The growing impact of pandemics and infectious disease outbreaks has highlighted the need for vaccine allocation strategies that balance risk-profile protection, equity, and operational feasibility under limited healthcare resources. However, many existing approaches rely on predefined population groups and do not sufficiently integrate data-driven risk-profile information into constrained allocation planning. To address this issue, this study proposes an equity-aware decision-support framework that combines machine learning-based risk-profile stratification with multi-objective vaccine allocation. Publicly available French COVID-19 hospital-surveillance data are reorganized into analytical records to construct operational risk-profile classes for allocation-scenario analysis. A Light Gradient Boosting Machine model classifies these records into ordered risk-profile groups, which are then incorporated into a constrained allocation model. The model aims to maximize protection of higher-priority risk profiles, promote equity across predicted risk-profile groups, and minimize vaccination delays under supply and capacity constraints. The resulting optimization problem is solved using a binary Particle Swarm Optimization algorithm with constraint-handling mechanisms. Computational experiments assess algorithmic performance under a common objective-evaluation budget and examine the repair strategy, policy-weight configurations, Pareto-based compromises, classification uncertainty, scalability, and resource-capacity sensitivity. Overall, the framework supports the exploration of risk-profile-based vaccine allocation policies under constrained pandemic-response settings.

In this paper, we introduce a global framework integrating predictive analytics and multi-objective optimization for the purpose of home healthcare logistics optimization. First, several machine learning approaches such as Multinomial Logistic Regression, Support Vector Machines, Random Forest, AdaBoost, and Gradient Boosting are implemented to predict and classify patients' care requirements. This categorization not only separates professional-grade nurses from primary-grade nurses but also decides whether one caregiver or two caregivers are to be deployed depending on the condition of the patient (bedridden or semi-dependent). Secondly, we create a Multi-Objective Vehicle Routing Problem with Time Windows (MOVRPTW) to schedule the caregivers efficiently and reduce transport costs. Since the corresponding optimization problem is NP-hard, we take two advanced genetic algorithms Non-dominated Sorting Genetic Algorithm II (NSGA-II) and Strength Pareto Evolutionary Algorithm 2 (SPEA2) to find good-quality solutions. To solve the problems of bedridden patient care with multiple visits per day, our model incorporates synchronization constraints to ensure continuity of care and coordination among single-caregiver teams in case dedicated double teams are not possible. By combining predictive analytics with strong optimization techniques, our framework not only improves resource allocation effectiveness and facilitates timely service delivery but also decreases operating expenses, thus providing a holistic solution to the changing needs of home healthcare logistics.

As the need for emergency care services increases, healthcare facilities are recognizing the importance of tailored layout designs to improve patient care efficiency. Strategic layout planning is vital for managing variable productivity and meeting fluctuating demand effectively. The primary aim of tackling the emergency department layout (EDL) problem is to identify a facility configuration that satisfies both internal organizational needs and global healthcare certification standards. A novel mathematical model presented in the article offers a fresh approach to Emergency Department Layout (EDL) optimization, considering patient movement and process flow simultaneously. Our contribution lies in the development of a tailored solution to enhance the efficiency of healthcare facility layouts, setting our work apart from existing methods. The Particle Swarm Optimization (PSO) technique is proposed as a solution to the EDL problem, using a constructive heuristic to provide practical options. The technique's practical applicability is demonstrated through a real-world case study at Roanne Hospital in France, offering insights for improved healthcare delivery.

Vehicular ad hoc networks (VANETs) have attracted considerable interest in recent years, with a focus on enhancing road safety and reducing traffic accidents. The adoption of VANETs in healthcare holds great promise, especially during emergencies like accidents. The issue concerns reducing the travel time for transporting patients from their accident location to hospitals using Vehicle-to-Everything (V2X) technology. This article aims to address this issue and conduct a new study based on V2X communication technologies to analyze and reduce ambulance response times. We integrate together Vehicle-to-Vehicle (V2V) with Vehicle-to-Infrastructure (V2I) communications technologies, allowing ambulances to communicate with both other vehicles and roadside infrastructure. Collectively, these technologies can be employed to shorten the emergency services’ arrival time at accident scenes. In VANETs, connectivity to the ambulance system can be established seamlessly, enabling direct communication with the patient without the need for any other intervention. Sensors installed on the roads (Roadside Units - RSU) act as routers to transmit messages to the ambulance receiver. Furthermore, we aim to garner more attention for this field, which is expected to reshape the future of urban mobility. Additionally, we will focus on various studies related to vehicular communication issues and their resulting outcomes to conclude with an analysis aimed at enhancing patient routing management.

Operating Theater Layout Problem (OTLP) has a great impact on the productivity and the efficiency of the health process. While solving OTLP, Real-life Operating Theater (OT) sizes are larger than exact methods capacity, this lead to explore other methods as heuristics, metaheuristics or parallel treatment looking for approximate solutions. In this paper we developed a novel approach using a Multi-Agent (MA) Decision Making System (DMS) based on Mixed Integer Linear Programming (MILP) for large-sized OTLP with objective of minimizing total traveling costs. The DMS generates exact solutions in reasonable time and gives the final OT layout in a graphic interface.

The focus of this paper is on facilities with multiple sections where the material transport between sections occurs through corridors. A Mixed Integer Linear Programming (MILP) formulation for the Operating Theater Layout problem is proposed. The formulation uses a multi-goal approach to optimize two objectives: the first quantitative objective minimizes the interdepartmental traveling costs, whereas the second qualitative objective maximizes the closeness of the facilities. The presented model determines the position and orientation of each activity according to the OT international standards. The applicability of the model is demonstrated on four illustrative examples using commercial optimization software.

This paper presents an assignment problem in the home health care structures. In this problem, we search to assign caregivers to patients during a mid-term and long-term planning horizon while considering the caregivers' skills and capacity. Moreover, we take into account the randomness of thepatients' demands due to a change in their profiles or to addition of new patients through the planning horizon. As aim we balance the caregivers' workload and secure the continuity of the care. We usethe Monte Carlo method in a deterministic way to represent the randomness of the patients' demand in the mixed integer programming model we developed.

In the French context of the health expenses control, the operating theatre planning, that represents 9% of hospital's annual budget, presents a stake of first importance. The realization of the operating theatre planning is the fruit of the negotiation between the different actors of the block whose constraints and interests are often different. In this paper, we propose an algorithm of construction of an estimable planning, based on constraints satisfaction programming and whose objective is the minimization of the risk of no realization (RNR) of the planning. We present the results obtained at the time of the implementation of this model on a whole scenario of interventions to plan. Finally, we simulate the proposed planning in order to study the variations of RNR: robustness indicator and other performance productivity indicators.