Since some pharmaceutical companies made available effective vaccines to fight the spread of the SARS-CoV-2 virus, the race for vaccinations has begun, implying the development of mass vaccination campaigns in a short time all around the globe. Implementing vaccination campaigns with such an entity, however, is not straightforward and requires several strategic decisions to be taken by the authority. The last step of this decision-making process is the planning and scheduling of vaccinations. Having a huge amount of input parameters and constraining factors to consider, its proper management is complex and fundamental to optimally implement the vaccination campaign. The study presented by this thesis aims to provide support to this last phase of the vaccination decision-making process to allow the implementation of a rapid and efficient mass vaccination campaign. The proposed Mass Vaccination Campaign Planning (MaVaCaP) framework is composed of an aggregated MILP (Mixed-Integer Linear Programming) model and has the goal to conclude the campaign as soon as possible by vaccinating the entire population and minimizing its transport effort. The model is subject to the constraints deriving from the antecedent phases of the vaccination decision-making process, such as the configuration of the vaccination network (number, capacity, and location of vaccination centers), the availability of vaccines (supply and administration constraints) and the configuration of the demand to be met (geolocation and quantity). The final output reveals the vaccine batches, indicating how many people are scheduled for vaccination in each center, in each week, with which type of vaccine, and with which type of inoculation (first or second dose). Moreover, in order to make the model able to deal with uncertainties, the use of the “rolling wave” planning methodology is presented. The proposed approach is then implemented and tested in the real case of the COVID-19 mass vaccination campaign of the Lombardy Region in Italy. Finally, a scenario-based analysis is carried out to validate the model's ability to manage uncertainties. Hence, the MaVaCaP framework is an aid to authority for planning and scheduling of vaccination through-out the entire campaign to adapt and modify the decisions according to the actual plan advancement even in case of disruptive scenarios.
Da quando alcune aziende farmaceutiche hanno messo a disposizione vaccini efficaci per combattere la diffusione del virus SARS-CoV-2, è iniziata la corsa alle vaccinazioni, che ha implicato lo sviluppo di campagne di vaccinazione di massa in breve tempo in tutto il globo. L'attuazione di campagne di vaccinazione con tale entità, tuttavia, non è semplice e richiede l'adozione di diverse decisioni strategiche da parte dell'autorità. L'ultimo passo di questo processo decisionale è la pianificazione e la programmazione delle vaccinazioni. Avendo un gran numero di parametri in input e fattori vincolanti da considerare, la sua corretta gestione è complessa e fondamentale per attuare in modo ottimale la campagna vaccinale. Lo studio presentato da questa tesi si propone di fornire un supporto a quest'ultima fase del processo decisionale vaccinale per consentire l'attuazione di una campagna di vaccinazione di massa rapida ed efficiente. Il framework proposto per la Pianificazione della Campagna di Vaccinazione di Massa (MaVaCaP) è composto da un modello MILP (Mixed-Integer Linear Programming) aggregato e ha l'obiettivo di concludere la campagna il prima possibile vaccinando l'intera popolazione e riducendo al minimo il suo sforzo di trasporto. Il modello è soggetto ai vincoli derivanti dalle fasi antecedenti del processo decisionale vaccinale, quali la configurazione della rete vaccinale (numero, capacità e ubicazione dei centri di vaccinazione), la disponibilità di vaccini (vincoli di fornitura e somministrazione) e la configurazione della domanda da soddisfare (geo localizzazione e quantità). L'output finale rivela i lotti di vaccino, indicando quante persone sono programmate per la vaccinazione in ciascun centro, in ogni settimana, con quale tipo di vaccino e con quale tipo di inoculazione (prima o seconda dose). Inoltre, al fine di rendere il modello in grado di affrontare le incertezze, viene presentato l'utilizzo della metodologia di pianificazione “rolling wave”. L'approccio proposto viene quindi implementato e testato nel caso reale della campagna di vaccinazione di massa COVID-19 della Regione Lombardia in Italia. Infine, viene eseguita un'analisi basata su scenari per convalidare la capacità del modello di gestire le incertezze. Quindi, il framework MaVaCaP è un aiuto all'autorità per la pianificazione e la programmazione della vaccinazione durante l'intera campagna per adattare e modificare le decisioni in base all'effettivo avanzamento del piano anche in caso di scenari dirompenti.
Planning the COVID-19 mass vaccination campaign : problem framing and optimization model
Tedesco, Daniela
2020/2021
Abstract
Since some pharmaceutical companies made available effective vaccines to fight the spread of the SARS-CoV-2 virus, the race for vaccinations has begun, implying the development of mass vaccination campaigns in a short time all around the globe. Implementing vaccination campaigns with such an entity, however, is not straightforward and requires several strategic decisions to be taken by the authority. The last step of this decision-making process is the planning and scheduling of vaccinations. Having a huge amount of input parameters and constraining factors to consider, its proper management is complex and fundamental to optimally implement the vaccination campaign. The study presented by this thesis aims to provide support to this last phase of the vaccination decision-making process to allow the implementation of a rapid and efficient mass vaccination campaign. The proposed Mass Vaccination Campaign Planning (MaVaCaP) framework is composed of an aggregated MILP (Mixed-Integer Linear Programming) model and has the goal to conclude the campaign as soon as possible by vaccinating the entire population and minimizing its transport effort. The model is subject to the constraints deriving from the antecedent phases of the vaccination decision-making process, such as the configuration of the vaccination network (number, capacity, and location of vaccination centers), the availability of vaccines (supply and administration constraints) and the configuration of the demand to be met (geolocation and quantity). The final output reveals the vaccine batches, indicating how many people are scheduled for vaccination in each center, in each week, with which type of vaccine, and with which type of inoculation (first or second dose). Moreover, in order to make the model able to deal with uncertainties, the use of the “rolling wave” planning methodology is presented. The proposed approach is then implemented and tested in the real case of the COVID-19 mass vaccination campaign of the Lombardy Region in Italy. Finally, a scenario-based analysis is carried out to validate the model's ability to manage uncertainties. Hence, the MaVaCaP framework is an aid to authority for planning and scheduling of vaccination through-out the entire campaign to adapt and modify the decisions according to the actual plan advancement even in case of disruptive scenarios.| File | Dimensione | Formato | |
|---|---|---|---|
|
2021_10_Tedesco.pdf
non accessibile
Descrizione: Testo della tesi
Dimensione
7.61 MB
Formato
Adobe PDF
|
7.61 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/179731