Article
Management
Ilhem Slama, Oussama Ben-Ammar, Simon Thevenin, Alexandre Dolgui, Faouzi Masmoudi
Summary: This paper addresses the capacitated disassembly lot-sizing problems under uncertain refurbishing durations. By modeling the problem as a two-stage stochastic Mixed-Integer Linear Program (MILP) and utilizing a reformulation of the inventory constraint and Monte-Carlo sampling, scalability issues are effectively alleviated. Experimental results demonstrate the effectiveness of the proposed models and the convergence of the resulting Sample Average Approximation (SAA) estimator.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2022)
Article
Management
Kerem Akartunali, Stephane Dauzere-Peres
Summary: This paper proposes and studies a novel approach to modeling uncertainty in demand for the single-item dynamic lot sizing problem. The uncertainty is not related to the demand quantity, but rather to the demand timing. The paper introduces a modeling technique that naturally correlates uncertain demands in different periods, and presents dynamic programming algorithms for the general case and special cases with stochastic demand timing. Additionally, the paper proves that the most general case, where the backlog cost depends both on the time period and the stochastic demand, is NP-hard.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2022)
Article
Computer Science, Interdisciplinary Applications
Huseyin Tunc
Summary: The study addresses the capacitated stochastic lot-sizing problem under alpha service level constraints. A static uncertainty strategy and dynamic cut generation approach are proposed to solve the problem, and their superior computational performance is demonstrated through extensive numerical study.
COMPUTERS & OPERATIONS RESEARCH
(2021)
Article
Computer Science, Interdisciplinary Applications
Ilhem Slama, Oussama Ben-Ammar, Alexandre Dolgui, Faouzi Masmoudi
Summary: This research focuses on proposing optimization methods for the stochastic multi-period disassembly lot-sizing problem, studying a specific end-of-life product and a two-level disassembly system. Three approaches were developed to solve the problem, including a two-stage mixed-integer linear programming model, a sample average approximation approach based on Monte Carlo simulation, and an optimization approach based on Monte Carlo simulation and a genetic algorithm. Experimental results show the effectiveness of the proposed models in supporting decision-making for replenishment and disassembly plans.
COMPUTERS & INDUSTRIAL ENGINEERING
(2021)
Article
Computer Science, Interdisciplinary Applications
Ming Liu, Hao Tang, Feng Chu, Feifeng Zheng, Chengbin Chu
Summary: This study addresses a multi-product joint lot-sizing and pricing problem with backlogging and limited production capacity, aiming to maximize the total company profit over a finite planning horizon. The study provides a mixed integer nonlinear programming (MINLP) formulation, proposes a model based heuristic approach, and develops a genetic algorithm with a new progressive repair strategy to efficiently solve instances of different sizes.
COMPUTERS & INDUSTRIAL ENGINEERING
(2022)
Article
Engineering, Industrial
Rafael A. Campos, Aakil M. Caunhye, Douglas Alem, Pedro Munari
Summary: This article presents a novel fragility-based approach for a production lot-sizing problem in the veterinary pharmaceutical industry, which addresses some issues in traditional robust optimization methods and achieves greater model stability and cost reduction.
INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH
(2023)
Article
Engineering, Industrial
Eduardo Curcio, Vinicius L. de Lima, Flavio K. Miyazawa, Elsa Silva, Pedro Amorim
Summary: Interest in integrating lot-sizing and cutting stock problems has been increasing, but few studies have considered the importance of uncertainty in optimizing these integrated decisions. This work addresses this gap by incorporating demand uncertainty through stochastic programming and robust optimization approaches, and proposes computational experiments to test the efficiency of these methods.
INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH
(2023)
Article
Computer Science, Interdisciplinary Applications
Simon Thevenin, Yossiri Adulyasak, Jean-Francois Cordeau
Summary: This study delves into lot sizing with component substitution under demand uncertainty, proposing a stochastic programming formulation using stochastic dual dynamic programming (SDDP) to optimize the problem. Computational experiments show that the heuristic version of SDDP outperforms other methods in terms of efficiency and dynamic planning capabilities.
INFORMS JOURNAL ON COMPUTING
(2022)
Article
Management
Narges Sereshti, Yossiri Adulyasak, Raf Jans
Summary: Dealing with demand uncertainty in multi-item lot sizing problems is challenging, but extended stochastic formulations and mathematical models for different types of service levels have been proposed to address this issue. Computational experiments have been conducted to analyze the impact of aggregate service levels and demonstrate the value of the proposed formulations.
OMEGA-INTERNATIONAL JOURNAL OF MANAGEMENT SCIENCE
(2021)
Article
Engineering, Industrial
Antoine Perraudat, Stephane Dauzere-Peres, Scott Jennings Mason
Summary: This study proposes a novel approach to help industrial firms reduce production costs by including electricity pricing in a multi-product lot-sizing problem. Critical parameters for manufacturers in curtailment requests include setup cost ratio, capacity utilization rate, number of products, and timing.
INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH
(2022)
Article
Management
Viktor Bindewald, Fabian Dunke, Stefan Nickel
Summary: This paper studies a new variant of the lot sizing problem with uncertain demand and planning horizon. It proposes a rolling horizon procedure to dissolve the multistage problem into a series of snapshot problems under uncertainty. Different approaches such as online optimization, stochastic programming, and robust optimization are evaluated to model and solve the snapshot problems based on available data and risk disposition. The impact of the selected methodology on the overall solution quality is evaluated using a methodology-agnostic framework for multistage decision-making under uncertainty. Computational results on lot sizing within a rolling horizon are provided regarding different types of uncertainty, solution approaches, and the value of available information about upcoming demands.
INTERNATIONAL TRANSACTIONS IN OPERATIONAL RESEARCH
(2023)
Article
Management
Oeykue Naz Attila, Agostinho Agra, Kerem Akartunah, Ashwin Arulselvan
Summary: This paper investigates a lot-sizing problem with parameter uncertainties on demands and returns, focusing on remanufacturing. A min-max decomposition approach and two novel extended reformulations are proposed, showing superior performance in comparison with the standard formulation. The impact of problem parameters on computational performance is also discussed.
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH
(2021)
Article
Computer Science, Interdisciplinary Applications
Franco Quezada, Celine Gicquel, Safia Kedad-Sidhoum
Summary: This paper investigates the uncapacitated lot-sizing problem with uncertain demand and costs, proposing a novel extension of the stochastic dual dynamic integer programming algorithm to improve computational efficiency. The proposed algorithm outperforms the existing SDDiP algorithm in providing good-quality solutions for the stochastic uncapacitated lot-sizing problem within computation time limits. Extensive computational experiments show the effectiveness of the proposed algorithm in solving the combinatorial optimization problem.
INFORMS JOURNAL ON COMPUTING
(2022)
Article
Operations Research & Management Science
Sheng- Chen, Delvinia Su
Summary: This study focuses on lot-sizing and scheduling with machine eligibility, sequence-dependent setups, and uncertain demands. It proposes multi-stage stochastic programming and compares deterministic and stochastic models to evaluate cost performance under different demand scenarios.
ANNALS OF OPERATIONS RESEARCH
(2022)
Article
Engineering, Industrial
Marco Caserta, Stefan Voss
Summary: In a recent paper, Bayley et al. (2018) proposed a Benders-based approach for a capacitated lot-sizing problem, but further investigation revealed that the inequalities used in their method were not valid as originally thought.
INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH
(2021)
Article
Computer Science, Software Engineering
Xiao Liu, Fatma Kilinc-Karzan, Simge Kucukyavuz
MATHEMATICAL PROGRAMMING
(2019)
Article
Mathematics, Applied
Hao-Hsiang Wu, Simge Kucukyavuz
SIAM JOURNAL ON OPTIMIZATION
(2019)
Article
Engineering, Industrial
Merve Merakli, Simge Kucukyavuz
Article
Operations Research & Management Science
Hao-Hsiang Wu, Simge Kucukyavuz
OPERATIONS RESEARCH LETTERS
(2020)
Article
Operations Research & Management Science
Qimeng Yu, Simge Kucukyavuz
Summary: The research focuses on minimization problems with bisubmodular objective functions, proposing valid inequalities and providing a complete linear description of the convex hull of the epigraph of a bisubmodular function. Additionally, a cutting plane algorithm for constrained bisubmodular minimization based on the poly-bimatroid inequalities is developed, showing the ability to solve highly non-linear problems in computational experiments.
OPERATIONS RESEARCH LETTERS
(2021)
Article
Computer Science, Software Engineering
Nam Ho-Nguyen, Fatma Kilinc-Karzan, Simge Kucukyavuz, Dabeen Lee
Summary: This study focuses on enhancing existing mixed-integer programming formulations for distributionally robust chance-constrained programs with random right-hand sides over Wasserstein ambiguity sets. By revealing connections and providing improved formulations, the research shows significant reductions in overall solution times, enabling larger problem sizes to be handled effectively in practical applications.
MATHEMATICAL PROGRAMMING
(2022)
Article
Computer Science, Software Engineering
Fatma Kilinc-Karzan, Simge Kucukyavuz, Dabeen Lee
Summary: This paper examines the intersection of mixing sets with common binary variables in joint linear chance-constrained programs featuring random right-hand sides and finite sample space. By establishing a strong connection to submodularity, the paper introduces a new class of valid inequalities and extends existing results concerning the convex hull of these intersections.
MATHEMATICAL PROGRAMMING
(2022)
Article
Operations Research & Management Science
Qimeng Yu, Simge Kucukyavuz
Summary: The paper investigates maximization problems with k-submodular objective functions and introduces valid linear inequalities for solving these problems. By designing a novel optimization algorithm, significant improvements are achieved in solving general k-submodular maximization problems.
DISCRETE OPTIMIZATION
(2021)
Article
Computer Science, Software Engineering
Linchuan Wei, Andres Gomez, Simge Kucukyavuz
Summary: Motivated by modern regression applications, this paper studies the convexification of a class of convex optimization problems with indicator variables and combinatorial constraints on the indicators. Different from previous work, this study simultaneously considers nonlinear non-separable objective, indicator variables, and combinatorial constraints. The results include the convex hull description of the epigraph of the composition of a one-dimensional convex function and an affine function under arbitrary combinatorial constraints, as well as ideal convexifications for different constraint problems.
MATHEMATICAL PROGRAMMING
(2022)
Article
Operations Research & Management Science
Simge Kucukyavuz, Ruiwei Jiang
Summary: Chance-constrained programming is one of the most difficult classes of optimization problems that has attracted the attention of researchers since the 1950s. This survey focuses on limited information scenarios and introduces recent developments in mixed-integer linear formulations of chance-constrained programs and distributionally robust CCP.
EURO JOURNAL ON COMPUTATIONAL OPTIMIZATION
(2022)
Article
Computer Science, Software Engineering
Peijing Liu, Salar Fattahi, Andres Gomez, Simge Kucukyavuz
Summary: In this paper, we study convex quadratic optimization problems with indicator variables and sparse matrix. By using a graphical representation of the support of the matrix, we prove that if the graph is a path, the associated problem can be solved in polynomial time. This allows us to construct a compact extended formulation for the closure of the convex hull of the mixed-integer convex problem's epigraph. Furthermore, we propose a novel decomposition method for a class of sparse strictly diagonally dominant matrices, inspired by inference problems with graphical models, leveraging the efficient algorithm for the path case. Our computational experiments demonstrate the effectiveness of the proposed method compared to state-of-the-art mixed-integer optimization solvers.
MATHEMATICAL PROGRAMMING
(2023)
Article
Computer Science, Software Engineering
Qimeng Yu, Simge Kucukyavuz
Summary: This study investigates the polyhedral convex hull structure of a mixed-integer set in a class of cardinality-constrained concave submodular minimization problems. The proposed three classes of strong valid linear inequalities are effective in obtaining valid inequalities for general non-negative vector and binary vector. The computational experiments demonstrate the effectiveness of the proposed inequalities in a branch-and-cut framework for mean-risk optimization problem.
MATHEMATICAL PROGRAMMING
(2023)
Article
Computer Science, Software Engineering
Linchuan Wei, Alper Atamturk, Andres Gomez, Simge Kucukyavuz
Summary: This paper discusses convex quadratic optimization problems with indicator variables and arbitrary constraints. The authors prove that the convex hull description of the associated mixed-integer set can be transformed into a problem with an extended space consisting of a (n + 1) x (n + 1) positive semidefinite constraint and linear constraints. They also provide a description in the original space using finitely generated conic-quadratic inequalities, which allows for characterizing whether a given inequality is necessary to describe the convex hull.
MATHEMATICAL PROGRAMMING
(2023)
Article
Management
Fatma Kilinc-Karzan, Simge Kucukyavuz, Dabeen Lee, Soroosh Shafieezadeh-Abadeh
Summary: In this study, a general conic mixed-binary set is considered, where each homogeneous conic constraint involves an affine function of continuous variables and an epigraph variable associated with a nonnegative function. A convex hull description of this set is provided, relying on simultaneous characterization of the epigraphs of the nonnegative functions. The result unifies and generalizes existing results in two important directions.
OPERATIONS RESEARCH
(2023)
Article
Operations Research & Management Science
Hao-Hsiang Wu, Simge Kucukyavuz
COMPUTATIONAL OPTIMIZATION AND APPLICATIONS
(2018)
