A non-anticipative learning-optimization framework for solving multi-stage stochastic programs

被引:1
|
作者
Yilmaz, Dogacan [1 ]
Buyuktahtakin, I. Esra [2 ]
机构
[1] New Jersey Inst Technol, Dept Mech & Ind Engn, 200 Cent Ave, Newark, NJ 07114 USA
[2] Virginia Tech, Grad Dept Ind & Syst Engn, 250 Perry St, Blacksburg, VA 24061 USA
基金
美国国家科学基金会;
关键词
Multi-stage stochastic programming; Machine learning; Lot-sizing; Knapsack; Sequential decision-making; Combinatorial optimization; Relax-and-fix heuristics; LOT-SIZING PROBLEMS; ALGORITHMS; FIX; FORMULATIONS; UNCERTAINTY; NETWORKS; WIND;
D O I
10.1007/s10479-024-06100-7
中图分类号
C93 [管理学]; O22 [运筹学];
学科分类号
070105 ; 12 ; 1201 ; 1202 ; 120202 ;
摘要
We present a non-anticipative learning- and scenario-based prediction-optimization (ScenPredOpt) framework that combines deep learning, heuristics, and mathematical solvers for solving combinatorial problems under uncertainty. Specifically, we transform neural machine translation frameworks to predict the optimal solutions of scenario-based multi-stage stochastic programs. The learning models are trained efficiently using the input and solution data of the multi-stage single-scenario deterministic problems. Then our ScenPredOpt framework creates a mapping from the inputs used in training into an output of predictions that are close to optimal solutions. We present a Non-anticipative Encoder-Decoder with Attention (NEDA) approach, which ensures the non-anticipativity property of multi-stage stochastic programs and, thus, time consistency by calibrating the learned information based on the problem's scenario tree and adjusting the hidden states of the neural network. In our ScenPredOpt framework, the percent predicted variables used for the solution are iteratively reduced through a relaxation of the problem to eliminate infeasibility. Then, a linear relaxation-based heuristic is performed to further reduce the solution time. Finally, a mathematical solver is used to generate the complete solution. We present the results on two NP-Hard sequential optimization problems under uncertainty: stochastic multi-item capacitated lot-sizing and stochastic multistage multidimensional knapsack. The results show that the solution time can be reduced by a factor of 599 with an optimality gap of only 0.08%. We compare the results of the ScenPredOpt framework with cutting-edge exact and heuristic solution algorithms for the problems studied and find that our framework is more effective. Additionally, the computational results demonstrate that ScenPredOpt can solve instances with a larger number of items and scenarios than the trained ones. Our non-anticipative learning-optimization approach can be beneficial for stochastic programming problems involving binary variables that are solved repeatedly with various types of dimensions and similar decisions at each period.
引用
收藏
页数:41
相关论文
共 50 条
  • [41] Multi-stage Stochastic Optimization for a PV-Storage Hybrid Unit in a Household
    Hafiz, Faeza
    de Queiroz, Anderson Rodrigo
    Husain, Iqbal
    2017 IEEE INDUSTRY APPLICATIONS SOCIETY ANNUAL MEETING, 2017,
  • [42] A Deep Reinforcement Learning Framework for Multi-Stage Optimized Object Detection
    Siamak, Sobhan
    Mansoori, Eghbal
    2022 10TH RSI INTERNATIONAL CONFERENCE ON ROBOTICS AND MECHATRONICS (ICROM), 2022, : 132 - 138
  • [43] A multi-stage stochastic optimization model for energy systems planning and risk management
    Cano, Emilio L.
    Moguerza, Javier M.
    Alonso-Ayuso, Antonio
    ENERGY AND BUILDINGS, 2016, 110 : 49 - 56
  • [44] A Barrier Lagrangian Dual Method for Multi-stage Stochastic Convex Semidefinite Optimization
    Gafour, Asma
    Alzalg, Baha
    VIETNAM JOURNAL OF MATHEMATICS, 2024,
  • [45] Model-agnostic multi-stage loss optimization meta learning
    Xiao Yao
    Jianlong Zhu
    Guanying Huo
    Ning Xu
    Xiaofeng Liu
    Ce Zhang
    International Journal of Machine Learning and Cybernetics, 2021, 12 : 2349 - 2363
  • [46] Multi-Stage Optimization of Deep Learning Model to Detect Thoracic Complications
    Ratul, Rizwanul Hoque
    Husain, Farah Anjum
    Purnata, Tajmim Hossain
    Pomil, Rifat Alam
    Khandoker, Shaima
    Parvez, Mohammad Zavid
    2021 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN, AND CYBERNETICS (SMC), 2021, : 3000 - 3005
  • [47] Model-agnostic multi-stage loss optimization meta learning
    Yao, Xiao
    Zhu, Jianlong
    Huo, Guanying
    Xu, Ning
    Liu, Xiaofeng
    Zhang, Ce
    INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS, 2021, 12 (08) : 2349 - 2363
  • [48] Optimization of dynamic multi-objective non-identical parallel machine scheduling with multi-stage reinforcement learning
    Guo, Liangxun
    Zhuang, Zilong
    Huang, Zizhao
    Qin, Wei
    2020 IEEE 16TH INTERNATIONAL CONFERENCE ON AUTOMATION SCIENCE AND ENGINEERING (CASE), 2020, : 1215 - 1219
  • [49] A Branch and Cut Framework for Multi-Stage Stochastic Programming Problems Under Endogenous Uncertainty
    Colvin, Matthew
    Maravelias, Christos T.
    10TH INTERNATIONAL SYMPOSIUM ON PROCESS SYSTEMS ENGINEERING, 2009, 27 : 255 - 260
  • [50] A multi-stage competitive swarm optimization algorithm for solving large-scale multi-objective optimization problems
    Shang, Qingxia
    Tan, Minzhong
    Hu, Rong
    Huang, Yuxiao
    Qian, Bin
    Feng, Liang
    EXPERT SYSTEMS WITH APPLICATIONS, 2025, 260