Due to access and flow challenges within acute care, nearly 3,000 long-term care residents in Nova Scotia, Canada are on a waitlist to be transferred to their preferred facilities. Reasons for transferring include moving closer to families, cultural communities, and to reunite with a life partner in a different facility. We developed a new method to identify an optimal series of transfers which considers residents’ care needs (i.e. required bed type, smoking, life partner), priority level, facility preferences, and the features of a vacant bed. We model this by generating a network of every resident on the waitlist and solving it as an elementary path problem, generating a chain of transfers starting with a vacancy. We use dynamic programming to solve this problem in practice. In this talk, the mathematical models and algorithms used to generate these chains, as well as an overview of their potential impact, will be discussed.

We study the problem of designing vaccine distribution strategies that maximally mitigate the negative impact of an infectious disease outbreak. This is achieved through a multiperiod optimization-based framework that embeds important subject-specific risk and contact information into the decision-making process. By analyzing the structure of the resulting optimization problem, we identify key structural properties which we use to construct a globally convergent solution scheme (suitable for smaller problem instances) and two, more scalable, heuristic schemes. We demonstrate the benefits of the considered framework through a case study on COVID-19 in Texas. Our results highlight the importance of considering risk and contact information as doing so substantially reduces the total expected number of fatalities over conventional compartmental-based approaches. These findings indicate that customization can have a significant benefit, particularly for community-scale planning.

Most cancer patients receive chemotherapy. This type of treatment relies on dangerous and expensive drugs. The quality of drug preparation, which must be delivered in the right quantities and on time, is therefore a key point in a patient's treatment pathway. To address this, a multi-objective scheduling method has been developed for drug manufacturing in a chemotherapy clinic, to reduce patient waiting times and minimize preparation expenses. A lexicographic approach coupled with a linear programming model that approximates scheduling is used. The objective is to jointly minimize the maximum delay of a task, the associated losses of raw materials, and overall lead times during production days. The model also enables the selection of medications to be produced in advance, aiming to increase the production buffer for the following day while also minimizing raw material losses. The resulting solution is then reconstructed to clarify the scheduling details and establish a schedule that can be used by pharmacists within clinical constraints. This reconstruction makes it possible to reduce delays, select drugs for early preparation, and improve productivity while reducing costs. This scheduling method contributes to a more efficient and cost-effective workflow in chemotherapy clinics, resulting in improved quality of patient care and resource utilization.

Hospital pharmacies receive thousands of prescriptions per day, consequently, it is crucial that the hospital medication circuit be as optimal as possible. Nevertheless, hospital inventory management is highly complex for various reasons, including space limitations, demand uncertainty, and human resource limitations. Furthermore, the inventory policy must be adaptable to real-time fluctuations in patient demands, with a simultaneous emphasis on preventing backorders and stockouts, given their potential to harm patients' well-being. Additionally, it is not operationally effective for care units (CUs) to repeatedly request new supplies every day to avert situations of insufficient stock. To address this problem, we present a real-time inventory control model based on the characteristics of medication demand while considering uncertainty in demand and space limitations, with the main objective of minimizing CU replenishment frequency. First, we categorize the medications into two groups; fast-moving and slow-moving medications, then, we utilize a continuous and dynamic review inventory control policy for these categories respectively. For optimizing the inventory control parameters we propose a stochastic optimization model for each category. To efficiently solve the proposed inventory policy in real-time, we employ a receding-horizon control (RHC) strategy, where the models are solved iteratively over a predetermined time horizon. To validate the effectiveness of our proposed approach, we use a real-world inventory management setting for a hospital in Montréal and conduct a comparative analysis between the proposed model and the existing state of the inventory policy to demonstrate the advantages of this research.