This study explores how digital leadership shapes psychological well-being among service employees in China, focusing on the mediating role of digital communication quality and the moderating influence of knowledge sharing willingness. Drawing on Social Exchange Theory, we develop and test a moderated mediation model using data collected from 308 service-sector employees across multiple Chinese cities. Confirmatory factor analysis confirmed construct validity, and structural equation modeling demonstrated that digital leadership positively affects communication quality, which in turn enhances psychological well-being. Moderation analysis revealed that the relationship between digital leadership and communication quality is stronger when employees are more willing to share knowledge. Further, PROCESS analysis supported a significant conditional indirect effect, suggesting that the mediating path is amplified under high knowledge sharing willingness. These findings underscore the importance of employee disposition in shaping the outcomes of digital leadership. The study contributes to leadership and organizational behavior literature by integrating communication and knowledge behaviors into a relational leadership framework, with implications for improving well-being in digitally intensive service settings.
Clients seeking paramedical and rehabilitation services require recurring treatment sessions over an extended period. Unlike single-visit problems, these services must assign each accepted client to a fixed, recurring day-time slot that remains occupied throughout the treatment program. This structure creates long-term capacity commitments that limit future scheduling flexibility and complicate acceptance decisions, particularly when clients differ in availability and required program durations. Motivated by an early intervention program for infants and toddlers with developmental delays, we study scheduling policies designed to address this combination of heterogeneity, uncertainty, and recurrence constraints. We model the multisession appointment scheduling problem as a Markov decision process in which requests arrive sequentially and decisions must consider both immediate feasibility and the long-term implications of blocking a slot across many periods. Our analysis identifies key structural elements of the scheduling decision, including a slot-selection guideline that assigns accepted clients to the least popular feasible slot and a duration-based threshold that characterizes acceptance behavior. These insights highlight the value of preserving flexibility and anticipating demand when scheduling recurring appointments under uncertainty. Building on these results, we develop a heuristic that groups schedule states into occupancy categories and applies simplified acceptance thresholds. Computational experiments show that this anticipatory approach outperforms first come, first served benchmarks, particularly when slot popularity is uneven or program durations vary widely. Whereas we do not model health outcomes directly, prior research links improved access, timely initiation, and continuity of care with better therapeutic results, underscoring the broader potential impact of more efficient scheduling.
Strategic and effective use of online review systems is a critical challenge for firms in the service industry, which can only participate in these systems while abiding by each system’s unique rules and designs. Through an experimental study, we extended the theoretical understanding and explored the practical benefits that firms can receive by writing managerial responses. We showed that topic-congruent responses have a strong positive influence on users’ purchase intention through increased trust and major trust antecedents, such as perceived diagnosticity and social presence. In positive reviews, both perceived diagnosticity and perceived social presence were significant mediators. On the other hand, for negative reviews, the only significant mediation was through perceived social presence. Our work complements previous research on personalized managerial responses, which mostly depends on archival data and does not provide causal insight into the individual mechanism behind the positive effects of writing managerial responses for the firm. The results of our study are useful to any firm that wants to optimize its managerial response strategy in online platforms by providing a detailed explanation of how potential customers can be influenced by the different strategies. We also expand the existing knowledge of how consumers develop trust and purchasing intentions in an electronic commerce environment when exposed to different managerial response strategies.
We consider the problem of optimally maintaining an offshore wind farm wherein wind turbines progressively degrade over time because of normal usage and exposure to a randomly varying environment. The turbines exhibit both economic and stochastic dependence due to shared maintenance setup costs and their common environment. However, even modestly sized farms can be difficult to analyze because of dimensionality considerations. Our aim is to identify near-optimal replacement policies that minimize the total expected discounted setup, replacement, and lost production costs over an infinite time horizon. The problem is formulated using a Markov decision process model that becomes intractable as the number of turbines increases. Therefore, we introduce a novel value function decomposition within the approximate linear programming (ALP) framework that exploits the unique features of the problem. Using a column generation algorithm to solve the dual of ALP (DALP) formulation, we establish lower and upper bounds for the value function, enabling us to quantify the optimality gap, which is under 1% for homogeneous farms and between 4% and 6% for heterogeneous farms. Furthermore, we bound the optimality gap between the exact value function and the approximate one using the optimal solutions of the dual of the linear programming formulation and the DALP formulation. Finally, we provide conditions under which an optimal policy can be retrieved from the approximate value function. Our results offer invaluable guidance to wind farm operators on managing the high costs of offshore wind farm maintenance. Because a significant portion of wind energy costs are linked to the replacement of major components, near-optimal replacement policies may help reduce these costs and make wind energy more cost-effective. Through sensitivity analyses, we explain the impact of three key factors-the environment, shared setup costs, and the number of turbines-on the characteristics of these policies. Additionally, we provide insights into the structure of the near-optimal policies and explain nonintuitive behaviors within the context of offshore wind turbine farms.
Despite efforts to increase the heart supply for transplantation, nearly 68.9% of donor hearts were unused. Current emphasis is placed on the information asymmetry between transplant candidates and the allocation system. Recently, a disruptive technology called the Organ Care System (OCS) has been shown to significantly extend heart preservation time. To improve heart utilization, this study develops an analytical and operational framework that jointly considers information asymmetry and OCS deployment. For patient self-selection of transplant centers with private information, we propose an incentive-compatible mechanism that encourages candidates to adhere to their declared willingness. For OCS deployment with a limited financial budget, we design a transplant network that maximizes patients' social welfare. Theoretical analysis for small-scale transplant networks identifies equilibrium transplant menus, encompassing transplant rates and waiting times, for each transplant center. Specifically, we find that time-insensitive patients on the queue may be strategically delayed from receiving transplants when the sensitivity gap between patient types is small. This sensitivity gap is determined by the heart supply and the rate of time-insensitive patients. Case studies show that the OCS technology can improve social welfare and reduce organ wastage. Moreover, time-sensitive patients benefit more than time-insensitive patients. Empirical evidence confirms the existence of strategic delays, as verified in the theoretical analysis.
Artificial intelligence (AI) is reshaping healthcare as a service system in which people, processes, and digital technologies cocreate value. This study conceptualizes AI-driven risk stratification not merely as a predictive classifier, but as a governed service mechanism that supports proactive patient-centered resource allocation under real-world clinical constraints. We instantiate this perspective through a long-term chronic disease follow-up service for individuals with metabolic dysfunction-associated steatotic liver disease (MASLD). MASLD is a prevalent chronic metabolic condition associated with elevated colorectal adenoma risk. Using routine, noninvasive health examination data, the service supports a risk-stratified follow-up plan. The proposed framework integrates knowledge-guided attention, explainable AI, and threshold governance to align model behavior with clinical reasoning and operational priorities. The framework further links calibrated risk estimates to decision curve analysis to support transparent, capacity-aware screening decisions, enabling healthcare providers to prioritize invasive examinations for high-risk individuals whereas reducing unnecessary procedures for lower-risk populations. Empirical evaluation using 3,971 colonoscopy cases from a health examination cohort demonstrated strong recall-oriented screening performance together with clinically interpretable decision support. Accordingly, the study provides a transferable framework for AI-enabled healthcare service redesign, supported by knowledge-aligned decision making, governed resource allocation, and explainable transparency for accountability in a long-term chronic disease follow-up plan.
Vehicles are crucial for sustaining socioeconomic activity and improving quality of life in modern cities by offering diverse services. These include passenger mobility, goods delivery, information acquisition, and acting as mobile servers such as food trucks and mobile lockers. At the same time, they also contribute to traffic congestion and air pollution. This tension fosters the rise of urban resource-conserving and sustainable service solutions. In this article, we introduce the concept of "Vehicle-Based Multi-Services" (VeMuS), in which a single vehicle offers multiple services simultaneously. Drawing on practical use cases, we examine service classification and integration for vehicles and the potential for the synergy effect and multitasking. We construct a general framework to study VeMuS and propose research questions in three key components: VeMuS design, VeMuS operations, and VeMuS evaluation. This article highlights the potential of VeMuS to create more sustainable, efficient, and adaptable urban services for future smart cities.
We design an automated decision-making system for inventory management using a Transformer-based neural network. Leveraging contextual information and historical data, the system makes two key decisions: (1) order timing and (2) order quantity. To accommodate random vendor lead times, we develop an imitation-learning framework, in which the Transformer imitates ex post optimal decisions computed from historical data to directly output inventory actions. The model adopts the GPT-2 architecture-an off-the-shelf large language model-for efficient fine-tuning under the inventory context. Our framework, InventoryGPT, addresses challenges faced by large e-commerce platforms that manage millions of stockkeeping units while serving customers with stochastic and nonstationary demand and leadtime patterns. By incorporating rich contextual data, the model learns to improve service levels and reduce costs. Empirical results using real-world data from a leading e-commerce platform show that InventoryGPT outperforms traditional and state-of-the-art benchmarks. Moreover, by carefully designing the Transformer's input and output structures, the proposed InventoryGPT model achieves good interpretability. Our study highlights the potential of Transformerbased neural networks for large-scale decision making in service operations.
Prior research identified range anxiety as a major factor limiting the adoption of electric vehicles (EVs). However, by driving over 15,000 kilometres (similar to 10,000 miles) in various electric vehicles in Canada, United States, and Europe and relying on public charging, we observed that today's EV drivers have charge anxiety instead. Charge anxiety comes from five factors: hardware issues: will the charger's plug fit my vehicle and, overall, "does it work"; software factors: will my app/card work at the specific charger; location issues: is a charger conveniently located; time issues: how long will charging take; and price issues: how much will it cost. Motivated by these observations, we present three empirically grounded analytical models, each fitted to real data from industry partners, to analyze key issues that we, as operations management scholars, see in the current state and potential trajectory of public charging infrastructure. The first is a discrete-event simulation (a "digital twin") to assess the realistic speed of fast charging. The second is a back-of-theenvelope Little's Law calculation to estimate the scale of a fast-charging station needed to match the throughput of a typical gas station. The third is another discrete-event simulation that incorporates realistic driver arrival patterns across different types of days in a year. We conclude with insights and research opportunities stemming from our observations and models.
We study a growing retail strategy called pickup partnership, where online retailers partner with physical stores to offer in-store pickup services. In practice, two main policies are used in these partnerships: (i) a fixed fee policy, where the retailer pays the offline partner a set fee per pickup order, and (ii) a coupon policy, where customers receive a coupon for use at the offline partner's store with each pickup order. Our goal is to evaluate these policies and determine which is most beneficial for online retailers. We develop a stylized model that captures the essential dynamics of pickup partnerships. We find that although the coupon policy allows the online retailer to gain greater market coverage compared with the fixed fee policy, it does not always lead to higher profits for the online retailer. The coupon policy is preferred when in-store fulfillment and pickup handling costs are low and direct-delivery costs are high, whereas the fixed fee policy is favored when these costs are moderate. We also find that both policies entail inefficiencies when the incentives of the two parties are not aligned. To alleviate such inefficiencies, we propose a new policy designed to better align incentives and improve partnership efficiency. This paper offers the first theoretical analysis of the in-store pickup partnership model and provides practical guidance for online retailers seeking to implement it. Our proposed policy aims to enhance the effectiveness and profitability of these partnerships beyond current industry practices.
In many cases, managers cannot rely on expected profit maximization to determine prices because of a lack of reliable sales data. To overcome this issue, the pricing literature suggests adopting strategies that are robust to model uncertainty. However, most of this literature addresses situations where firms sell directly to consumers, leaving a gap in understanding how to implement robust pricing strategies when products are sold through retailers. To address this gap, we propose a model that explores the impact of robustness on pricing strategies for both manufacturers and retailers. The model provides three novel insights. First, we find that in equilibrium, channel partners will seek different levels of robustness for their pricing decisions. Second, the model reveals that injecting robustness in channel pricing decisions can not only help manage unforeseeable demand shocks, as it should be, but also mitigate double marginalization. Finally, the model uncovers that a decentralized channel can generate more profit than the centralized channel when robust pricing strategies are implemented.
While community detection techniques reveal latent structures in complex networks, they pose privacy leakage risks. To achieve privacy protection, researchers have focused on community hiding methods that disrupt community structures through network structural perturbations with minimal costs. However, existing topology perturbation-based approaches fail to effectively characterize intrinsic network relationships due to their neglect of node multi-dimensional attribute features. To address this, this paper innovatively constructs an Attribute-enhanced Stochastic Block Model (AESBM) and proposes a Community Hiding Algorithm based on AESBM (HC-AESBM). First, by jointly modeling network topology and node attributes, an Expectation-Maximization algorithm iteratively estimates the node membership matrix, connection probability matrix, and attribute correlation matrix. Second, an edge probability generation model incorporating attribute similarity is established to reconstruct network structure formation mechanisms. Finally, precise perturbations are implemented via a critical edge identification strategy to conceal community structures. Comprehensive experiments on real-world network datasets demonstrate the effectiveness of the proposed method. Results show significant advantages in community hiding metrics: Normalized Mutual Information (NMI), Adjusted Rand Index (ARI), and Jaccard similarity coefficient decrease by 12.3
Stock trend prediction remains a challenging task due to the complex interplay of multi-source information and inherent non-stationary characteristics of financial markets. To address the limitations of existing methods in handling noisy textual data and capturing multi-scale temporal patterns, this paper proposes a novel news-driven framework that synergizes structured news representation with adaptive time-frequency analysis. Specifically, we classify financial news into local and global news, and design a sparse attention module to capture both fine-grained local news correlations and macro-level global interactions. Then a cross-attention mechanism is introduced to dynamically modulate local news features under global contextual guidance. To handle non-stationary price sequences, a wavelet transform attention module is proposed to decompose stock technical indicators into high-frequency fluctuations and low-frequency trends, enabling the extraction of transient signals and long-term dependencies through a dual-path attention mechanism. The framework further employs cross-modal attention to fuse news and price features, and a temporal aggregation module to model temporal dependencies. Experimental results on our self-constructed A-share datasets demonstrate that our model outperforms state-of-the-art baselines, highlighting its effectiveness in resolving the ambiguity of traditional text representations and improving prediction performance in low signal-to-noise ratio price sequences.
In customs supervision scenarios, import and export cargo data exhibits significant non-uniform distribution characteristics due to variations in cargo types and seasonal fluctuations. Traditional anomaly detection methods, which rely on uniform distribution assumptions, face performance limitations. To address this critical issue, this paper proposes the HADN framework, which enhances anomaly detection accuracy for non-uniform time series through density-adaptive preprocessing and multimodal feature modeling. The core innovations of this study include: (1) A dynamic density discrimination mechanism based on variance analysis, enabling real-time partitioning of sparse and dense subsequences; (2) A dual-path feature processing architecture combining Fourier decomposition with resampling (for periodic dense data) and context-aware imputation (for sparse data with semantic mutations); (3) A local-global representation contrastive decision module that optimizes anomaly criteria through multidimensional feature alignment. Existing methods primarily designed for uniform distribution scenarios struggle to address feature drift caused by density heterogeneity. Experimental results demonstrate that HADN achieves a 1.62
Fog computing serves as an intermediary between cloud computing and end users, enhancing latency and Quality of Service (QoS) by deploying resources at the network’s edge. The challenge of efficient service placement in fog environments arises from resource heterogeneity, network instability, and task dependencies. This paper introduces a hierarchical service placement framework based on deep reinforcement learning (HA3C). The framework employs a hierarchical decision mechanism to divide the service placement problem (SPP) into two subproblems: community selection and node selection. Our approach incorporates a dynamic multi-objective reward function, introducing a Load Balancer index while optimizing service delay and resource utilization. Extensive simulations on the iFogSim platform reveal that our method significantly outperforms basic algorithms in fog node placement rate while maintaining moderate to high resource utilization. In particular, HA3C exhibits enhanced robustness and scalability in large-scale tasks and complex network topologies. The experimental results confirm the efficacy of our method in optimizing service placement decisions in fog environments.
Cloud-network convergence, as an emerging technology in the service computing domain, has created novel development prospects for network services. Understanding user intent constitutes a key component in this technological framework, yet existing research efforts remain substantially inadequate in addressing this critical aspect, necessitating in-depth investigation. This study therefore focuses on the analytical investigation of user service intent data within cloud-network convergence environments, establishing a comprehensive service intent interpretation and translation framework to facilitate the proactive transformation of Internet-related services. To address the challenge of directly acquiring user intent information identified in current research, this paper proposes an integrated solution framework. Specifically targeting the prevalent issue of data scarcity in network intent datasets within cloud-network convergence systems, the present research develops a novel dataset generation scheme for user service intent characterization.
With the continuous development of deep learning in the field of speech, intelligent applications such as speech recognition have become indispensable in human life. But the training of high-performance speech recognition models requires extensive hardware resources, making the constructed models likely to be deployed as services. Some research has shown that speech recognition services are vulnerable to backdoor attacks, where adversaries poison the training process to implant malicious behavior into the service. Research on backdoor attack methods not only helps reveal potential security risks, but also provides valuable insights for developing effective defense mechanisms. However, current backdoor attacks targeting speech recognition still face the issue of the trigger design being insufficiently stealthy. To address the problem, we propose a phoneme segmentation based backdoor attack method. Firstly, we segment audio samples at the phoneme level and enhance the amplitude of phoneme segments. Then, the frequency intensity is calculated to select the optimal frequency for embedding trigger. Finally, we generate a pure tone trigger with the optimal frequency, which is masked within the amplitude-enhanced regions, enhancing stealthiness by blending the trigger with natural variations in speech. The experimental results show that our method achieves an attack success rate greater than 94
With the explosive growth of satellite-based remote sensing missions, the image data generated on satellites has exponentially increased. However, limited space-to-ground communication bandwidth is a critical bottleneck, hindering swift transmission of large volumes of valuable image data to ground stations. Environmental noise during observations further degrades data quality, reducing its practical utility. Consequently, efficiently selecting and prioritizing satellite image transmission based on ground station queries has become a key challenge in space information retrieval. In this paper, we introduce a groundbreaking Space-Oriented Image Retrieval (SOIR for short) task, aimed at retrieving high-quality satellite images using ground-transmitted queries. To tackle this task, we categorize common satellite noise perturbations into two types: coarse and fine-grained perturbations. Based on this, we employ a sophisticated data augmentation strategy that constructs different types of perturbation masks to simulate varying levels of perturbation. After this, we introduce a Multi-task based Dual Generative Model (DGM for short) designed to rapidly filter high-quality satellite image data in response to image queries, significantly improving data retrieval and transmission. We conduct comparative experiments against a range of established image retrieval models. The experimental outcomes reveal that DGM offers substantial benefits in both retrieval efficiency and effectiveness.
Recent decades have witnessed the prosperous development of the Internet of Things (IoT). To handle the bottlenecks of the traditional cloud-based IoT framework, edge computing is widely introduced in IoT, to provide cloud services for terminal devices in a distributed and low-latency manner, forming a cloud-edge-terminal collaborative architecture. The heterogeneity, unreliability, and mobility of both IoT and edge devices raise great security and privacy concerns about cross-layer communications and collaborations, highlighting the necessity for a cross-layer authentication scheme. Existing authentication approaches suffer from heavy key management overheads and the need for a centralized authority. Besides, the current works mainly focus on terminal authentication while neglecting edge-terminal integrated authentication. To bridge these gaps, we propose a blockchain-based cross-layer IoT authentication scheme. Specifically, blockchain is utilized to construct trust among different entities and provide a platform where formatted transactions are broadcast and stored. Identity-based cryptography (IBC) is exploited for mutual authentication among edges and between edge and IoT devices. Experimental evaluation based on a prototype is conducted to show the effectiveness and efficiency of the proposed scheme.
