The extreme congestion in complex healthcare systems, particularly in hospital emergency departments (ED), often leads to catastrophic rejections of high-risk patients. To address this critical resource allocation challenge, we propose a novel data-driven Automated Admission Control framework based on a Dynamic Threshold Downgrading Policy (DTDP). Modeled via a Markov Decision Process (MDP), our approach systematically automates the decision-making process for multi-tier bed allocation. Unlike traditional studies relying on synthetic parameters, this paper pioneers the calibration of the MDP environment using empirical data from the National Hospital Ambulatory Medical Care Survey (NHAMCS) administered by the U.S. CDC. By extracting precise arrival ratios and service rates, we capture the authentic capacity friction between standard and high-risk patients. Extensive simulation results demonstrate that the proposed automated DTDP algorithm exhibits a robust switching curve structure. Compared to the traditional First-Come-First-Served (FCFS) heuristic, our data-driven control policy achieves a 42.4% reduction in total penalty costs while simultaneously eliminating 87.5% of catastrophic high-risk rejections under severe demand surges. These findings provide highly actionable deployment guidelines for automated hospital information systems.

The rapid growth in private car ownership has exacerbated social and environmental problems. A carsharing system offers an effective solution to mitigate these problems. However, a one-way carsharing system can cause a vehicle imbalance. This study examines the operation of a multi-region one-way station-based carsharing system using staff relocation. We consider the percentage of demand served and revenue as performance measures and analyze the influence of several dynamic operator-based relocation policies. A case study of 1020 vehicles at 273 stations in Vancouver, Canada, is used to test the policies. The results indicate that the rolling-horizon policy significantly enhances the percentage of served demand and system revenue, which can assist managers in scheduling relocators.

Accurate short–horizon forecasts of the queue length at downstream merge stations—where multiple upstream desks converge—are vital for dynamic triage, load balancing and anticipatory staffing in outpatient clinics. We model patient arrivals to each upstream desk as a non-homogeneous Poisson process and encode real-time queue states into a compact feature vector. Within this supervised-learning framework we benchmark five algorithms—k-nearest neighbors, random forest, Bayesian ridge regression, single-layer perceptron and multi-layer perceptron (MLP)—for one-step-ahead prediction of the merge-station queue length. The results indicate that a lightweight MLP consistently delivers the best accuracy across all load segments and markedly reduces under-forecasting during peak periods, while deeper or wider networks offer diminishing returns. The proposed predictor operates with modest computational overhead and can be embedded in real-time routing and staffing policies to shorten queues and enhance service consistency during congestion.

The cement industry is a significant contributor to global environmental impacts, and Life Cycle Assessment (LCA) has emerged as a critical tool for evaluating and managing these burdens. This review uniquely synthesizes recent advancements in the LCA methodology and provides a detailed comparison of cement production impacts across major producing regions, notably highlighting China’s role as the largest global emitter. It covers the core LCA phases, including goal and scope definition, inventory analysis, impact assessment, and interpretation, and emphasizes the role of LCA in quantifying cradle-to-gate impacts (typically around 0.9–1.0 t CO2 per ton of cement), evaluating the emissions reductions provided by alternative cement types (such as ~30–45% lower emissions using limestone calcined clay cements), informing policy frameworks like emissions trading schemes, and guiding sustainability certifications. Strategies for environmental load reduction in cement manufacturing are quantitatively examined, including technological innovations (e.g., carbon capture technologies potentially cutting plant emissions by up to ~90%) and material substitutions. Persistent methodological challenges—such as data quality issues, scope limitations, and the limited real-world integration of LCA findings—are critically discussed. Finally, specific future research priorities are identified, including developing country-specific LCI databases, integrating techno-economic assessment into LCA frameworks, and creating user-friendly digital tools to enhance the practical implementation of LCA-driven strategies in the cement industry.

This study introduces a specialized Artificial Neural Network (ANN) modeling approach to accurately forecast nitrogen oxides (NOx) and particulate matter (PM) emissions in cement kiln operations. Drawing upon 73 carefully curated input parameters-spanning raw material consumption, production rates, intricate chemical compositions, and diverse operational indicators-the methodology encompassed rigorous data preprocessing steps, including outlier exclusion, normalization, and feature selection. The resulting ANN-based models demonstrated commendable predictive prowess, achieving SMAPE values of 5.32 % for NOx and 6.26 % for PM, accompanied by R2 scores of 0.84 and 0.68, respectively. These metrics attest to the models' ability to capture subtle nonlinear interdependencies governing emission dynamics. When benchmarked against three established machine learning counterparts (XGBoost, Linear Regression, and Random Forest), the ANN consistently surpassed them, revealing superior adaptability and accuracy. Further enhancing its practical utility, a sensitivity analysis uncovered a set of pivotal factors wielding pronounced influence on emission outcomes, thereby guiding strategic adjustments for improved environmental stewardship. By weaving together advanced analytical techniques, robust data preparation, and astute parameter prioritization, this work not only underscores the potency of ANN-driven solutions within the cement industry's complex landscape, but also lays a scalable foundation for informed, proactive process management and heightened sustainability in similar industrial contexts.

This paper proposes a reinforcement learning-based approach to address the patient overflow problem in hospitals. We model the patient diversion decision-making process as a Markov Decision Process (MDP) and employ the Proximal Policy Optimization (PPO) algorithm to find an optimal policy. To handle the cyclical nature of patient flow, we introduce a novel temporal representation using sine and cosine encoding for the time of day. Furthermore, we enhance the stability and performance of the learning process by implementing orthogonal initialization of the neural network weights and normalizing the state inputs using a running mean and standard deviation. The agent's state is defined by the current patient count, queue length, and number of patients awaiting discharge across five different departments. The actor network outputs the probability of transferring an overflowing patient to other permissible departments, while the critic network evaluates the state value. Both actor and critic are modeled as Multilayer Perceptron (MLP) networks. The objective is to minimize a cost function composed of waiting costs for patients remaining in their original, full departments and overflow costs for patients transferred to other departments. Experimental results demonstrate that our proposed method can effectively learn a diversion policy that reduces overall costs and improves hospital operational efficiency.

Surgical cases assignment problem (SCAP) is among the most investigated interests in operating room planning. The studies related to the SCAP are mainly focused on the single objective optimization. To concern multiple criteria in the management, this paper reformulates a multi-objective SCAP with minimizing total operating cost and maximizing scheduled surgeries number, simultaneously. Although dozens of multi-objective methods have been introduced in recent years, these cannot be applied into the multi-objective SCAP directly considering the characteristics of the problem. To tackle the multi-objective SCAP, a problem-specified multi-objective squirrel search algorithm (MOSSA) is presented. First, a targeted single-list scheme is devised to encode individuals and an efficient decoding scheme integrating a repair strategy is proposed to construct feasible scheduling plan. Second, two simple heuristics are incorporated into the algorithm to imitate the behaviors of squirrels and improve the search ability. Third, an addition archive is employed to store the non-dominated solutions. Moreover, a DOE method is performed to investigate the influence of parameter settings. The performance of the MOSSA is assessed by a representative benchmark dataset with three different evaluating indicators. Extensive experiments and analysis results demonstrate the effectiveness of the proposed algorithm and the superiority over compared approaches in addressing the multi-objective SCAP.

Download This Paper Open PDF in Browser Add Paper to My Library Share: Permalink Using these links will ensure access to this page indefinitely Copy URL Copy DOI

Carsharing systems have attracted considerable market interest owing to their positive impact on resolving social and environmental pollution problems such as traffic congestion, lack of parking spaces, and greenhouse gas emissions. However, one-way carsharing systems often struggle with vehicle imbalance issues, which can affect customer satisfaction. To address this, a one-way electric carsharing model was designed to maximize operator revenue by introducing customer space flexibility. A mixed integer nonlinear program was formulated to optimize consumer flexibility, relocation, and employee-hiring decisions. The variation in the average vehicle charge at each station was captured using the state of change (SOC) model. Real data from the Evo carsharing company in Vancouver, Canada, were used to conduct numerical tests, verifying the effectiveness of the rolling horizon method in producing high-quality solutions. The results indicate that customer space flexibility can enhance the operational efficiency of carsharing systems without increasing the number of vehicles or parking spaces. This approach can also boost total system revenue and customer satisfaction, with departure space flexibility showing a significantly greater impact than arrival space flexibility.

Life cycle assessment (LCA) is an effective tool to quantitatively assess the environmental impacts of cement products in the whole life cycle. The establishment of the indigenized and recognized weight coefficients of LCA plays an important role in promoting the application and development of LCA in China's cement industry. An online questionnaire survey was carried out across the cement academia and industry of China, and a total of more than 200 valid samples were collected. Descriptive analysis, Chi-square test, multiple response analysis and Pareto chart were then conducted to analyze the real application situation of carbon footprint accounting, LCA and carbon reduction measures in the cement industry. Based on the scoring by senior experts and feedback from cement academic and industry, the analytic hierarchy process was used to determine the weight coefficients of LCA. The findings suggested that the most effective measures to reduce carbon emissions in the cement industry included development of low-carbon cementitious materials, fuel substitution, utilization of alternative raw materials for clinker production, application of carbon capture-utilization-storage (CCUS) and so on. For the cement production stage, the clinker calcination was considered to have the largest potential contribution to reducing the overall environmental impacts of cement, accounting for 39.04%. Further calculation indicated the weight coefficients of abiotic depletion potential (ADP), global warming potential (GWP) and human toxicity potential (HTP) were relatively large, which were 0.3005, 0.2663 and 0.2502, respectively.

通过对生命周期评价在水泥生产中的应用研究进行梳理和归纳,总结当前研究取得的成果、存在的问题以及未来的研究方向.生命周期评价主要包括目标和范围定义、清单分析、影响评价以及结果解释四个阶段.当前研究主要存在数据来源不清晰、排放统计不完善等问题,建立本土水泥生产数据库、评估数据质量、开发适用于我国国情的生命周期评价软件等是未来值得关注的研究领域.

Based on the transmission mechanism of the COVID-19 and the time-varying characteristics of the patient population,an adaptive susceptible infected recovered model that considered critically ill patients was constructed to simulate the spread of the COVID-19 in Wuhan.On this basis,a dynamic medical resource allocation model was established,an improved immune algorithm was designed to solve the problem,and real data from Wuhan was used for verification. The results show that the model optimizes the allocation of medical staff,and the improved greedy strategy can improve service levels and reduce the number of dead. The parameter sensitivity analysis shows that the hospital’s service level is highly correlated with the emergency response time and the number of medical staff.

The surgical case assignment problem (SCAP) is vital to the operating room planning problem. Although several methods have been applied, the solution accuracy can be improved further. In this paper, an efficient genetic programming hyper-heuristic (GP-HH) algorithm is proposed for the SCAP to minimize the total operating cost. First, eight simple and adaptive heuristic rules are devised to constitute a set of low-level heuristics (LLHs). Second, genetic programming is employed as a high-level heuristic to dynamically manage LLHs applied to the solution domain. Third, effective solution encoding and the corresponding decoding schemes are developed to represent individuals and construct valid schedules. To investigate the influence of parameter settings, we performed a design-of-experiment (DOE). The effectiveness of GP-HH is executed on a typical benchmark dataset. The experimental results demonstrate the superiority of the proposed GP-HH scheme over existing approaches.

骨盆骨折是一种高能量创伤且通常伴有多发伤,手术是主要的治疗方法,本文制定并优化创伤患者的手术分配计划.首先根据患者入院时的生命稳定状态,将手术患者分为恢复期和计划期两类;针对两类患者的不同特点,为恢复期患者设置随机恢复时间,以最大期望收益为目标建立马尔可夫决策过程模型;根据医院实际情况设计实验,采用后向迭代算法求解得到最优分配策略;改变惩罚函数形式和恢复期患者数量,制定不同场景的分配策略,提高医疗资源利用率.采用二次惩罚函数时,最优分配曲线呈现开关曲线形式;恢复期患者数量越多,其享有的优先权越高.

The rapid increase in the ownership of private vehicles has exacerbated social and environmental problems, such as traffic congestion and exhaust emissions, and carsharing services are regarded as an effective method for solving these problems. In a one-way carsharing system, customer demand is highly variable, and the volume of demand in different locations changes continuously. As a result, the facility planning for a one-way carsharing system is a difficult issue for the managers. In this study, a dynamic optimization method based on a mixed integer programming model is proposed to solve this issue. In the model, various costs, such as station construction, parking space operation, vehicle maintenance, and vehicle rescheduling, are considered, and the objective of optimization is to maximize the profits of the operators. To solve the proposed model, an adaptive Jaya algorithm is proposed. Finally, the case of Vancouver, Canada, is used to verify the effectiveness of the proposed model, and the experimental results show that compared with those of the traditional single-period static planning method, the dynamic planning method proposed in this study can increase the average revenue of the operators by 10.8% and the average customer service rate by 2.95%.

Pelvic fracture is a severe trauma and is often seen in the traffic accidents, which are associated with complications or multiple injuries. Surgery is the main treatment for patients with serious conditions, while conservative treatment is adopted for older or minor-illness patients. Surgery resources, such as doctors, nurses, and operating rooms, are shared by all pelvic fracture patients. From the perspective of patient state, this paper divides patients who require surgery into two types, convalescent patients and scheduled patients. Convalescent patients' life states are always unstable, and they require recovery time to meet the condition of surgery. The recovery time is usually stochastic due to different patient situations. Scheduled patients have stable life states, and the pelvic fracture surgical plan is scheduled days or weeks in advance. Considering the characteristics of the two types of patients, a finite-horizon Markov decision process (MDP) model is established. With data collected from the hospital, parameters are set and experiments are designed to reveal the dynamic priority rules for receiving patients into surgery. Performances of different scenarios are compared, and the optimal policies obtained from the MDP are analyzed.

Surgical cases assignment problem (SCAP) is among the most investigated parts in healthcare scheduling and assignment problems, in which a set of surgical cases are assigned to operating rooms within a specified planning horizon. Several methods have been developed to provide approximate solutions for SCAP. Nevertheless, existing methods underperform at the large-scale instances. In this paper, a discrete squirrel search algorithm (DSSA) is proposed for SCAP with the objective of minimizing total operating cost. First, four heuristics are presented to improve quality and diversity of initial population. Second, a surgical case sequence vector is employed to encode individuals, and a corresponding decoding scheme is designed to construct feasible schedules. Third, several efficient heuristics are embedded into DSSA to enhance the search capacity. Moreover, the Taguchi method of design-of-experiment (DOE) is adopted to explore the influence of parameter settings. To the best of our knowledge, it is the first application of the squirrel search algorithm for SCAP. The effectiveness of DSSA is conducted on a typical benchmark dataset. Computational results and comparisons demonstrate the superiority of the proposed scheme over the existing methods in solution accuracy and consuming time for solving SCAP.