Drug shortages mitigation of supply chain in the Canadian hospital pharmacy

Drug shortage is a complicated issue worldwide, which also causes several negative impacts to the whole Canada’s pharmaceutical supply chain because of various factors that include an unforeseen increase in demand, trouble obtaining direct raw materials or labour, sudden production problems, sole sourcing manufacturing issues, legislative and regulatory problems, distribution factors, and natural disasters. Furthermore, the existing inventory management with manual operation and inventory management in the hospital pharmacy cannot prevent drug shortage. In fact, this issue is very critical in the Canadian healthcare system which is recently calling several investigation and research to mitigate negative impact and risk to the health care system. \n \nIn order to highlight the critical factors which could lead to the drug shortages and affecting the supply chain and the inventory management strategies in Canadian hospital pharmacies, our thesis is firstly to contribute a whole picture of the Canadian hospital pharmacies by presenting an analysis of Drug Shortage in Canada from 2016 –2021 and a comprehensive Systematic Literature Review (SLR) to extract the critical factors performing a wide review of the Canadian hospital pharmacies and to understand how the continued perturbation occurred in this process and affected to the drug shortages. We have used the open data from Canadian Medical Association (CMA) database and analytical stochastic methods to illustrate the survey result. \n \nBased on the intensive survey, we next go in the practical model of drug shortage in the Hospital Supply Chain (HSC) in which we contribute an optimization model to avoid the medicine shortage problem in a hospital and propose a learning method to automatically manage the inventory. Specifically, a Deep Reinforcement Learning (DRL) mechanism is designed based on the optimization model to operate the inventory under an online fashion in which refilling drug decision is automatically determined based on the observation of price, demand, current level of drugs. A penalty cost is also added in the objective function to minimize not only the drug shortage issue but also the over-provisioning situation. \n \nFurthermore, In this thesis, a numerical result has been presented to verify the performance of the proposed approach which outperforms the online benchmarks such as (Over-provisioning, Ski-rental, and Max-min) in terms of the refilling cost and the shortage rate. \n \nTo conclude, we focus on the medication shortage in Canada in this thesis to provide an extensive analysis over a lengthy period of time, from 2016 to 2021. We contribute to the study of this topic by using an actual data set and stochastic analytical approaches. We propose an intensive literature study as a follow-up to the analysis to sketch the drug shortage picture of the hospital’s pharmacy supply chain. The findings of the study and survey will aid us in furthering our research by constructing an inventory management optimization model. Besides that, and based on DRL, a deep learning strategy was developed to manage autonomously inventories in the hospital pharmacy in order to prevent drug shortages.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.