[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-119717-en":3,"doc-seo-119717-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},119717,1099513958762,"Logic","https://ap-avatar.wpscdn.com/avatar/1000023916a998db790?x-image-process=image/resize,m_fixed,w_180,h_180&k=1784791008015729253",8,"Research & Report","SHIP-D - Ship Hull Dataset for Design Optimization Using Machine Learning","SHIP-D provides a large, publicly available dataset designed to accelerate data-driven ship hull design. It includes 30,000 ship hulls with parameterization, mesh, point-cloud, and image representations, together with 32 hydrodynamic drag measures across operating conditions. The dataset combines human and computational use, enabling computational workflows as well as interactive input. The work also demonstrates an end-to-end pipeline using twelve CAD-derived hulls, training a surrogate model to predict wave drag and applying it in a genetic algorithm study to cut total drag by 60% while preserving key hull geometry constraints.","SHIP-D: SHIP HULL DATASET FOR DESIGN OPTIMIZATION USING MACHINE  \nLEARNING  \nNoah J. Bagazinski 􀀃  \nDepartment of Mechanical Engineering Massachusetts Institute of Technology Cambridge, Massachusetts, 02139 Email: [noahbagz@mit.edu](noahbagz@mit.edu)  \narXiv :2305 .08279v2  \nsign could be identified and optimized. However, the lack of publicly available ship design datasets currently limits the potential for leveraging machine learning in generalized ship design. To address this gap, this paper presents a large dataset of 30,000 ship hulls, each with design and functional performance information, including parameterization, mesh, point-cloud, and image representations, as well as 32 hydrodynamic drag measures under different operating conditions. The dataset is structured to allow human input and is also designed for computational methods. Additionally, the paper introduces a set of 12 ship hulls from publicly available CAD repositories to showcase the pro posed parameterization's ability to accurately reconstruct existing hulls. A surrogate model was developed to predict the 32 wave drag coefficients, which was then implemented in a genetic algorithm case study to reduce the total drag of a hull by 60% while maintaining the shape of the hull's cross section and the length of the parallel midbody. Our work provides a comprehensive dataset and application examples for other researchers to use in advancing data-driven ship design.  \nINTRODUCTION  \nRecent advancements in machine learning for engineering design have shown the ability to create novel designs [1], and  \n􀀃 Address all correspondence to this author.  \nFaez Ahmed  \nDepartment of Mechanical Engineering  \nMassachusetts Institute of Technology  \nCambridge, Massachusetts, 02139  \nEmail: [faez@mit.edu](faez@mit.edu)  \nhigh-performing systems level designs with significantly reduced cycle time [2] . The design of ships can greatly benefit from these advancements in machine learning methods as they have years long design cycles and are produced as one-off designs or in small batches. A well-designed machine learning tool for ship design could learn design trade-offs for ships through the continual design of many different types of ships. This can streamline the ship design process, which currently requires large teams of naval architects to balance all the trade-offs in a single ship's design. The current lack of a publicly available dataset for designing ships impinges this possibility. In order to create a machine learning tool capable of generalized ship design, a dataset of ships is needed that represents the vast array of current existing ships. Literature review was unable to find engineering datasets for machine learning for ship design that encompassed the full spectrum of ship shapes needed to generalize the ship design process. The lack of available public datasets is likely due to prior computational limitations, which have now been solved with time as computation has become cheaper and faster.  \nThis paper presents groundwork for the creation of a dataset of diverse ship hulls to implement machine learning methods for ship hull design. Hull design was chosen as a starting point for the creation of a dataset as this is the traditional starting point in ship design [3] . The hull shape affects several key aspects of a ship's performance, including the buoyancy, upright stability, hydrodynamics, and general arrangements of the ship. In addition, the shape of the hull has a direct impact on over 70% of the cost of a ship [4] . A ship's hull has a significant impact on many aspects of an overall ship system, making it a great candidate to apply machine learning methods to its design to balance overall design trade-offs with a data driven approach.  \nThe following sections detail the literature review of previous work, the methodology for generating a dataset of ships,  \nmeasures of the dataset, optimization of a ship hull using a trained surrogate model from the dataset, and","cbCaijh2R27GtIiR","https://ap.wps.com/l/cbCaijh2R27GtIiR","pdf",1862176,1,15,"English","en",105,"# Introduction\n## Motivation and dataset gap\n## Why ship hulls\n# Dataset methodology and representations\n## Parameterization and geometric coverage\n## Generated meshes, images, and measures\n# Dataset contributions and case study\n## CAD hull reconstruction showcase\n## Surrogate-based optimization with wave drag prediction\n# Previous work\n## Engineering dataset design principles\n## Ship hull representation methods\n## Machine learning approaches for hydrodynamic resistance","[{\"question\":\"What does the SHIP-D dataset provide for machine learning ship design?\",\"answer\":\"SHIP-D provides 30,000 ship hulls with parameterization plus geometric representations (mesh, point-cloud, images) and 32 hydrodynamic drag measures under different operating conditions.\"},{\"question\":\"How is the dataset parameterization structured to support generalized hull design?\",\"answer\":\"The parameterization is designed to represent a broad spectrum of hull geometric features and to allow human and computer inputs to coexist within the same data framework.\"},{\"question\":\"How was machine learning used to optimize hull performance in the paper’s case study?\",\"answer\":\"A surrogate model was trained to predict wave drag coefficients and then used within a genetic algorithm to reduce total drag by 60% while maintaining constraints on the cross-section and parallel midbody length.\"}]","SHIP-D - Ship Hull Dataset for Design Optimization Using Machine Learning | PDF",1785725945,38,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"ship-d-ship-hull-dataset-for-design-optimization-using-machine-learning","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/ship-d-ship-hull-dataset-for-design-optimization-using-machine-learning/119717/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-03",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What does the SHIP-D dataset provide for machine learning ship design?","Question",{"text":75,"@type":76},"SHIP-D provides 30,000 ship hulls with parameterization plus geometric representations (mesh, point-cloud, images) and 32 hydrodynamic drag measures under different operating conditions.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How is the dataset parameterization structured to support generalized hull design?",{"text":80,"@type":76},"The parameterization is designed to represent a broad spectrum of hull geometric features and to allow human and computer inputs to coexist within the same data framework.",{"name":82,"@type":73,"acceptedAnswer":83},"How was machine learning used to optimize hull performance in the paper’s case study?",{"text":84,"@type":76},"A surrogate model was trained to predict wave drag coefficients and then used within a genetic algorithm to reduce total drag by 60% while maintaining constraints on the cross-section and parallel midbody length.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]