Abstract:Under the context where extreme weather affects the freshness of agricultural products,this paper establishes dynamic game models for a two-echelon agricultural products supply chain across five cases,in which the government always grants a tax reduction to the retailer and may additionally provide an order-based or sales-based subsidy to either the processor or the retailer.The optimal tax reduction and subsidy strategy is analyzed in terms of supply chain operations,government financial performance,and public welfare construction.Results show that intensifying extreme weather or increasing the tax-reduction rate both push the supply chain further away from optimal operation.Retailers'engagement in public welfare can improve supply chain performance,though overly intense public welfare reduces retailer profits.Moreover,compared to the single tax reduction policy,the strategy of tax reduction followed by subsidy can improve supply chain operations,enhance government financial performance,and strengthen the effectiveness of public welfare performance,with order-based subsidies outperforming sales-based subsidies.From the perspectives of optimizing supply chain decisions,increasing government tax revenue,enhancing fiscal incentive effects,improving consumer surplus,strengthening social welfare,and reinforcing public-welfare implementation,the government should adopt a"retailer tax reduction+processor order-based subsidy"policy.From the perspective of improving the efficiency of fund utilization,the optimal policy becomes"retailer tax reduction+retailer order-based subsidy".

Abstract:With the deep integration of next-generation information technologies and high-end equipment manufacturing,traditional operations and maintenance(O&M)models are encountering serious challenges in terms of real-time performance,intelligence,and full lifecycle management.From the perspective of digital and intelligent transformation,this study systematically explores a novel digital and intelligent O&M management paradigm driven by emerging technologies,aiming to construct a technology-function-organization collaborative framework that promotes high-quality industrial development.The study reveals the core characteristics of digital and intelligent O&M:data-driven decision-making,intelligent technology penetration,full lifecycle management,cross-domain collaboration,and diversified value creation.Based on this,the study proposes a three-layer O&M management model of"technology-function-organization"and combines the technological foundations,functional closed loops,and industrial ecosystems to achieve a paradigm shift from passive response to proactive prevention and value co-creation.Case studies demonstrate that Goldwind Technology establishes a closed-loop system through technology integration,verifying the significant effectiveness of the proposed O&M model in enhancing efficiency,optimizing resources,and promoting collaboration.The study further proposes a pathway for constructing an O&M service ecosystem,providing theoretical support and practical guidance for advancing digital transformation in the high-end equipment manufacturing industry.

Abstract:To tackle the key challenges in product quality modeling for complex manufacturing systems,such as nonlinear dependencies across multiple processes,intricate structural relationships,long-range error propagation,and insufficient process engineering knowledge,this paper proposes a modelling method based on an improved graph attention network called Inter-layer Contrastive Loss Filtering Graph Attention Network(ICLF-GAT).Initially,a data-driven deep learning framework is adopted to avoid reliance on prior physical knowledge.Subsequently,a directed graph model of the manufacturing system is constructed based on Graph Attention Networks(GAT)to effectively capture complex structural features and nonlinear dependencies between processes.On this basis,a novel inter-layer contrastive loss filtering mechanism is introduced,which dynamically evaluates and filters the quality of node features to significantly mitigate the over-smoothing problem in deep GATs and enhance the modeling capability for long-range error propagation.Finally,a target attention decoder is designed to further improve the modeling accuracy of the systematic complex dependencies.Simulation experiments and a practical industrial case demonstrate that ICLF-GAT significantly reduces the Root Mean Squared Error(RMSE)compared to existing benchmark methods,with particularly advantages in long-range error propagation.

Abstract:With the rapid development of smart and connected products,the vast volume of user-generated online reviews has become a vital source for understanding user preferences and tracking market demand dynamics.To reveal the evolution of user opinions from high-dimensional,heterogeneous,and temporally structured reviews,this study develops a multimodal analytical model capable of dynamically capturing user needs and opinion evolution to support product iteration and enhancement.Specifically,the study first integrates textual and visual features of online reviews to construct the initial nodes of the review network using spectral clustering.Subsequently,by incorporating the temporal memory structure of long short-term memory(LSTM)networks and a weight-learning mechanism into the DeGroot model,a dynamic DeGroot-LSTM model is introduced to adaptively update influence relationships among nodes and capture the nonlinear dynamics of opinion propagation.By analyzing node opinions and weight matrices,the prediction of user demand and opinion evolution trends across different product features is conducted.A case study on the Echo Dot smart speaker demonstrates that the proposed model achieves the average MSE,MAE,and R2 of 0.003 2,0.027 6,and 0.829 5,respectively,across all features,significantly outperforming the static DeGroot model and the linear Ridge regression model.The findings not only provide data-driven decision support for the optimization of smart and connected products but also offer a novel methodological framework for studying opinion evolution from a multimodal online review perspective.

Abstract:To explore the collusion and loan fraud behavior of low-carbon supply chain enterprises with financial constraints in the context of green finance,a tripartite game model is constructed among banks,core enterprises,and small and medium-sized enterprises.The model identifies the key factors that drive collusion between core enterprises and small and medium-sized enterprises to defraud green loans from banks in low-carbon supply chains.Also,the impact of bank regulatory cost,regulatory success rates,and collusion penalty levels on enterprise collusion decisions and bank regulatory strategies is analyzed.Results show that the"operational ease"and"concealment"of collusion among enterprises,as well as the regulatory cost and penalty levels of banks,are key factors influencing collusion behavior in low-carbon supply chains.With the continuous increase in bank regulatory cost,its motivation and intensity are weakening,leading to a high probability of enterprise collusion.When the green lending rate of a bank is relatively high,that is,when its gap with the ordinary lending rate is small,the probability of collusion among enterprises increases.By investing in and deploying emerging technologies to enhance regulatory efficiency(i.e.,higher regulatory success rates)and by strengthening penalties for collusion,banks can effectively deter such behavior and reduce both the motivation and probability of collusion.

Abstract:In a two-echelon supply chain that consists of a single manufacturer and a single retailer,the decision efficiency of push-and pull-based supply chains with wholesale-price contracts led by either the manufacturer or the retailer is studied,across four ownership structures:no cross-holding,retailer unilateral holding,manufacturer unilateral holding,and mutual cross-holding.Based on the newsvendor model,the first order optimality conditions for decentralized decisions are derived for both push and pull supply chains under each ownership structure.A comparative analysis of order quantities,wholesale prices and supply chain efficiency is conducted,and the impacts of manufacturer/retailer's shareholding ratio as well as product margin rate are further investigated.Results show that,for both push and pull supply chains,efficiency is highest when the contract leader holds shares in the follower,and lowest when the follower holds shares in the leader,with mutual cross-holding leading to intermediate efficiency.Furthermore,under the cases of no cross-holding or retailer unilateral holding,supply chain efficiency of the retailer-led pull configuration is higher than that of the manufacturer-led push configuration.In contrast,under the other two cases,the manufacturer-led push configuration becomes more efficient only when the manufacturer's shareholding ratio is relatively high and the product margin rate is relatively low.

Abstract:In the context of e-commerce supply chains,taking the no-credit case as the benchmark model,this study incorporates a consumer payment choice model to characterize payment preference.Decision models for credit service introduction are then developed under scenarios with and without payment preference.The Karush-Kuhn-Tucker(KKT)condition and backward induction are used to derive the optimal decisions for all scenarios.The impact of consumers'payment preference behavior on the introduction of credit services and its fee rate strategy of e-commerce platform is explored through comparative analysis.Furthermore,based on the consideration of payment preference,the study innovatively explores two additional scenarios:differentiated default rates and a combined payment mode of"cash+credit".Results show that:1)In the credit-service scenario without payment preference,the e-commerce platform introduces credit services when the trouble cost is low;when the default rate is high or both the default rate and the trouble cost are low,an interest-charging strategy is adopted;otherwise,a subsidy strategy is adopted.2)In the credit-service scenario with payment preference,only consumers with sufficient budgets choose the credit services introduced by the platform when the pain-of-paying coefficient for cash is high and the trouble cost is low,even if the platform only adopts an interest-charging strategy.3)Under the condition that is more conducive to consumers using credit payment,the higher the total demand under payment preference,the greater the possibility of achieving a win-win situation for both parties in the e-commerce supply chain.4)Differentiated default rates do not change the platform decisions regarding credit introduction and fee rate strategies;however,in the combined payment scenario,as the proportion of cash payment gradually decreases,the platform shifts from an interest-charging strategy to a subsidy one.

Abstract:To enhance mission safety and cost-effectiveness,this study aims to address the challenge of reliability assessment for reusable rocket engines across multiple mission phases.Key operational phases,such as ignition and launch,maximum dynamic pressure,stage separation,secondary orbit insertion,and landing capture,are analyzed using the goal-oriented flow(GO-FLOW)method to construct a comprehensive system reliability model.Physical components are abstracted into operators,with signal propagation mechanisms employed to evaluate the dynamic evolution of propellant supply,combustion,and control subsystems across these phases.A performance consistency metric is proposed,with calculation results demonstrating a system-wide performance consistency of 80.6%,indicating high reliability.Sensitivity analysis conducted using MATLAB involves adjusting the key operator parameters from 0.95 to 1.00.Results identify the nozzle as the most influential component,contributing up to 14.9%to system performance consistency,followed by the propellant tank(11.7%),fuel pipeline(8.1%),control valve(7.1%),and combustion chamber(14.6%).These findings provide a quantitative insight for pinpointing critical vulnerabilities in the system.The key components impacting reliability at the five mission phases are identified as igniters,control valves,combustion chambers,gimbal mechanisms,and turbopumps.Compared to conventional reliability analysis methods,the GO-FLOW approach demonstrates superior dynamic modeling capabilities and computational efficiency.This study offers novel insights for the design and operational planning of reusable rocket engines.

Abstract:This study investigates the alignment between government subsidy policies and enterprise behavior preferences in blockchain-based agricultural assistance.It aims to reveal the impact of different subsidy schemes on the decisions of self-interested and altruistic enterprises,thereby optimizing policy effectiveness and the sustainability of agricultural assistance.A tripartite Stackelberg game model involving the government,enterprises,and farmers is constructed to analyze equilibrium decisions under cost-based and sales-based subsidies across various scenarios.Sensitivity analysis and strategy comparisons are conducted to validate the results.Findings demonstrate that sales-based subsidies effectively incentivize self-interested enterprises to enhance traceability efforts,expand market scale,and significantly increase farmer income;in contrast,cost-based subsidies are more suitable for altruistic enterprises with high research and development costs,which can alleviate their financial pressure.Further analysis shows that when technology costs are low and consumer traceability preferences are high,sales-based subsidies achieve multi-win outcomes;otherwise,the system may fall into a"prisoner's dilemma".The study innovatively proposes a"policy-preference"matching mechanism,providing theoretical guidance and policy recommendations for designing targeted and differentiated blockchain-based agricultural subsidy policies.

Abstract:With the rapid aging of population in China,the medical resource imbalance is becoming increasingly prominent.In this condition,the"Internet+Healthcare"policy has emerged as an important measure to activate the potential of primary healthcare.This study focuses on the pricing strategies of primary care physicians on online healthcare platforms,aiming to explore the impact mechanism of physician service prices on patient choice behavior.Based on the signal theory,regression analysis is applied using data from Haodf.com online platform,including basic information,behavioral data,and patient numbers of primary care physicians from January to December 2021.The effect of physician service prices on patient choice is thoroughly analyzed,along with the moderating effects of physician titles,city tiers,online knowledge sharing behavior,and the number of patients previously served.Results indicate that physician service prices have a significantly positive effect on patient choice behavior.Patients are more likely to choose higher-priced physicians in first-tier cities.Moreover,both online knowledge sharing and the number of patients already served positively affect physician service prices for patient choice,whereas physician titles do not significantly influence this relationship.These results contribute to understanding how service prices of primary care physicians affect patient medical choices in online healthcare,and how the offline characteristics and online behavior of physicians moderate this relationship.These findings guide community physicians in adjusting service prices and managing online consultations,while providing empirical support for platform operators to address resource allocation imbalances in China.

Abstract:In addressing the inventory allocation problem in automated storage and retrieval systems(AS/RS),a multi-objective optimization model was established based on factors such as the operational path of the stacker crane,acceleration,and product categories.The objectives of this model are to reduce the energy consumption of the stacker crane,enhance the spatial aggregation of related products,and ensure the overall stability of the shelves.The standard multi-objective whale optimization algorithm(NSWOA)and an improved multi-objective whale optimization algorithm(INSWOA)were used to solve the model.Simulations were conducted using MATLAB on a real-world instance of inventory allocation,leading to specific allocation results.Comparative analysis proved that the proposed model and algorithms effectively improve the issues of high energy consumption,low correlation among similar products,and disorganized product placement in the operational processes of automated storage systems,thereby providing an effective solution to the inventory allocation problem in AS/RS environments.

Abstract:With the rise of the resale market,it makes resale transactions a common phenomenon in the luxury industry,while the strategic interactions between brand owners and second-hand dealers remain underexplored.To analyze the optimal resale strategies of brand owners,this study constructs a game theory model incorporating consumer conspicuous behavior to explore the typical strategies employed by brand owners in response to second-hand dealer intrusion.Three strategies are included,which are boycott,competition,and brand licensing.Optimal pricing and profits of both brand owners and second-hand dealers are derived using different strategies.Results show that brand owners consistently benefit from the status-seeking conspicuous behavior of consumers.When second-hand dealers enter the market with products of equally low(or high)quality and authenticity,they either damage the brand image and dilute brand value or harm the profits of brand owners due to the strong substitution of second-hand products.Licensing fees,as a regulatory tool,can enhance channel efficiency,and the double marginalization effect in decentralized channels under brand licensing can actually benefit both brand owners and second-hand dealers,leading to a win-win outcome.Additionally,the core conclusion remain robust when the consumer utility function is extended,and a reduction in profit may occur only when the brand owner chooses the competition strategy.The conclusions of this study provide theoretical foundations and strategic recommendations for luxury brands to formulate scientific resale decisions.

Abstract:According to whether customers purchase on-time delivery services or are willing to pay additional fees for early delivery,on-demand delivery orders are classified into different types.In addition to the initial order requirements,new orders may arise during the delivery process.A mathematical model is established with the objective of minimizing the sum of vehicle delivery costs and customer time costs,considering constraints such as delivery time windows,vehicle capacity,and the service range of crowdsourced vehicles.An improved hybrid tabu search algorithm based on rolling time horizons is designed to solve the problem,which incorporates a dynamic adjustment mechanism for the tabu step size and a diversification strategy for solutions.Parameter analysis shows that,in order to effectively reduce costs,transportation companies should avoid setting long update intervals.Instead,they should prioritize the delivery of heterogeneous orders of the second and third categories,and expand the service range of crowdsourced vehicles while fully utilizing the vehicles within this range.Multiple case studies of different scales demonstrate that the improved hybrid tabu search algorithm based on rolling time horizons can effectively solve cases of various scales.

Abstract:In order to study the platform advertising strategy of merchants on e-commerce platform with membership programs,this paper establishes game models under two scenarios:one without platform advertising and the other with platform advertising through e-commerce platform membership.The reverse induction method and comparative analysis are used to identify the mechanism of merchants selecting a platform advertising strategy and the impact of the strategy on the decisions and profits of supply chain members.Results show that merchants should not adopt platform advertising if the membership discount is below a threshold,or if it exceeds a threshold while the commission rate remains low.If both membership discount and the commission rate exceeds their respective thresholds,introducing platform advertising is a"win-win"strategy for merchants and the platform from the perspectives of both profit and market share maximization.Platform advertising is always detrimental to nonmembers but can benefit members when the commission rate and the membership discount satisfy specific conditions.To ensure member rights and maintain the diversity of product categories with membership discounts,the platform should set membership discounts based on the commission rate for each product category.

Abstract:This study addresses the home health care routing and scheduling problem(HHCRSP),considering the high randomness in patient service time and caregiver travel time,coupled with the differentiated priorities among patient groups.Traditional deterministic optimization approaches struggle to balance the robustness and efficiency in scheduling under such conditions.To tackle these challenges,this study proposes innovative solutions at both modeling and algorithmic levels.At the modeling level,a distributionally robust optimization(DRO)framework is introduced to construct an ambiguity set based on first-order moments and absolute deviation moments,where the distributional uncertainty of random variables are captured.This allows the establishment of a DRO model that maximizes total priority-based revenue while controlling time-related risks,without relying on exact probability distributions.At the algorithmic level,an exact solution approach is designed to address the computational difficulties arising from the complex constraints.By efficiently generating cutting planes and implementing convergence strategies,the algorithm enhances solving efficiency and solution stability.Through comprehensive numerical experiments,the proposed DRO model is compared against classical stochastic programming and deterministic models.Results demonstrate that the DRO model exhibits superior robustness under uncertainty.It effectively balances service efficiency and risk control by adjusting confidence levels,enabling decision-makers to achieve a trade-off between service quality and operational costs based on actual risk preferences.Furthermore,the proposed exact algorithm exhibits notably superior efficiency over commercial solvers in test cases involving complex parameter combinations,providing efficient and reliable decision support for HHCRSP.

Abstract:Enhancing the rationality of urban electric taxi charging station planning and layout,and alleviating the problem of low utilization efficiency,are of great significance in reducing driver time costs and promoting green mobility.To this end,this paper first simulates the operation process of electric taxis to obtain random charging demand points.Then,the Dijkstra algorithm and a queuing simulation model are used to simulate the process of taxis searching for charging stations and waiting for charging.An optimization model for charging station locations is established with the objective of minimizing the searching time,queuing time and loss costs.The model is solved using an improved adaptive large neighborhood search(ALNS)algorithm.Experimental results show that the improved ALNS algorithm demonstrates superior performance,achieving the lowest costs with a reasonable allocation of station locations and charging piles.

Abstract:To address the issues of excessive emissions and inadequate emission reduction management in the contract manufacturing industry,this study explores practical pathways to simultaneously achieve economic development and green production.Two types of contract manufacturing structures are constructed including an original equipment manufacturer(OEM)model focusing on green emission reduction,and an original design manufacturer(ODM)model characterized by cost-sharing of emission reduction efforts.This paper compares the heterogeneous performance of these two models in terms of economic and environmental benefits,with particular attention to the key factors affecting the optimal decision-making of participating firms.Results show that the ODM model with cost-sharing achieves a higher optimal emission reduction level.As the level of green production improves,the economic benefits of the cost-sharing ODM model become increasingly pronounced,encouraging a transition towards greener manufacturing structures.However,considering the rising costs associated with green technology investment,firms may prioritize economic benefits while neglecting emission reduction effectiveness.Therefore,technological innovation is crucial.

Abstract:To precisely match medical needs and optimize resource allocation,a doctor-patient matching model in two-way referrals under the medical alliance model has been established.Firstly,the doctor evaluation framework was constructed based on the iceberg theory,and the doctor evaluation model of bidirectional referral was obtained by combining the evaluation indexes considered by up-and-down referrals.The support vector machine algorithm was used to predict the patient's disease type,and the random forest algorithm was used to obtain the patient evaluation model after mining the risk factors,thus constructing a complete evaluation system for the doctor competency model and patient characteristic profile at the referral stage.Secondly,the first round of doctor-patient matching was conducted based on the Gale-Shapley algorithm,incorporating the patients' disease risks.Additionally,by integrating personalized rankings,patients' requirements were fulfilled to enhance matching satisfaction.Then,the dynamic multi-objective optimization algorithm was employed to achieve precise matching for all patients.Finally,the feasibility and accuracy of the model were verified by numerical experiments using the real data of Haodaifu website and offline hospitals.

Abstract:Against the backdrop of the diagnosis-intervention packet(DIP)payment reform,this paper explores the upcoding behavior of hospitals,and provides insights for the implementation of medical insurance payment reform.An extensible asymmetric dynamic evolutionary game model is developed and simulation analysis is conducted based on real-world cases to reveal the key factors of medical violations.Results show that increasing penalty intensity and expanding reputational damage are conducive to curbing the upcoding behavior of hospitals.Lowering regulatory cost while increasing penalty intensity and regulatory returns,encourages insurers to exercise strict supervision.Reducing complaint cost and improving complaint benefits provide protection for patient complaints.The probability of patient complaints increases with the probability of hospital upcoding,while hospital upcoding decreases as insurers intensify supervision,and insurer supervision strengthens with the decrease of patient complaints.The ultimate driving force of upcoding comes from extra benefits,which is independent of whether insurers or patients are in a supervisory state.Under the current management regulations of DIP in China,the internal motivation for hospital upcoding is insufficient,and there is no stable equilibrium point in reality.

Abstract:The rapid development of data technologies has significantly expanded the application of data-driven approaches in blood inventory management.To elucidate the research evolution,identify key hotspots,and clarify future trends,a comprehensive review of existing research is conducted.Based on a systematic search across multiple authoritative databases,92 relevant studies are selected and analyzed using bibliometric and knowledge graph techniques.The analysis provides a macro-level examination from multiple perspectives,including publication volume,journal distribution,author collaboration networks,and keyword co-occurrence patterns.Three critical areas are focused:blood demand forecasting,inventory level control,and inventory allocation and distribution,while applications of data-driven methods in these areas are systematically summarized.Based on the limitations of existing research and the practical needs of the industry,future research directions are proposed from four aspects:research problems,methodological frameworks,data foundations,and system-level applications.To the best of our knowledge,no comprehensive review has yet been conducted focusing on data-driven blood inventory management.This study fills this gap by providing valuable insights into literature integration and structured analysis.The findings contribute to a better understanding of the current research progress and offer theoretical and methodological support for the intelligent transformation of blood inventory management.