Inaccurate pharmaceutical demand forecasting undermines hospital inventory management, causing drug shortages, overstocking, and avoidable logistics costs. Classical “World Healthcare Organization” methods, particularly the consumption method, rely solely on historical consumption data and cannot account for the operational context driving demand. This paper presents a machine learning-based causal forecasting framework applied to monthly medicine consumption data from five French hospitals. Using twelve operational features, including patient volumes, bed capacity, and staff levels, Random Forest and Artificial Neural Network models are trained and tested on ten medicines. Random Forest achieves a mean R² of 0.81 on the test set and substantially outperforms the WHO consumption method across all medicines. These results demonstrate the operational feasibility and accuracy gain of machine learning-based causal forecasting for hospital pharmaceutical supply chain management.

Efficient ambulance dispatch is essential for timely emergency medical care under limited fleet capacity and uncertain patient demand. This paper presents an event driven semi Markov decision model in which dispatch decisions are triggered by patient call arrivals or newly idle ambulances. The model allows strategic ambulance withholding and ambulance redirection. To solve the problem, we develop a transformer reinforcement learning framework that learns contextual ambulance and patient location assignments through attention based communication. We further distill the learned policy into a surrogate decision tree to support explainable dispatch recommendations. Experiments on synthetic cases and Toronto paramedic data show that the proposed approach reduces patient waiting and improves ambulance operating efficiency.

Hybrid outpatient clinics must expand access while maintaining reliable operational performance. This is difficult when telemedicine visits share clinical capacity with in-person care and are affected by infrastructure disruptions. Unlike service time variability, telemedicine disruptions can add consultation time or prevent visits from being completed. This paper studies how disruption conditions and appointment slot length affect such systems. An analytical model is developed to separate added workload from incomplete telemedicine visits and to derive a reference slot length for a given disruption setting. The mechanism is tested using a discrete-event simulation of a hybrid outpatient clinic with scheduled patients, walk-ins, emergencies, no-shows, and carryovers. Results show that short disruptions increase overtime and utilization, while longer disruptions reduce congestion because some telemedicine visits are not completed. Longer slots reduce waiting, but increase carryover by reducing daily appointment capacity. Resilient hybrid outpatient scheduling, therefore, requires balancing congestion, capacity, and telemedicine appointment completion.

Healthcare systems face persistent challenges in delivering consistent service quality despite rapid technological transformation. Traditional Healthcare Service Quality (HCSQ) models treat quality dimensions descriptively and assume equal importance, limiting structured prioritization in technology-enabled settings. This study proposes a Relational Framework for HCSQ (RF-HCSQ) and Decision Support Tool (DST) integrating validated quality domains and Key Performance Indicators (KPIs) with hybrid multi-criteria weighting. A mixed-method approach combined expert interviews, a three-round Delphi validation, a practitioner survey (N=127), and a patient survey (N=206). Ten domains and 47 KPIs were validated and weighted via hybrid techniques integrating the Analytic Hierarchy Process (AHP), Arithmetic Mean aggregation, and Shannon entropy, while patient priorities were modeled using the Plackett-Luce method. Bootstrap resampling confirmed robustness. Findings identify Patient Safety as the most influential domain and Clinical Quality as the second, with Technology Impact functioning as an enabling rather than outcome-driving domain. A preliminary end-user evaluation (n=7) indicated that the DST supports structured prioritization and actionable performance review.