检索结果-内蒙古大学图书馆

data-driven optimization for drone delivery service planning with online demand

TRANSPORTATION RESEARCH PART E-LOGISTICS AND TRANSPORTATION REVIEW 2025年 198卷

作者： Paula, Aditya Levin, Michael W. Waller, S. Travis Rey, David UNSW Sydney Sch Civil & Environm Engn Sydney NSW 2052 Australia Univ Minnesota Dept Civil Environm & Geoengn Minneapolis MN 55455 USA Tech Univ Dresden Fac Transport & Traff Sci Lighthouse Professorship Transport Modelling & Sim Dresden Germany Univ Cote dAzur SKEMA Business Sch Sophia Antipolis France

In this study, we develop an innovative data-driven optimization approach to solve the drone delivery service planning problem with online demand. Drone-based logistics are expected to improve operations by enhancing flexibility and reducing congestion effects induced by last-mile deliveries. With rising digitalization and urbanization, however, logistics service providers are constantly grappling with the challenge of uncertain real-time demand. This study investigates the problem of planning drone delivery service through an urban air traffic network to fulfill dynamic and stochastic demand. Customer requests-if accepted-generate profit and are serviced by individual drone flights as per request origins, destinations and time windows. We cast this stochastic optimization problem as a Markov decision process. We present a novel data-driven optimization approach which generates predictive prescriptions of parameters of a surrogate optimization formulation. Our solution method consists of synthesizing training data via lookahead simulations to train a supervised machine learning model for predicting relative link priority based on the state of the network. This knowledge is then leveraged to selectively create weighted reserve capacity in the network and via a surrogate objective function that controls the trade-off between reserve capacity and profit maximization to maximize the cumulative profit earned. Using numerical experiments based on benchmarking transportation networks, the resulting data-driven optimization policy is shown to outperform a myopic policy. Sensitivity analyses on learning parameters reveal insights into the design of efficient policies for drone delivery service planning with online demand.

关键词： data-driven optimization Online demand Markov decision process Drone delivery Service planning

来源：评论

学校读者我要写书评

暂无评论

data-driven optimization for production planning with multiple demand features

引用

KYBERNETES 2025年第1期54卷 110-133页

作者： Su, Xiaoli Zeng, Lijun Shao, Bo Lin, Binlong Fuzhou Univ Sch Econ & Management Fuzhou Peoples R China Univ Wisconsin Dept Ind & Syst Engn Madison WI USA

PurposeThe production planning problem with fine-grained information has hardly been considered in practice. The purpose of this study is to investigate the data-driven production planning problem when a manufacturer can observe historical demand data with high-dimensional mixed-frequency features, which provides fine-grained ***/methodology/approachIn this study, a two-step data-driven optimization model is proposed to examine production planning with the exploitation of mixed-frequency demand data is proposed. First, an Unrestricted MIxed data Sampling approach is proposed, which imposes Group LASSO Penalty (GP-U-MIDAS). The use of high frequency of massive demand information is analytically justified to significantly improve the predictive ability without sacrificing goodness-of-fit. Then, integrated with the GP-U-MIDAS approach, the authors develop a multiperiod production planning model with a rolling cycle. The performance is evaluated by forecasting outcomes, production planning decisions, service levels and total *** results show that the key variables influencing market demand can be completely recognized through the GP-U-MIDAS approach;in particular, the selected accuracy of crucial features exceeds 92%. Furthermore, the proposed approach performs well regarding both in-sample fitting and out-of-sample forecasting throughout most of the horizons. Taking the total cost and service level obtained under the actual demand as the benchmark, the mean values of both the service level and total cost differences are reduced. The mean deviations of the service level and total cost are reduced to less than 2.4%. This indicates that when faced with fluctuating demand, the manufacturer can adopt the proposed model to effectively manage total costs and experience an enhanced service ***/valueCompared with previous studies, the authors develop a two-step data-driven optimization model by directly incorporating a potentially

关键词： Big data data-driven optimization Production planning Mixed-frequency data

来源：评论

学校读者我要写书评

暂无评论

Experimental and data-driven optimization of vortex-induced vibration for marine energy harvesting

引用

OCEAN ENGINEERING 2025年 324卷

作者： Xu, Peng Jia, Shanshan Zhang, Yuan Li, Xin Jiang, Changqing Shanghai Jiao Tong Univ State Key Lab Ocean Engn Shanghai 200240 Peoples R China Zhejiang Ocean Univ Sch Naval Architecture & Maritime Zhoushan 316022 Peoples R China Univ Duisburg Essen Inst Sustainable & Autonomous Maritime Syst D-47057 Duisburg Germany

Marine current energy, particularly harnessed through vortex-induced vibration (VIV), offers a cost-effective method for renewable energy generation, especially in low-flow velocity environments. In practical applications, VIV harvester performance is influenced by a free surface, making its dynamic behavior challenging to predict, which differs from the extensively studied VIV phenomenon in an infinite water depth. In this study, we propose an alternative approach to predict the optimal VIV working conditions using an artificial neural network (ANN). Physical model tests with varying parameters were conducted, and an ANN algorithm established a nonlinear mapping between these parameters and the energy conversion ratio. Utilizing the established ANN model, we identify the optimal combinations of working parameters that maximize the conversion ratio of the VIV. Experimental validation of the predicted optimal working conditions is conducted, and our predictions agree favorably with experimental measurements. Our results demonstrate that the neural network model, trained with the experimental data, exhibits high prediction accuracy and strong fitting ability. It offers a practical direction for optimizing VIV working parameters and predicting vibration performance at low flow velocities with a free surface, enhancing the potential for efficient marine current energy harvesting.

关键词： Vortex-induced vibration Artificial neural network Experimental analysis data-driven optimization Marine current energy

来源：评论

学校读者我要写书评

暂无评论

A case-individual data-driven optimization approach for surgery planning

引用

OR SPECTRUM 2025年 1-37页

作者： Rottmann, Janine Guender, Kai Pibernik, Richard Univ Wurzburg D-97070 Wurzburg Germany Zaragoza Logist Ctr Zaragoza 50018 Spain

Operating rooms generate more than 40% of a hospital's total revenue, making effective surgery planning a highly relevant problem for many hospitals. Surgery planning is a challenging task that requires a decision maker to solve multiple interrelated (sub-)problems under uncertainty. An important source of uncertainty are surgery durations-their realizations can vary substantially, depending on a number of factors related to the type of surgery, as well as patients' and surgeons' characteristics. This paper proposes a data-driven approach to solve a complex and practically relevant surgery planning problem using case-individual information contained in electronic health records of past surgeries. Our approach combines machine learning models (Quantile Regression Forests) for estimating durations of individual surgeries and data-driven optimization techniques to assign a set of planned surgeries to operating rooms, and to simultaneously determine their sequence and starting times for a given workday. Numerical analyses based on a real-world dataset of a large university hospital demonstrate the value of our approach. Our approach leads to a cost improvement of 35% compared to a naive approach that reflects the current practice of our case hospital, and we can show that the largest part (approximately 23% points) of this improvement can be attributed to the sequencing and room assignment extensions in our model. We are able to quantify the value of case-individual information, identify the most relevant co-variates ("features"), and demonstrate how better uncertainty estimations affect the individual planning tasks, i.e., room assignment, sequencing, and appointment scheduling.

关键词： Surgery planning data-driven optimization Quantile regression forests Stochastic optimization

来源：评论

学校读者我要写书评

暂无评论

A hybrid approach based on multi-criteria decision making and data-driven optimization in solving portfolio selection problem

引用

OPSEARCH 2025年第1期62卷 1-36页

作者： Doaei, Meysam Dehnad, Kazem Dehnad, Mahdi Islamic Azad Univ Dept Finance Esfarayen Branch Esfarayen Iran Islamic Azad Univ Dept Finance Qom Branch Qom Iran Khatam Univ Dept Finance TEIAS Tehran Iran

In this paper, a two-phase approach based on multi-criteria decision making and multi-objective optimization models is proposed to select portfolio optimally. In the first phase, potential companies for investment are selected initially by considering the criteria extracted from the literature review. In the second phase, a multi-objective optimization model is proposed to optimize the investment in selected companies according to risk and return objectives. In order to deal with uncertainty conditions, a data-driven approach is used, which is one of the newest applied methods in this field. According to the obtained results, it is observed that cash adequacy ratio with score 0.1604 is the most important criterion and operating profit with score 0.004 is the least important one. In the alternative prioritization section, it is concluded that Shraz, Shavan, Shenft and Vanft companies have a high priority for investment. In solving the mathematical model under certain conditions, it is observed that the Pareto members 152, 154 and 193 have the smallest distance from the ideal solution (0.0121) and therefore each of them can be used as the final solution. In solving the problem under uncertain conditions, numerical scenarios resulting from changes in the prioritization of companies based on the coefficient v is used in the VIKOR model. Based on the results, it is observed that the impact of different scenarios on corporate investment is not negligible and consequently investors need to pay attention to this fact.

关键词： MCDM Multi-objective optimization data-driven optimization Risk Return

来源：评论

学校读者我要写书评

暂无评论

data-driven dynamic optimization for real-time parking reservation considering parking unpunctuality

引用

IISE TRANSACTIONS 2025年第1期57卷 90-103页

作者： Yan, Pengyu Bai, Mingyan Cai, Xiaoqiang Chen, Zhibin Xie, Hongke Univ Elect Sci & Technol China Sch Management & Econ Chengdu Peoples R China Chinese Univ Hong Kong Shenzhen Key Lab IoT Intelligent Syst & Wireless N Shenzhen Peoples R China Shenzhen Res Inst Big Data Shenzhen Guangdong Peoples R China NYU Shanghai Shanghai Frontiers Sci Ctr Artificial Intelligence Shanghai Peoples R China NYU Shanghai Shanghai Key Lab Urban Design & Urban Sci Shanghai Peoples R China

This article examines a real-time parking reservation service that dynamically allocates a limited number of parking slots (which can be reused) to serve parking requests in a crowded area. The study takes into account two stochastic factors: (i) random arrivals of parking requests and (ii) driver parking unpunctuality, which are seldom simultaneously considered in the existing literature. This study formulates the dynamic parking resource allocation problem as a joint chance-constrained model to maximize the expected total revenue over a finite horizon. Due to the complex nature of the model considering the two stochastic factors, the study proposes a data-driven approach using joint chance constraint decomposition and sample average approximation to approximate the model to a deterministic mixed-integer programming model. Furthermore, a stratified ranked set sampling method is introduced to construct a high-quality sample set as the input to the deterministic model, and valid inequalities are designed to accelerate the solution process. Numerical experiments based on real-world data are conducted to validate the performance of the proposed approach under multiple parking scenarios and provide findings.

关键词： Parking reservation random demand parking unpunctuality chance constraints data-driven optimization

来源：评论

学校读者我要写书评

暂无评论

A multi-criteria data-driven study/optimization of an innovative eco-friendly fuel cell-heat recovery process, generating electricity, cooling and liquefied hydrogen

引用

ENERGY 2025年 314卷

作者： Hai, Tao Sharma, Kamal Mahariq, Ibrahim El-Shafai, W. Fouad, H. Sillanpaa, Mika Ankang Univ Sch Elect & Informat Engn Ankang Peoples R China Qiannan Normal Univ Nationalities Sch Comp & Informat Duyun 558000 Guizhou Peoples R China Nanchang Inst Sci & Technol Sch Informat & Artificial Intelligence Nanchang Peoples R China GLA Univ Dept Mech Engn Mathura 281406 UP India Gulf Univ Sci & Technol GUST Engn & Appl Innovat Res Ctr GEAR Mishref Kuwait Appl Sci Private Univ Appl Sci Res Ctr Amman Jordan China Med Univ China Med Univ Hosp Dept Med Res Taichung Taiwan Yildiz Tech Univ Fac Mech Engn TR-34349 Istanbul Turkiye Gulf Univ Sci & Technol Funct Mat Grp Mubarak Al Abdullah 32093 Kuwait UPES Sch Adv Engn Sustainabil Cluster Dehra Dun 248007 Uttarakhand India Woxsen Univ Sch Technol Hyderabad Telangana India Chitkara Univ Inst Engn & Technol Ctr Res Impact & Outcome Rajpura 140401 Punjab India Chandigarh Univ Univ Ctr Res & Dev Dept Civil Engn Mohali Punjab India

The current paper aims to develop a multi-heat integration structure for a solid oxide fuel cell, focusing on methods that reduce thermodynamic irreversibility and address environmental concerns. Hence, the suggested method comprises a bi-evaporator refrigeration-organic flash cycle, a water electrolyzer cycle, a reverse osmosis cycle, and a Claude cycle producing electricity, cooling load, and liquefied hydrogen simultaneously. Furthermore, intelligent data-driven study/optimization focusing on thermodynamic, environmental, and economic aspects are performed to highlight potential areas for enhancement. Hence, two different multi-objective scenarios using a detailed sensitivity analysis are defined. Accordingly, artificial neural networks are developed for learning and verifying objectives related to energetic and exergetic performances, the cost of liquefied hydrogen, and the reduction of CO2 emissions. Subsequently, a multi-objective grey wolf optimization is used in energy- cost-environmental and exergy-cost-environmental scenarios. The results reveal a significant sensitivity index of 0.619 for fuel cell operating temperature. Notably, the first scenario provides the most appropriate optimization way, showing an energy efficiency of 62.91 %, a liquefied hydrogen cost of 3.177 $/kg, and a CO2 emission reduction of 101.9 kg/MWh. Also, an exergy efficiency of 45.42 % and a payback time of 2.45 years are the other notable findings.

关键词： Solid oxide fuel cell Heat recovery Liquefied hydrogen Eco-friendly design Artificial neural network data-driven optimization

来源：评论

学校读者我要写书评

暂无评论

Balanced uncertainty sets for Closed-Loop supply chain Design: A data-driven robust optimization framework with Fairness considerations

引用

EXPERT SYSTEMS WITH APPLICATIONS 2025年 276卷

作者： Zhao, Yuchen Ghiasvand, Mohsen Roytvand Tosarkani, Babak Mohamadpour Univ British Columbia Sch Engn Kelowna BC Canada

Recycling is crucial for minimizing the environmental impact of plastic waste and plays a key role in sustainable supply chains. However, optimizing these networks is challenging due to uncertainties in parameters such as demand and return rates. This research focuses on designing robust Closed-Loop Supply Chain (CLSC) networks to enhance resource efficiency, reduce costs, and increase recycling rates, making the system more resilient and sustainable. A novel data-driven Robust optimization (DDRO) approach is developed, incorporating historical data to manage uncertainties, such as fluctuating demand and return rates. The research introduces the concept of "balanced uncertainty sets," where boundaries are equidistant from uncovered data, ensuring a fair and true representation of uncertainty. A Kernel Weight Adjustment (KWA) approach is proposed to balance uncertainty sets across varying levels of conservatism. Additionally, a Robust Fairness (RF) index is proposed to evaluate the balance of uncertainty sets, and a data-driven algorithm is developed to compute the RF index efficiently. Numerical results show that the developed DDRO approach generates more balanced uncertainty sets, and the RF index allows for effective comparison without added computational complexity.

关键词： Supply Chain data-driven optimization Robust optimization Machine Learning Balanced Uncertainty Sets

来源：评论

学校读者我要写书评

暂无评论

data-driven optimization for Dynamic Shortest Path Problem Considering Traffic Safety

引用

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS 2022年第10期23卷 18237-18252页

作者： Jiang, Shan Zhang, Yilun Liu, Ran Jafari, Mohsen Kharbeche, Mohamed Johnson & Johnson Supply Chain Bridgewater NJ 08807 USA Shanghai Jiao Tong Univ Dept Ind Engn Shanghai 200240 Peoples R China Rutgers Univ New Brunswick Dept Ind & Syst Engn Piscataway NJ 08854 USA Qatar Univ Qatar Transportat & Traff Safety Ctr Doha Qatar

Traffic congestion is an inescapable problem that frustrates drivers in megacities. Although there is hardly a way to eliminate the congestion, it is possible to mitigate the impact through predictive methods. This paper develops a data-driven optimization approach for the dynamic shortest path problems (DSPP), considering traffic safety for urban navigations. The dynamic risk scores and travel times at different times and locations are estimated by the Safe Route Mapping (SRM) methodology and Long Short-Term Memory (LSTM) with Autoencoder, respectively, where possible variations in the future are considered. The DSPP is formulated as a mixed-integer linear programming problem under risk constraints to minimize the total travel cost, defined as the weighted sum of distance and travel time. To improve the efficiency of the DSPP, we design an improved tabu search with alternative initial-solution algorithms to accommodate various problem scales. Moreover, subgraph and self-adaptive insertion techniques are adopted as acceleration strategies to enhance computational efficiency further. Numerical experiments investigate the computational performance and the solution quality of our algorithm. The result shows satisfactory solution quality and computational efficiency with the proposed acceleration strategies compared to the CPLEX solver, a label-setting algorithm, and a state-of-the-art algorithm. Our algorithm can also compete with Google Maps regarding the travel cost in a real network in Manhattan, NY, USA, which is promising for Urban Navigations.

关键词： Navigation Routing Safety Heuristic algorithms Shortest path problem Real-time systems Stochastic processes data-driven optimization urban navigation dynamic shortest path problem tabu search risk prediction travel time prediction

来源：评论

学校读者我要写书评

暂无评论

data-driven optimization for Commodity Procurement Under Price Uncertainty

引用

M&SOM-MANUFACTURING & SERVICE OPERATIONS MANAGEMENT 2023年第2期25卷 371-390页

作者： Mandl, Christian Minner, Stefan Tech Univ Munich TUM Sch Management Logist & Supply Chain Management D-80333 Munich Germany

Problem definition: We study a practice-motivated multiperiod stochastic commodity procurement problem under price uncertainty with forward and spot purchase options. Existing approaches are based on parametric pricemodels, which inevitably involve price model misspecification and generalization error. Academic/practical relevance: We propose a nonparametric, data-driven approach (DDA) that is consistent with the optimal procurement policy structure but without requiring the a priori specification and estimation of stochastic price processes. In addition to historical prices, DDAis able to leverage real-time feature data, such as economic indicators, in solving the problem. Methodology: This paper provides a framework for prescriptive analytics in dynamic commodity procurement, with optimal purchase policies learned directly from data as functions of features, via mixed integer linear programming (MILP) under costminimization objectives. Hence, DDAfocuses on optimal decisions rather than optimal predictions. Furthermore, we combine optimization with regularization from machine learning (ML) to extract decision-relevant data fromnoise. Results: Based on numerical experiments and empirical data, we show that there is a significant value of feature data for commodity procurement when procurement policy parameters are learned as functions of features. However, overfitting deteriorates the performance of data-driven solutions, which asks for ML extensions to improve out-ofsample generalization. Compared with an internal best-practice benchmark, DDA generates savings of on average 9.1 million euros per annum (4.33%) for 10 years of backtesting. Managerial implications: A practical benefit of DDA is that it yields simple but optimally structured decision rules that are easy to interpret and easy to operationalize. Furthermore, DDA is generalizable and applicable to many other procurement settings.

关键词： commodity procurement data-driven optimization machine learning prescriptive analytics

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：