Proposta de dissertação do MEI Título: Predictive Algorithms for Managing Community Pharmacies Inventory Proponente(s): Jorge Cruz Marco Correia (ANF) Créditos: 42 ECTS Área científica: Decision Support and Artificial Intelligence Início preferencial: 1º Semestre URL: Já estão em curso trabalhos preliminares executados pelo alunos: Breve descrição: Managing a pharmacy shop involves making purchasing and replenishment decisions on a daily basis. Factors such as product demand, seasonality, supplier lead time, or product reduced availability must be taken into consideration for an optimal inventory management. The goal of this project is to develop a reliability optimization algorithm for computing the stock levels of each product (how many units of it should be on stock at a given time) that results in the best compromise between overestimation and underestimation, assessed with a predefined loss function. To model uncertainty, predictive modelling approaches will be used based on the available product sales history. Successful developments will be integrated with an inventory management software product to be deployed in Portuguese community pharmacies. Observações: Thesis carried out in business collaboration with Associação Nacional de Farmácias (ANF) with the joint supervision of Marco Correia and supported by a scholarship in an exclusive regime.