We study a retail system in which a brand owner sells through both a direct online channel and an offline retailer under uncertain market demand. We investigate the strategic interaction between the Buy-Online-and-Pick-up-in-Store (BOPS) channel and the retailer’s demand information sharing. We find that the brand owner’s direct online sales mitigate the double marginalization effect and improve total supply chain profitability. Without BOPS, the retailer always withholds demand information. With BOPS, however, the retailer may voluntarily share demand information when BOPS primarily creates incremental online-originated demand and the retailer retains sufficient cross-selling benefits, as the competition mitigation effect and the cross-channel value-added effect together outweigh the double marginalization effect. From the brand owner’s perspective, BOPS adoption and information sharing are always strategic complements. For the retailer, complementarity holds when the online sales increment and competition intensity jointly satisfy certain conditions. To resolve cases where voluntary sharing fails, we propose two mechanisms: a fixed payment scheme and a bilateral revenue-sharing (BRS) contract. We show that the BRS contract can achieve a Pareto improvement, and its performance is primarily driven by the cross-selling rate and the online sales increment. We also extend the model to several settings including consistent pricing, competition mitigation, market expansion, and a screening game, providing a more comprehensive characterization of omnichannel retailing under demand information asymmetry.
With the rapid development of social networks, the speed and volume of information spread are increasing. However, the spread of misinformation in social networks may lead to public opinion deviation and have negative influence on the society. With the scale expansion of social networks, it is impossible to have a central controller to detect the misinformation and minimize its influence in social networks. To this end, this paper proposes a decentralized method to handle the misinformation in social networks, which enables users to cooperatively shift the attention of the target user from the misinformation, so as to reduce the influence of misinformation to the society. In our method, an intelligent agent based model is established to simulate users and their relationships in social networks. Then, intelligent agents can explore message sending paths to the target user in a decentralized manner. Finally, suitable messages and their sending paths are dynamically selected to efficiently and effectively shift the attention of the target user. The experiments on real social network datasets (i.e., Twitter, Wiki-Vote, Epinions) indicate the good performance of the proposed method in terms of the attention shift of the target user and minimizing the influence of misinformation on the society.
Faced with the various and massive information resources, it is prominent to provide users with accurate, personalized service content efficiently and comprehensively, addressing their diverse retrieval needs. To this end, for University Digital Libraries (UDLs), we propose a Personalized Information Retrieval method for UDLs based on Probabilistic Graphical Model (PIRPGM). This method integrates both keyword retrieval and semantic retrieval to enhance the relevance and precision of search results. We introduce a “Spike and Slab” prior and design a novel Collapsed Variational Bayesian (CVB) inference algorithm to estimate model parameter. The PIRPGM not only offers deep insights into topic but also alleviates data sparsity. Its effectiveness is verified across different lengths (e.g., short texts and long texts) and scenarios, particularly benefiting cold start users.
Multinational firms (MNFs) have traditionally relied on retail channels to distribute products globally. Cross-border e-commerce now enables direct consumer access, but long-distance deliveries face disruption risks, highlighting the need for suitable logistics modes. This paper examines two common cross-border logistics modes: direct shipping and bonded-warehouse shipping, to identify the optimal approach for MNFs facing disruption risks. We analyze a dual-channel supply chain where the direct channel faces disruption risk in direct shipping but avoids it with bonded-warehouse shipping. We find that, when the disruption probability is not excessively high, the MNF prefers bonded-warehouse shipping if its bargaining power is weaker or stronger than the local retailer’s. In contrast, direct shipping is more beneficial. Moreover, in the scenario where the probability of channel disruption is low and the local retailer’s bargaining power is weaker than that of the MNF, the local retailer reaps greater benefits from bonded-warehouse shipping.
The increasing complexity of Air Traffic Management (ATM) systems calls for systematic methods that can support forward architecture design and early verification. This paper proposes a model-based systems engineering (MBSE) approach that integrates a multi-architecture modeling language, developed from the GOPPRRE meta-metamodel, with satisfiability modulo theory (SMT) for static verification. The proposed method provides two key contributions: a unified KARMA-based framework that organizes strategic, operational, service, and resource perspectives into a consistent multi-domain architecture, and an embedded SMT mechanism that enables automated detection of inconsistencies and constraint violations at the early design stage. A case study illustrates the construction of a tailored UAF-based ATM architecture, where SMT checking validates capability coverage, communication integrity, and logical consistency. These findings demonstrate that integrating multi-architecture modeling with early-stage formal verification not only enhances model consistency but also provides a repeatable pathway for life cycle management – from conceptual design through system validation – thereby supporting continuous evolution of ATM architectures within a model-based development environment. Quantitative results show that the method supports architecture development with more than 350 defined meta-elements and achieves static verification in a few seconds. The integration of multi-architecture modeling with formal verification enhances traceability, reduces design rework, and strengthens the robustness of ATM system development.
Product recommendation is vital for improving the experience of users and boosting sales. Many conventional product recommender systems depend on the website browsing history of customers for precisely predicting the preferences of online users for product recommendations. Various existing methods have been developed for examining the recommendation of a product to the user, but they are not effective. Thus, this work devises a new technique, named Stacked Convolutional Neural Network with Secretary Gannet Bird Optimization Algorithm (SCNN_SGBOA). Here, the input taken from E-commerce data is demonstrated in three representations, like Context-aware model, Purchase Behavior, and User rating model. In the user rating model, the user review of the product based on a review rating is considered. In the purchase behavior model, the product’s user review based on purchasing behavior is regraded, and the context-aware model considers the user review for the product in terms of the day and time at which the product is bought. Next, all three models are combined to obtain a feature vector. Following this, context-aware personalized recommendation is executed when a new user is encountered by employing SCNN, where SCNN is tuned using the SGBOA approach. Furthermore, the newly presented SCNN_SGBOA technique measured a maximum precision of 91.877
The pressure of environmental responsibility, cost reduction and emission reduction has driven the rapid development of the remanufacturing industry. Manufacturers are now faced with the critical issue of how to achieve common development through cooperation or competition with other remanufacturers. Additionally, due to the lack of transparency in information, the decision of whether to introduce blockchain technology to alleviate consumers’ distrust in the quality of remanufactured products is also a crucial consideration for supply chain members. In light of this, four models were constructed to examine the impact of different remanufacturing modes (outsourcing and authorization) and blockchain strategies (adoption and non-adoption). The optimal decisions for each model were determined, and the selection of remanufacturing modes and blockchain adoption strategies for the manufacturer under various factors were discussed. The results indicate that, regardless of the mode, the price and demand for remanufactured products increase with consumers’ acceptance of remanufactured products, and decrease with their risk aversion and the quality fluctuation of remanufactured products. While blockchain technology may initially reduce the price and demand for remanufactured products, the authorized mode with blockchain adoption can ultimately bring higher profits to manufacturers when the blockchain service fee is low. However, when the blockchain service fee is high, manufacturers can maximize economic and environmental benefits, as well as social welfare, by choosing the outsourcing mode with blockchain adoption.
Since the advent of economic globalization, the pursuit of economic development has had a detrimental effect on the environment, resulting in serious pollution and increased greenhouse gas (GHG) emissions. Clean technology can reduce GHG emissions and mitigate environmental damage. However, their uneven development and technological barriers between advanced and unadvanced regions hinder widespread adoption. This research develops a multi-agent cooperative game model (including central government, advanced region, and unadvanced region) to investigate strategic decision-making in cross-regional clean technology diffusion. The model uses numerical simulation to evaluate how technology transfer costs, environmental benefits, and economic returns influence stakeholder behavior strategies. The simulation’s outcomes indicate the following: 1) Cooperative clean technology diffusion between advanced and unadvanced regions is feasible when both parties achieve greater economic and environmental benefits or face significant penalties for failing to adopt clean technology. 2) Insufficient participation by advanced and unadvanced regions often prompts the central government to adopt stricter regulatory measures. 3) Factors such as increased central government incentives, higher subsidies for clean technology in advanced regions, and rising costs of alternative technology in unadvanced regions significantly enhance interregional technology diffusion. The conclusions of this research provide policy recommendations and management implications for policymakers.
Amid an aging population, smart pension services provide an innovative solution to the challenges of elderly care, but they also heighten the risk of privacy breaches among older adults. Balancing privacy protection with the growth of the smart pension service market has increasingly attracted the attention of both government and society. This paper develops an evolutionary game model of regulation, grounded in prospect theory and evolutionary game theory, to examine the dual objectives of privacy protection and industry development. Simulation results demonstrate the effects of key parameters on evolutionary outcomes. The findings indicate that: 1) government guidance during the initial and developmental stages is essential for the sustainable growth of smart senior care; 2) in the mature stage, reputation mechanisms become the primary drivers of market participants’ behavior; 3) moderate government penalties are effective in regulating privacy protection, whereas subsidies exert limited influence; and 4) enhancing value perception and adjusting risk attitudes among smart aging platforms and government agencies can improve governance outcomes.
E-commerce is a rapidly growing industry that stands to gain the most from emerging technologies. Online shopping and trading have become the preferred activities for customers and other stakeholders in the e-commerce value chain. The primary issues stakeholders face include trust, secure payments, transaction costs, and transparency in supply chains and financial transactions. Blockchain technology has the potential to solve all of these issues, providing a robust framework for a more secure and efficient e-commerce ecosystem. This research delves into the potential applications of blockchain in e-commerce and presents a new model of an e-commerce ecosystem based on blockchain. In addition, a survey was conducted to gauge customers’ readiness to adopt blockchain-based e-commerce services. The acceptance study used the UTAUT2 model, and data analysis was performed with SmartPLS. The study results suggest that blockchain significantly enhances consumer trust and satisfaction. Furthermore, the analysis revealed that the statistically significant variables affecting the adoption of blockchain in e-commerce are Effort expectancy, Perceived efficiency, Price value, and Trust. It is crucial to educate individuals about the effort and trust required when implementing new technologies. The research provides a theoretical contribution by refining the UTAUT2 model and proposing a blockchain-based e-commerce ecosystem model. Its practical implications offer businesses and policymakers strategies to enhance trust, improve usability, and establish regulatory clarity for real-world blockchain adoption.
As environmental awareness grows among consumers, technology-intensive firms increasingly face a strategic dilemma of whether to retain proprietary green technologies for exclusive competitive advantage or license them to direct competitors to diversify revenue streams. In this paper, we examine this tension in a duopoly where a technology developer may license her green technology to a competing manufacturer in return for licensing fees. In our model, consumers are segmented into eco-conscious and ordinary types, and social welfare explicitly accounts for the environmental damage caused by the total carbon footprint. By analyzing equilibrium outcomes under both the licensing and no-licensing regimes, we identify the technology developer’s initial carbon label as a pivotal determinant of the optimal licensing strategy. In particular, licensing is profitable only when the initial carbon label falls below a critical threshold; above this threshold, licensing erodes profitability and is therefore not offered. Importantly, we demonstrate that, in some cases, licensing the green technology to the competitor enhances both consumer surplus and social welfare, thereby generating Pareto-improving outcomes for all stakeholders. Our sensitivity analysis further clarifies how key factors, such as the proportion of eco-conscious consumers, the intensity of consumer environmental concern, and the degree of product substitutability, shape the optimal licensing strategy. Finally, we extend the baseline model to a setting with multiple competing manufacturers and confirm the robustness of our key findings.
Fake reviews on e-commerce platforms mislead consumer decisions, damage merchant reputations, and undermine market order. Detecting fake review has become vital for protecting customer rights and maintaining platform fairness. Existing fake review datasets often suffer from severe class imbalance, which degrades the performance of detection model. To address this issue, we propose a data augmentation method based on Large Language Models with Multi-Stage Prompting (LLM-MSP) for fake review detection. First, the original reviews are parsed by LLMs to extract key elements such as review targets, detailed descriptions, and sentiment tendencies, and converted into structured data. Second, prompts are then constructed based on the extracted structures to guide LLMs to generate natural and semantically coherent reviews, and the high-quality generated reviews are selected to augment the original dataset. Finally, different classification models based on BERT are adopted to test the effectiveness of LLM-MSP for fake review detection. Experimental results show that, compared with the single-stage prompting, LLM-MSP improves the novelty and diversity of the generated fake reviews. Furthermore, among various data augmentation methods, LLM-MSP achieved the best classification performance. Specifically, compared with the original imbalanced dataset, the classification accuracy increased about 10
An edge cloud framework is employed to decentralize power to network’s edges. Edge computing brings data processing closer to the source. To avoid overloading or underutilization, which can result in hardware failures or execution delays, effective load balancing is crucial. Hence, this research proposes a novel approach for optimal load balancing within Cloud-Edge Environment (CEE). The process begins by defining cloud-edge model. Next, an improved Bidirectional Long Short-term Memory (Bi-LSTM) model is proposed, incorporating architectural improvements and a normalized loss function to achieve accurate load prediction. Following this, optimal load balancing is performed at edge execution layer using Botox Optimization Incorporated Secretary Bird Optimization (BTI-SBO) algorithm that integrates Secretary Bird Optimization (SBO) and Botox Optimization (BTO), taking into account of constraints such as makespan, server load, response time, Turnaround Time (TAT), reliability, and execution time. The proposed BTI-SBO algorithm enabling faster convergence to high-quality solutions.
Given the information overload brought about by the widespread application of e-commerce questions and answers (Q A), it has become necessary to identify valuable Q As in a sea of Q As. This study is the first to introduce the concept of popularity in e-commerce Q As to distinguish the value of Q As, and construct an effect framework for the impact of popular Q A on subsequent reviews based on signal theory and expectation disconfirmation theory to prove the rationality of this identification method. We collect data from the JD.com and propose the information and confirmation effects of popular Q A. Through deep learning and rule-based methods, the mediating variables of the explanation mechanism are constructed. Firstly, we analyse the most popular Q A and the results show that for different product type, (1) the information effect of the most popular Q A reduces subsequent negative reviews related to product quality (search products) and product fit (experience products), leading to an increase in ratings; (2) the confirmation effect of the most popular Q A reduces the content about product quality (search products) and product fit (experience products) in reviews, leading to a decrease in review length. Secondly, this study further proposes reference thresholds for identifying what is a popular Q A. Finally, we repeat the above analysis for other less popular Q As to verify the effectiveness of the identification method. The results provide practical guidance for identifying the value of Q As.
In supply chains characterized by both cooperation and competition, quality disclosure strategies become increasingly complex and critical for enterprise. To address this issue, we examine a three-echelon supply chain in which an upstream supplier possesses superior quality information and may vertically integrate by encroachment, while a downstream retailer can open a store brand (SB) to enter the market. We find that opening a direct channel intensifies competition, which reduces the supplier’s incentive to disclose quality information and thus affects market transparency. In addition, the disclosure cost not only directly influences the supplier’s disclosure strategy, but also influences the supplier’s and the retailer’s encroachment strategies. Interestingly, we find that the supplier may prefer the retailer to enter the market alone rather than encroaching itself, even though sole encroachment would yield higher standalone profits. Finally, we find that a high disclosure cost may lead the supplier to reveal less information, yet consumer surplus nevertheless increases. This study provides insights for managers to understand the interaction between product quality information disclosure strategies and market entry strategies among enterprises in the supply chain.
The performance-based warranty can not only protect customers against uncertainty regarding product performance but also assist manufacturers in establishing credibility and promoting sales. This study first develops a novel performance-based warranty policy that integrates limits on the number of repairs for both performance failure and customer-induced accidental failure, aiming to safeguard customer rights and mitigate potential risks for the manufacturer. When the total number of repair actions for performance failures exceeds the threshold, we offer customers flexible compensation options. We use the expected total maintenance cost during the cycle to determine the optimal warranty length and guaranteed performance level from the manufacturer’s perspective. Through numerical experiments, we have derived several interesting practical implications: (i) The manufacturer’s top priority is to continuously improve product performance to reduce the potential number of performance failures. (ii) When limits on the number of repairs for performance failures are lower, the manufacturer is advised to provide more reliability information to help customers evaluate repair expenses more accurately. This mutual exchange benefits both parties by reducing the expected cost during the maintenance cycle.
Online debates are often reduced to binary categories, such as favor, against, and none, by using traditional opinion analysis methods. This oversimplification fails to capture the nuances of opinions, such as the intensity of feelings or the degree to which a stance varies in favor or against. To address this limitation, a new method, SFBiLT, is proposed, which leverages a fuzzy logic-based BiLSTM model optimized using the Teacher-Learning-Based Optimization (TLBO) technique to provide more nuanced categories for opinions. This approach enables a more accurate understanding of people’s views, particularly in the context of ordered-class classification. The performance of the proposed model is evaluated on two new datasets and the benchmark SemEval-2016 TaskA dataset. The SFBiLT model demonstrates significant improvements over State-of-the-Art (SOTA) models, achieving accuracy and F1-score improvements of 6.11
Manufacturers increasingly face the challenge of aligning profitability with sustainability in evaluating multiple recycling modes. This paper investigates a monopolistic manufacturer’s production, recycling and pricing strategies under four modes: no-recycling, trade-in, buyback, and hybrid recycling. Through comparative analysis of optimal decisions across these recycling modes, this paper derives key conclusions from dual lenses of economic and environmental performances. 1) Not all recycling modes boost economic and environmental performances. In terms of economic performance, trade-in and hybrid recycling modes outperform no-recycling mode in profit, but buyback mode underperforms when the proportion of existing customers is low. In terms of environmental performance, while recycling enhances material recovery, it also stimulates demand for new products. Notably, when the unit environmental impact of used products in extended use stage is lower than in other stages, no-recycling mode achieves superior environmental performance overall. 2) Among three recycling modes, the hybrid recycling mode is more outstanding in both economic and environmental performances. The comparison regarding economic performance between the two single recycling modes (trade-in and buyback) mainly depends on customers’ discount coefficient for used product valuation. The environmental performances of buyback and hybrid recycling modes are identical and superior to that of trade-in mode. These findings provide theoretical guidance and management insights for manufacturers to select the appropriate recycling mode.
Resource recycling is a cornerstone of sustainable supply chain development. However, information asymmetry among multiple stakeholders often hinders effective consensus, significantly impairing decision-making efficiency in reverse logistics, particularly in resource recycling. This study proposes a blockchain-based closed-loop supply chain (CLSC) system to elucidate the reconstruction mechanism of the CLSC network and promote sustainable supply chain development. The reconstruction focuses on two core issues: fault-tolerant consensus and profit distribution. Using a trust-enhanced Byzantine Fault Tolerance (BFT) consensus mechanism and a Shapley value-based profit distribution model, this study deciphers the operational logic underlying CLSC reconstruction. Findings indicate that a two-stage approach integrating fault-tolerant consensus and profit distribution effectively promotes the integration of forward and reverse chains. A cooperative governance model is essential for transforming supply chain resource recycling, and policy regulation is necessary to ensure the stable operation of distributed recycling systems. Together, these findings advance the theory and practice of dual-chain integration and closed-loop supply chains.
This paper examines how manufacturers’ fairness concerns and government subsidies affect carbon emission mitigation and operational performance in a closed-loop supply chain. To analyze the effect of power asymmetry, we develop game-theoretic models under three typical supply chain power structures: Manufacturer-led Stackelberg (MS), Retailer-led Stackelberg (RS) and Vertical Nash (VN). Several key findings emerge from the equilibrium analysis. First, optimal emission reduction is achieved in the VN mode under low fairness concerns and in the RS mode under high subsidies. Second, the power structure critically determines the optimal subsidy: higher subsidies are warranted in RS mode, as retailer dominance dampens the manufacturer’s voluntary investment, whereas lower subsidies suffice in MS mode, where manufacturer leadership incentivizes greater investment. Finally, contrary to conventional expectations, a manufacturer’s fairness concerns do not invariably improve its profits. Under MS mode, strong fairness concerns can reduce profits by distorting pricing and investment decisions, as the manufacturer may prioritize equitable distribution over economic optimality. This study offers insights into the interplay of power structures, behavioral preferences, and environmental policies in sustainable closed-loop supply chains.
