analysis. Additionally, we develop a separation-based decomposition algorithm to solve the resulting multimodal decision-dependent DRO models with finite convergence and optimality guarantee under certain settings. We provide a detailed computational study over two example problem settings, the facility location problem and shipment planning problem with pricing, to illustrate our results, which demonstrate that omission of multimodality or decision-dependent uncertainties within DRO frameworks result in inadequately performing solutions with worse in-sample and out-of-sample performances under various settings. We further demonstrate the speed-ups obtained by the solution algorithm against the off-the-shelf solver over various instances.

math.OC↗

arXiv2024

Adaptive Multistage Stochastic Programming

Multistage stochastic programming is a powerful tool allowing decision-makers to revise their decisions at each stage based on the realized uncertainty. However, in practice, organizations are not able to be fully flexible, as decisions cannot be revised too frequently due to their high organizational impact. Consequently, decision commitment becomes crucial to ensure that initially made decisions remain unchanged for a certain period. This paper introduces adaptive multistage stochastic programming, a new optimization paradigm that strikes an optimal balance between decision flexibility and commitment by determining the best stages to revise decisions depending on the allowed level of flexibility. We introduce a novel mathematical formulation and theoretical properties eliminating certain constraint sets. Furthermore, we develop a decomposition method that effectively handles mixed-integer adaptive multistage programs by adapting the integer L-shaped method and Benders decomposition. Computational experiments on stochastic lot-sizing and generation expansion planning problems show substantial advantages attained through optimal selections of revision times when flexibility is limited, while demonstrating computational efficiency of the proposed properties and solution methodology. Optimizing revision times in a less flexible case can outperform arbitrary selection in a more flexible case. By adhering to these optimal revision times, organizations can achieve performance levels comparable to fully flexible settings.

math.OC↗

arXiv2023

Optimizing Strategic and Operational Decisions of Car Sharing Systems under Demand Uncertainty and Substitution

Optimizing car sharing systems under demand uncertainty is an emerging problem for ensuring profitable and sustainable operations of these services while taking into account quality of service concerns. With the increasing adoption of electric vehicles and environmental awareness, this problem requires consideration of a mix fleet of vehicles with gasoline-powered and electric, complicating the strategic and operational planning as the demand of each vehicle type are observed. To address this problem, we propose a two-stage stochastic mixed-integer program leveraging spatial-temporal networks that capture the strategic and operational decisions of these systems over a multi-period planning horizon. We optimize the location decisions of regions to serve with purchasing decisions of the vehicles in the first-stage problem under budget and carbon emission considerations in designing the fleet, while considering parking capacities, satisfying one-way and round-trip car rental requests, and relocating cars between open regions under each demand realization in the second-stage problem. We then introduce demand substitution to this problem by extending and generalizing the multi-commodity formulation, and allowing satisfaction of customer demand of each vehicle type with its alternatives. We further prove that the corresponding second-stage problem has a totally unimodular constraint matrix. By benefiting from this result, as our solution approach, we provide a branch-and-cut based decomposition algorithm with enhancements. We present an extensive computational study highlighting the value of the proposed models from different perspectives and demonstrating the performance of the proposed solution algorithm with significant speedups. Introducing substitution to the car sharing operations leads to higher quality of service and flexibility in operations with lower costs under various settings.

math.OC↗

arXiv2023

Path-Based Formulations for the Design of On-demand Multimodal Transit Systems with Adoption Awareness

This paper reconsiders the ODMTS Design with Adoptions problem (ODMTS-DA) to capture the latent demand in on-demand multimodal transit systems. The ODMTS-DA is a bilevel optimization problem, for which Basciftci and Van Hentenryck (2022) proposed an exact combinatorial Benders decomposition. Unfortunately, their proposed algorithm only finds high-quality solutions for medium-sized cities and is not practical for large metropolitan areas. The main contribution of this paper is to propose a new path-based optimization model, called P-Path, to address these computational difficulties. The key idea underlying P-Path is to enumerate two specific sets of paths which capture the essence of the choice model associated with the adoption behavior of riders. With the help of these path sets, the ODMTS-DA can be formulated as a single-level mixed-integer programming model. In addition, the paper presents preprocessing techniques that can reduce the size of the model significantly. P-Path is evaluated on two comprehensive case studies: the mid-size transit system of the Ann Arbor Ypsilanti region in Michigan (which was studied by Basciftci and Van Hentenryck (2022)) and the large-scale transit system for the city of Atlanta. The experimental results show that P-Path solves the Michigan ODMTS-DA instances in a few minutes, bringing more than two orders of magnitude improvements compared to the existing approach. For Atlanta, the results show that P-Path can solve large-scale ODMTS-DA instances (about 17 millions of variables and 37 millions of constraints) optimally in a few hours or in a few days. These results show the tremendous computational benefits of P-Path which provides a scalable approach to the design of on-demand multimodal transit systems with latent demand.

math.OC↗

arXiv2022

Bilevel Optimization and Heuristic Algorithms for Integrating Latent Demand into the Design of Large-Scale Transit Systems

Capturing latent demand has a pivotal role in designing transit services as omitting these riders can lead to poor quality of service and/or additional costs. This paper explores this topic in the design of transit networks by considering the perspectives of both the transit agencies and riders. The paper presents a generic bilevel optimization model, namely the Transit Networks Design with Adoptions (TN-DA), that considers the network design decisions in the leader problem, and routing of the riders in the follower problem under the given network design, while allowing a black-box choice function for representing the adoption behavior of latent demand. The paper then identifies structural properties of the optimal solution of the TN-DA problem, which are desirable for transit agencies for capturing adoption behavior of the riders. The paper further provides guideline metrics for the transit agencies based on these desired adoption properties. Due to the computational complexity of this bilevel problem, the paper proposes five efficient heuristic algorithms to solve large-scale instances, which leverage an iterative procedure by solving a simpler version of the TN-DA problem and integrating the evaluation of rider choices. These algorithms either satisfy the desired properties of the optimal solution or provide fast approximations. The paper presents extensive large-scale case studies on two different transit systems by utilizing real datasets: (i) On-demand Multimodal Transit Systems (ODMTS) and (ii) Scooters-Connected Transit Systems (SCTS). The results demonstrate that the heuristic algorithms can find high-quality solutions much faster than exact approaches over various instances, while satisfying key adoption properties of the optimal solutions.

math.OC↗

arXiv2022

Capturing Travel Mode Adoption in Designing On-demand Multimodal Transit Systems

This paper studies how to integrate rider mode preferences into the design of On-Demand Multimodal Transit Systems (ODMTS). It is motivated by a common worry in transit agencies that an ODMTS may be poorly designed if the latent demand, i.e., new riders adopting the system, is not captured. The paper proposes a bilevel optimization model to address this challenge, in which the leader problem determines the ODMTS design, and the follower problems identify the most cost efficient and convenient route for riders under the chosen design. The leader model contains a choice model for every potential rider that determines whether the rider adopts the ODMTS given her proposed route. To solve the bilevel optimization model, the paper proposes an exact decomposition method that includes Benders optimal cuts and nogood cuts to ensure the consistency of the rider choices in the leader and follower problems. Moreover, to improve computational efficiency, the paper proposes upper bounds on trip durations for the follower problems and valid inequalities that strenghten the nogood cuts. The proposed method is validated using an extensive computational study on a real data set from AAATA, the transit agency for the broader Ann Arbor and Ypsilanti region in Michigan. The study considers the impact of a number of factors, including the price of on-demand shuttles, the number of hubs, and accessibility criteria. The designed ODMTS feature high adoption rates and significantly shorter trip durations compared to the existing transit system and highlight the benefits in accessibility for low-income riders. Finally, the computational study demonstrates the efficiency of the decomposition method for the case study and the benefits of computational enhancements that improve the baseline method by several orders of magnitude.

math.OC↗

arXiv2022

Resource Distribution Under Spatiotemporal Uncertainty of Disease Spread: Stochastic versus Robust Approaches

We consider the problem of optimizing locations of distribution centers (DCs) and plans for distributing resources such as test kits and vaccines, under spatiotemporal uncertainties of disease spread and demand for the resources. We aim to balance the operational cost (including costs of deploying facilities, shipping, and storage) and quality of service (reflected by demand coverage), while ensuring equity and fairness of resource distribution across multiple populations. We compare a sample-based stochastic programming (SP) approach with a distributionally robust optimization (DRO) approach using a moment-based ambiguity set. Numerical studies are conducted on instances of distributing COVID-19 vaccines in the United States and test kits, to compare SP and DRO models with a deterministic formulation using estimated demand and with the current resource distribution plans implemented in the US. We demonstrate the results over distinct phases of the pandemic to estimate the cost and speed of resource distribution depending on scale and coverage, and show the ``demand-driven'' properties of the SP and DRO solutions. Our results further indicate that if the worst-case unmet demand is prioritized, then the DRO approach is preferred despite of its higher overall cost. Nevertheless, the SP approach can provide an intermediate plan under budgetary restrictions without significant compromises in demand coverage.

math.OC↗

arXiv2022

A Joint Chance-Constrained Stochastic Programming Approach for the Integrated Predictive Maintenance and Operations Scheduling Problem in Power Systems

Maintenance planning plays a key role in power system operations under uncertainty by helping system operators ensure a reliable and secure power grid. This paper studies a short-term condition-based integrated maintenance planning with operations scheduling problem while considering the unexpected failure possibilities of generators as well as transmission lines. We formulate this problem as a two-stage stochastic mixed-integer program with failure scenarios sampled from the sensor-driven remaining lifetime distributions of the individual system elements whereas a joint chance-constraint consisting of Poisson Binomial random variables is introduced to account for failure risks. Because of its intractability, we develop a cutting-plane method to obtain an exact reformulation of the joint chance-constraint by proposing a separation subroutine and deriving stronger cuts as part of this procedure. To solve large-scale instances, we derive a second-order cone programming based safe approximation of this constraint. Furthermore, we propose a decomposition-based algorithm implemented in parallel fashion for solving the resulting stochastic program, by exploiting the features of the integer L-shaped method and the special structure of the maintenance and operations scheduling problem to derive stronger optimality cuts. We further present preprocessing steps over transmission line flow constraints to identify redundancies. To illustrate the computational performance and efficiency of our algorithm compared to more conventional maintenance approaches, we design a computational study focusing on a weekly plan with daily maintenance and hourly operational decisions involving detailed unit commitment subproblems. Our computational results on various IEEE instances demonstrate the computational efficiency of the proposed approach with reliable and cost-effective maintenance and operational schedules.

math.OC↗

arXiv2020

Distributionally Robust Facility Location Problem under Decision-dependent Stochastic Demand

Facility location decisions significantly impact customer behavior and consequently the resulting demand in a wide range of businesses. Furthermore, sequentially realized uncertain demand enforces strategically determining locations under partial information. To address these issues, we study a facility location problem where the distribution of customer demand is dependent on location decisions. We represent moment information of stochastic demand as a piecewise linear function of facility-location decisions. Then, we propose a decision-dependent distributionally robust optimization model, and develop its exact mixed-integer linear programming reformulation. We further derive valid inequalities to strengthen the formulation. We conduct an extensive computational study, in which we compare our model with the existing (decision-independent) stochastic and robust models. Our results demonstrate superior performance of the proposed approach with remarkable improvement in profit and quality of service by extensively testing problem characteristics, in addition to computational speed-ups due to the formulation enhancements. These results draw attention to the need of considering the impact of location decisions on customer demand within this strategic-level planning problem.

math.OC↗

arXiv2019

Bilevel Optimization for On-Demand Multimodal Transit Systems

This study explores the design of an On-Demand Multimodal Transit System (ODMTS) that includes segmented mode switching models that decide whether potential riders adopt the new ODMTS or stay with their personal vehicles. It is motivated by the desire of transit agencies to design their network by taking into account both existing and latent demand, as quality of service improves. The paper presents a bilevel optimization where the leader problem designs the network and each rider has a follower problem to decide her best route through the ODMTS. The bilevel model is solved by a decomposition algorithm that combines traditional Benders cuts with combinatorial cuts to ensure the consistency of mode choices by the leader and follower problems. The approach is evaluated on a case study using historical data from Ann Arbor, Michigan, and a user choice model based on the income levels of the potential transit riders.

math.OC↗

arXiv2019

Adaptive Two-stage Stochastic Programming with an Analysis on Capacity Expansion Planning Problem

Multi-stage stochastic programming is a well-established framework for sequential decision making under uncertainty by seeking policies that are fully adapted to the uncertainty. Often such flexible policies are not desirable, and the decision maker may need to commit to a set of actions for a number of planning periods. Two-stage stochastic programming might be better suited to such settings, where the decisions for all periods are made here-and-now and do not adapt to the uncertainty realized. In this paper, we propose a novel alternative approach, where the stages are not predetermined but part of the optimization problem. Each component of the decision policy has an associated revision point, a period prior to which the decision is predetermined and after which it is revised to adjust to the uncertainty realized thus far. We motivate this setting using the multi-period newsvendor problem by deriving an optimal adaptive policy. We label the proposed approach as adaptive two-stage stochastic programming and provide a generic mixed-integer programming formulation for finite stochastic processes. We show that adaptive two-stage stochastic programming is NP-hard in general. Next, we derive bounds on the value of adaptive two-stage programming in comparison to the two-stage and multi-stage approaches for a specific problem structure inspired by the capacity expansion planning problem. Since directly solving the mixed-integer linear program associated with the adaptive two-stage approach might be very costly for large instances, we propose several heuristic solution algorithms based on the bound analysis. We provide approximation guarantees for these heuristics. Finally, we present an extensive computational study on an electricity generation capacity expansion planning problem and demonstrate the computational and practical impacts of the proposed approach from various perspectives.

math.OC↗