SCHIFFSTECHNIK | SHIP
SCHIFFSTECHNIK | SHIP TECHNOLOGYreal world and the virtual world are both acting concurrently duringthe evolution of the design, the construction, and the productionof the ship. Ideally, this world of possibilities is extended overthe entire lifecycle of the product.At the core of these processes, CAD tools have evolved, whichmay easily be linked to these new technologies.To remain competitive, it is mandatory for the shipbuilding industryto understand how these technologies can improve theprocesses and the workflow as well as the cooperation betweenstakeholders. There is no global solution for all industries. Digitalisationin the Maritime world is available, and it is matureenough to be implemented and become part of our strategy forthe future.Digital transformation requires suitable strategies to identifythe added value obtained in relation to the necessary investments.In the current uncertainties we have to face, a good strategyleading to good results is to start with small projects that havesufficient viability to be evaluated without large investments(Minimum Viable Products). It has been proven that this approachcan bring good results, even exceeding expectations,without compromising too much on risks or investment. The keyfor success is to clearly define objectives, as well as identifying thecorrect technologies to apply.Furthermore, using visual tools for the presentation of resultsfacilitates the work of engineers and decision makers. These toolsare complemented by dashboards that must be adapted and personaliseddepending on the users or the shipyard needs.Another important point: future smart ships and yards must beinterconnected. Otherwise, they can’t use their potential. Theconnection of smart devices within a smart ship or smart yardmust be controlled by humans. The information shared must bemanaged along the whole lifecycle of the ship and the yard (workshops,docks, berths...), starting from the beginning of the initialdesign.Digital TwinsListening to the presentations the importance of the Digital Twinconcept in modern design and production has evolved as one ofthe big topics in the engineering world, including the maritimeindustries. At the core of any Digital Twin will be a 3D geometricalrepresentation, which opens the door for numerous applicationsin simulating physical behaviour and Virtual Realityinteraction. Complex Ship new-building projects are expected tobe designed and built with a Digital Twin in the future.For ships without appropriate digital representation of the asbuiltstatus, 3D laser scanning has been found to be a very successfulway forward by Marius Blom of Blom Maritime from Lysakerin Norway. In his experience, »data at your fingertips« as isthe case with a good digital twin have saved up to 30 % on materialcost and up to 35 % on man-hours in retrofit projects. Initiallymotivated for retrofits, the geometrical digital twin created by advanced3D laser scanning has been found to be very versatile alsofor other applications, such as preventive maintenance or training.Integration of 3D scan point clouds in CAD software hasbeen streamlined and can now be seen best business practice alsoin the maritime world.An interesting example of digital twin was described by ThomasDeNucci of the U.S. Coast Guard.A study was triggered when the fishing vessel »Scandies Roise«capsized in extreme spray-icing conditions on 31 December 2019Thomas Hildebrand has received this year’s »Compit Award« for hisnumerous contributions and support to the conference for many yearswith the loss of five of seven fishermen onboard. Testimonyidentified current vessel regulations may not accurately accountfor vessel icing on crab pots – a porous surface of webbing andopen spaces where icing can accumulate in a non-uniform pattern.The lack of understanding of this situation may put marinersand the marine environment at undue risk. DeNucci describeda methodology for predicting ice accretion on vessels operatingin high latitudes.Machine LearningMachine Learning has many tasks in ship design. Anomalies canfor instance be identified at different moments of the lifecycle of theproject. Machine Learning also makes it possible to optimise the resourcesused, by ensuring the highest possible standardisation ofparts, standardising materials and manufacturing. This will ultimatelyresult in a reduction of design and construction cost.A good example of the capabilities of Machine Learning in designwas presented by Thomas Hildebrand of Numeca with thepaper »Accelerating Marine Propeller Development in Early DesignStages using Machine Learning«. A machine learning approachwas presented to accelerate the design of marine propellers.For open-water propeller characteristics, neural nets aretrained to predict both integral quantities of open-water performancecurves and local field values of velocities and pressureson two-dimensional planes for Wageningen B-Series propellers.The average difference between the predicted and CFD- obtainedvalues remained below 1.5% for integrated quantities (thrust,torque, efficiency).This demonstrates that the quasi- instantaneous response of atrained neural net may accelerate propeller design significantly.Thomas Hildebrand has received this year’s »Compit Award« forhis numerous contributions and support to the conference formany years.Openness and CollaborationThe topic of openness regarding data and the use of software hasrepeatedly come up in the discussions at Compit this year. BesidesRodrigo Perez Fernandez in his overview on Smart Tech-© PayerHANSA – International Maritime Journal 08 | 202261
SCHIFFSTECHNIK | SHIP TECHNOLOGYof inspection was declared equivalent to traditional class requirementsby the attending surveyor.A field trial on board a chemical tanker is described. At first thedrone was tested operated by wifi / bluetooth connection withouta tether. The mirror-like reflecting surfaces of stainless steel howeverposed problems with wireless transmission of the signalhere. Therefore a tethered solution was used. The drone also enabledvisual close-up inspection of the upmost parts of the compartment,a capability not available before. The tethered systemallowed for continuous inspection, saving time and eliminatingsignal loss during flight. An anti-collision system prevented contactwith the structure. Results of detecting and tracking cracks invideos are summarised.The official group photo of Compit 2022 in Pontignano Junenologies for Improving Ship Design Tools, also Enrique Gaspar ofNTNU Alesund closed his paper with »A Call for Openness andCollaboration in Ship Design«. He addressed his colleagues andstudents to consider implementing open and collaborativemethods in the everyday design tasks, both at academia as well asin industry. A GitHub page for a project – either public or private– is an experience highly recommended. As GitHub is used tomanage large software projects, it has functions like allocatingtasks, discussions on traceability in pair (or even better) thanmost of PDM/PLM solutions. Giving up proprietary data-files inexchange for a standard among all tools seems to be a feasiblepath.As a final call, it would be nice to see academia really incorporatingthe open science concept, where the data from publicfunded projects is not only found in hidden papers, but in a repository,able to be accessed, scrutinised and re-used. For softwaredevelopments, many models suggest that open standardsand interfaces are key to success in the market, and may be a lucrativeapproach.© PayerThe Virtual ClassroomTowards the end of the Conference, Tracy Plowman from DNVgave a lively lecture on her experience after more than two yearsdelivering maritime training digitally. She describes the disruptionfrom the COVID-19 pandemic as a problem and achance. Training had to be changed spontaneously, the key vehicleswere suddenly virtual classroom and e-learning. The problemdeveloped into an opportunity. There was the signal from theTop: Digital is here to stay.The emergency is now over, long term solutions have to befound. In-line with the DNV sustainability drive. Can we movefrom real to digital in training?What did we learn? Before the change, classroom materialswere just sufficient for the real classroom. Experience soonshowed, training hours per day had to be shortened. Teaching onlinerequires new skills and more preparation time. Also, postpandemicpeople have more experience with online training andare more demanding. Customers in certain remote locations willhowever request to keep the online learning option going. It is expected,training solutions will become more complex, with moreblended or hybrid options available.The Virtual Classroom is quick to develop. It has direct accessto experts and direct interaction between experts and students ispossible. Experience has shown that Live online training also hasits challenges: language and accent challenges, engineers mayhave charisma problems, people sometimes get Cyber Fatigue.Evaluations show, there may be a requirement for more livelinessin the course. Solution: a lot less text – more imagery. The finalresult: Online learning is not as interactive as classroom learning.There is less knowledge-sharing. Most pupils hope to be back toclassroom learning soon.Remote Inspections using DronesErik Stensrud of DNV and Kristian Klausen of ScoutDI reportedabout their development and experience with automatic inspectionof cargo and ballast tanks using drones already at Compit2019 and 2020. This year, two important forthcomings weredescribed: The drone pilot operated the drone from outside thetank, i.e. beyond his line of sight (BLOS). Additionally, this formRetrofit Installation – E3D, by Blom Maritime© Blom Maritime62 HANSA – International Maritime Journal 08 | 2022
