Hybrid algorithm based on Reinforcement Learning and DDMRP methodology for inventory management

Vista sencilla de ítem
dc.contributor.advisorAguilar, José Lisandrospa
dc.contributor.authorCuartas Murillo, Carlos Andrés
dc.contributor.authorAguilar, José Lisandro
dc.coverage.spatialMedellín de: Lat: 06 15 00 N degrees minutes Lat: 6.2500 decimal degrees Long: 075 36 00 W degrees minutes Long: -75.6000 decimal degreeseng
dc.creator.degreeMagíster en Ciencias de Datos y Analíticaspa
dc.creator.emailcacuartasm@eafit.edu.cospa
dc.creator.emailjlaguilarc@eafit.edu.cospa
dc.date.accessioned2021-09-15T00:33:15Z
dc.date.available2021-09-15T00:33:15Z
dc.date.issued2021
dc.description.abstractThis article proposes a hybrid algorithm based on Reinforcement Learning and on the inventory management methodology called DDMRP (Demand Driven Material Requirement Planning) to determine the optimal time to buy a certain product, and how much quantity should be requested. For this, the inventory management problem is formulated as a Markov Decision Process where the environment with which the system interacts is designed from the concepts raised in the DDMRP methodology, and through the Reinforcement Learning algorithm – specifically, Q-Learning. The optimal policy is determined for making decisions about when and how much to buy. To determine the optimal policy, three approaches are proposed for the reward function: the first one is based on inventory levels; the second is an optimization function based on the distance of the inventory to its optimal level, and the third is a shaping function based on levels and distances to the optimal inventory. The results show that the proposed algorithm has promising results in scenarios with different characteristics, performing adequately in difficult case studies with a diversity of situations such as scenarios with discontinuous or continuous demand, seasonal and non-seasonal behavior with high demand peaks, multiple lead times, among others.spa
dc.identifier.ddc658.78 C961
dc.identifier.urihttp://hdl.handle.net/10784/30241
dc.language.isospaspa
dc.publisherUniversidad EAFITspa
dc.publisher.departmentEscuela de Administraciónspa
dc.publisher.placeMedellínspa
dc.publisher.programMaestría en Ciencias de los Datos y Analíticaspa
dc.rights.accessrightsinfo:eu-repo/semantics/openAccessspa
dc.rights.localAcceso abiertospa
dc.subjectSistemas de gestión de inventariospa
dc.subjectDDMRPspa
dc.subject.keywordSmart inventoryspa
dc.subject.keywordInventory management systemspa
dc.subject.keywordReinforcement learningspa
dc.subject.keywordQ-Learningspa
dc.subject.lembALGORITMOSspa
dc.subject.lembALGORITMOS (COMPUTADORES)spa
dc.subject.lembINVENTARIOSspa
dc.titleHybrid algorithm based on Reinforcement Learning and DDMRP methodology for inventory managementspa
dc.typemasterThesiseng
dc.typeinfo:eu-repo/semantics/masterThesiseng
dc.type.hasVersionacceptedVersioneng
dc.type.localTesis de Maestríaspa
dc.type.spaArtículospa

Archivos

Bloque original
Mostrando 1 - 3 de 3
No hay miniatura disponible
Nombre:
carta-aprobacion-trabajo-grado-eafit f.pdf
Tamaño:
243.02 KB
Formato:
Adobe Portable Document Format
Descripción:
Carta de aprobación trabajo de grado
Descargar
No hay miniatura disponible
Nombre:
carta-derechos-autor-eafit f.pdf
Tamaño:
848.28 KB
Formato:
Adobe Portable Document Format
Descripción:
Formulario de autorización de publicación de obras
Descargar
No hay miniatura disponible
Nombre:
Tesis Carlos Andrés Cuartas Murillo.pdf
Tamaño:
2.31 MB
Formato:
Adobe Portable Document Format
Descripción:
Trabajo de grado
Descargar
Bloque de licencias
Mostrando 1 - 1 de 1
No hay miniatura disponible
Nombre:
license.txt
Tamaño:
2.5 KB
Formato:
Item-specific license agreed upon to submission
Descripción:
Descargar

Colecciones

Maestría en Ciencias de los Datos y Analítica (tesis)