This paper tackles a challenging Integrated Surgery Planning and Scheduling Problem that simultaneously considers operating rooms and recovery beds. The objective is to minimize the maximum daily closing time of the operating theatre, a key metric for optimizing hospital resource utilization and improving the quality of patient care. To address this NP-hard problem, we propose using the Red Deer Algorithm (RDA), a recently developed evolutionary metaheuristic inspired by the mating behavior of Scottish Red Deer during their breeding season. To evaluate the effectiveness of the proposed approach, computational experiments were carried out on a series of benchmark instances. The results highlight the effectiveness of the RDA in generating high-quality surgical schedules, contributing to improved operational efficiency and enhanced patient outcomes..

Efficient operating room (OR) management depends on the accurate prediction of surgical procedure durations to improve scheduling, enhance patient outcomes, and reduce operational costs. This study presents a hybrid machine learning model that combines Random Forest and K-Means clustering to predict the duration of cholecystectomy procedures. The model is trained using real-world data from the digestive surgery department at Mahmoud El Matri Hospital in Tunis, Tunisia, incorporating patient demographics, surgeon experience, and other contextual factors. Synthetic data generation was also applied to reinforce model reliability. The proposed approach achieved strong performance, with a root mean square error (RMSE) of 0.45 minutes, a mean absolute error (MAE) of 0.36 minutes, and a coefficient of determination (R2) of 0.99. Comparative analysis with individual models such as Random Forest, K-Means, decision trees, and linear regression confirms the hybrid model’s superior predictive capability. These results demonstrate the potential of the proposed hybrid model as a practical tool for optimizing OR scheduling and improving healthcare resource management.

Deep Reinforcement Learning (DRL) has showcased remarkable achievements across various domains, such as image recognition and automation. Nevertheless, its potential in the realm of logistics and transportation, particularly in tackling routing challenges, remains mostly untapped. On the contrary, Evolutionary Algorithms (EA) have enjoyed widespread adoption for solving combinatorial optimization problems. Unexpectedly, the combination of EA and DRL methods for tackling combinatorial optimization problems has not been extensively explored in the current body of literature. Driven by these gaps in research, this study presents a novel method called Evolutionary Reinforcement Learning (ERL) aimed at addressing the Traveling Salesman Problem (TSP). To enhance the policy generated by a deep neural network, we exploit the collaborative potential between the EA and DRL frameworks. Significantly, the weights linked to the actor component are crucial, especially in approaches that are not primarily focused on policy optimization. By harnessing the capabilities of EA, we establish a weight population and seamlessly integrate them into the DRL framework, aiming to substantially improve TSP results. Employing the Genetic Algorithm (GA) as our EA, we introduce a novel ERL-based approach, specifically the ERL-GA. Computational experiments conducted reveal that the ERL-GA outperforms the basic DRL framework in terms of performance.

This triangulation design-based study investigates the benefits and challenges of telemedicine adoption among 2875 patients and caregivers above 18 years old after the Covid-19 pandemic using a cross-sectional survey. In the quantitative part, we run logistic regression models to identify the predictors of the behavior (intention of telemedicine use after Covid-19) following the health belief model. For the qualitative part of the study, we use thematic analysis to identify the benefits and challenges of the same behavior. Positive experiences, convenience of telemedicine, unsafety during in-person visits, and extensive use of telemedicine during Covid-19 were positively correlated with the user’s behavior. In addition, insurance coverage of online visits and the safety of home-based care encourages patients to continue trusting telemedicine. Through the qualitative analysis, we found that factors encouraging patients and caregivers to continue using telemedicine after the pandemic are safety, access to care, convenience, trust in technology, and system and technology-related factors. The challenges of telemedicine adoption post-Covid-19 era included systems-related limitations, work environment issues, trust, and communication issues. More efforts must be made to improve telemedicine design and healthcare infrastructure to align with telemedicine requirements. Policies must consider the regulations’ updates needed to ensure successful telemedicine adoption.

The healthcare sector aims providing patients with the best quality of care. However, healthcare supply chain management is costly. In this paper, we focus on inventory drugs management. The main major problems regarding inventory management practice in healthcare system are: drugs shortage, overstock, unfunded forecasting technique due to the lack of drugs consumption information and absence of IT support as Decision Support System. In order to address this concern, this paper presents a real case study of a drug inventory management problem within the emergency service of a Tunisian Hospital: Charles Nicolle Teaching Hospital. The indicated service aims to optimize the drug stock management and procurement while reducing drug shortage and stock management costs. Precisely, we have investigated a set of the real hospital dataset and, in particular, the emergency service's pharmacy. This analysis phase has yield to develop a drugs classification and to establish adequate inventory policy for each drug family. The implementation of the developed policies reduces significantly the stock management costs. Then, a Decision Support System (DSS) was implemented to assists users to improve the drugs inventory management as well as to make reliable decisions by taking the adequate action at the right moment.

We investigate the Network Loading Problem (NLP) where a set of multicommodity demands (traffic) should be routed between pairs of nodes (customers). The problem consists of designing a network by installing capacities that allow simultaneous transmission of all the point-to-point demands with a minimum cost. In recent years, this challenging NP-hard problem has caught the interest of professionals and scientists, as it remains of crucial importance in telecommunications field. In order to solve the NLP to optimality, we use a path-based formulation to develop a tailored Benders decomposition scheme combined to a column generation approach for solving the master problem. Then, to accelerate the solution of the relaxed master program, we dynamically derive efficient cutset inequalities using a new max-cut-like integer programming model. To evaluate the efficacity of the proposed algorithms, computational results are reported using Benchmark instances from SNDLib.

We aim to derive effective lower bounds for the Discrete Cost Multicommodity Network Design Problem (DCMNDP). Given an undirected graph, the problem requires installing at most one facility on each edge such that a set of point-to-point commodity flows can be routed and costs are minimized. In the literature, the Lagrangian relaxation is usually applied to an arc-based formulation to derive lower bounds. In this work, we investigate a path-based formulation and we solve its Lagrangian relaxation using several non-differentiable optimization techniques. More precisely, we devised six variants of the deflected subgradient procedures, using various direction-search and step-length strategies. The computational performance of these Lagrangian-based approaches are evaluated and compared on a set of randomly generated instances, and real-world problems. The empirical results show that the Lagrangian relaxation of the path-based formulation requires less computation time than the arc-based formulation.