Show simple item record

dc.contributor.authorMashhadi, Farshad
dc.contributor.authorSalinas Monroy, Sergio A.
dc.date.accessioned2020-11-23T20:31:57Z
dc.date.available2020-11-23T20:31:57Z
dc.date.issued2020-10-13
dc.identifier.citationF. Mashhadi and S. A. Salinas Monroy, "Deep Learning for Optimal Resource Allocation in IoT-enabled Additive Manufacturing," 2020 IEEE 6th World Forum on Internet of Things (WF-IoT), New Orleans, LA, USA, 2020, pp. 1-6en_US
dc.identifier.isbn978-172815503-6
dc.identifier.urihttps://doi.org/10.1109/WF-IoT48130.2020.9221038
dc.identifier.urihttps://soar.wichita.edu/handle/10057/19649
dc.descriptionClick on the DOI link to access the article (may not be free).en_US
dc.description.abstractAdditive manufacturing is revolutionizing the way that we produce, deliver, and consume objects in many industries. Compared to traditional manufacturing, where large factories mass produce objects far away from consumers, additive manufacturing can build customized objects in many small factories closer to where they are needed. Building these simplified supply chains requires manufacturers to employ IoT technologies that can provide real-time monitoring of the 3D-printers, and automatically adjust their operation to meet a highly-dynamic object demand. To this end, researchers have proposed the Additive Manufacturing (AM) cloud that can automatically manage the manufacturing resources based on real-time data collected from IoT-enabled equipment. However, most existing works on the AM Cloud only focus on finding production decisions that minimize the operating costs of the AM Cloud, and disregard setting prices for their services, which limits the profits of the manufacturers. To bridge this gap, we design a deep-learning auction that maximizes the utility of the AM Cloud by assigning 3D-printers to the buyers who are willing to pay the highest prices for their objects, and by assigning production orders to the manufacturers that can build the objects at the lowest cost. The auction prevents buyers from unfairly affecting the results, and its assignments satisfy production capacity and raw material constraints. We conduct extensive simulations, and see that our proposed auction mechanism can both find production control decisions and improve the utility of the AM Cloud by up to 50% compared to existing approaches.en_US
dc.language.isoen_USen_US
dc.publisherIEEEen_US
dc.relation.ispartofseries2020 IEEE 6th World Forum on Internet of Things (WF-IoT);Category numberCFP2018V-ART:Code164055
dc.subject3D printersen_US
dc.subjectAdditivesen_US
dc.subjectBridgesen_US
dc.subjectCostsen_US
dc.subjectDeep learningen_US
dc.subjectEngineering educationen_US
dc.subjectInformation managementen_US
dc.subjectOperating costsen_US
dc.subjectPrinting pressesen_US
dc.subjectProduction controlen_US
dc.subjectSupply chainsen_US
dc.titleDeep Learning for Optimal Resource Allocation in IoT-enabled Additive Manufacturingen_US
dc.typeConference paperen_US
dc.rights.holder© 2020 IEEEen_US


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record