Page 21: of Maritime Reporter Magazine (November 2026)

AI IN MARITIME & OFFSHORE s of 2024, the maritime AI market stood puter vision or machinery control cannot rely on continuous at $4.13 billion, with a projected ? ve-year cloud connectivity, so the processing must happen on board.

CAGR of 23%. Market-size estimates should AI and digitalization should not be confused with automa- be treated cautiously, but the direction is tion in general. Maritime systems are usually a combination

A hard to miss – AI solutions have advanced of physics-based models, optimization, rules, and machine from trials into voyage planning, machinery monitoring, learning, not a black-box approach. And sensors, inspection, bridge decision support, port coordination, and digital and edge computing are as important as modeling. Lead- assurance. The main questions are where it is producing ing companies focus on high-value and repeated use cases, measurable operational value and where the evidence still hardware-software-AI integration, and actual customer data. lags the marketing. AI and digital tools also serve buyers differently. Table 1 maps key needs to the technologies most directly associat-

THE VESSEL HAS TO BE CONNECTED FIRST ed with them, distinguishing shipbuilders’ priorities from

AI does not create useful ship data – it consumes it. The those of ship operators.

connected vessel is thus a prerequisite for the AI-capable ves- sel. Propulsion load, fuel ? ow, vibration, temperature, pres- WHERE THE NUMBERS ARE ALREADY VISIBLE sure, weather, satellite, radar, and camera data are collected One of the clearest commercial applications is voyage through the Internet of Things (IoT) sensors and existing optimization because fuel consumption provides an imme- shipboard systems, aggregated and pre-processed onboard, diate KPI. In February 2026, NAPA Voyage Optimization and transmitted to shore via secure communications links. trials demonstrated fuel savings of up to 7.1% on both long

Data quality and context set the ceiling on model perfor- and shorter voyages. On the separate 34-day Lila Global mance. Sensor drift, missing values, inconsistent naming, voyage from Suez to China, NAPA reported $40,200 in and different operating pro? les can affect the lab-tested solu- voyage cost savings, 35 tons of fuel saved, 106.5 tons of tion reliability. Latency is another factor – any critical com- CO2 avoided, and almost one day less steaming time com-

Table 1. What shipowners and shipbuilders want from AI, digitalizaO on, and enabling tools

RoleNeed / use caseWhat they want from it Typical AI, digitalizaO on, and enabling technologies

Ship owner / OperaO onal & Lower fuel use, emissions and IoT sensors, edge compuO ng, rouO ng pla? orms, digital twins, opO mizaO on operator energy opO mizaO on voyage cost; improved safety soO ware

ComplianceConO nuous monitoring and reporO ng; IoT sensors, data pla? orms, emissions systems, regulatory databases, AI- reduced manual compliance burden assisted search

PredicO ve Detect faults earlier; improve availability, CondiO on sensors, O me-series data, predicO ve models, digital twins, maintenance and reduce unplanned downtime remote diagnosO cs

Lifecycle Maintain a traceable digital record Digital twins, Product Lifecycle Management (PLM) systems, Enterprise management of vessel/component changes Resource Planning (ERP) systems, asset- and document-management systems

Cybersecure Protect connected navigaO on, Security InformaO on and Event Management (SIEM) systems, intrusion-de- operaO ons machinery, and vessel data as tecO on systems, anomaly-detecO on systems, ? rewalls, network-segmenta- automaO on increases O on tools, idenO ty and access-management systems, encrypO on

Progressive Support decisions and reduce crew MulO modal sensors, sensor-fusion systems, computer-vision systems, autonomy burden through remote/semi- remote-control systems, autonomous-control systems, human-machine autonomous operaO ons interfaces (HMIs)

Shipbuilder / Connected design Reduce build cost, rework, energy Digital twins, Computer-Aided Design (CAD) systems, Computer-Aided shipyard & engineering use, and design risk Engineering (CAE) systems, Computer-Aided Manufacturing (CAM) systems,

ComputaO onal Fluid Dynamics (CFD) soO ware, PLM systems; virtual reality (VR), augmented reality (AR), and mixed reality (MR) systems

Digital capture & Connect design, as-built, 3D scanning, point-cloud-processing software, digital-imaging systems, lifecycle conO nuity commissioning, and handover data PLM, ERP

ProducO on planning Reduce bo? lenecks, delays, material Scheduling soO ware, producO on-management systems, digital twins, ERP, & yard logisO cs movements, and idle O me computer-vision systems, tracking systems

Automated Automate hazardous/repeO O ve Industrial and collaboraO ve robots, roboO c welding and painO ng systems, producO on fabricaO on and material handling automated material-handling systems, machine-vision systems, 3D prinO ng

InspecO on, quality Detect defects, verify work, and Machine-vision systems, drones, three-dimensional (3D) scanning, roboO c & commissioning support commissioning inspecO on systems, non-destrucO ve tesO ng (NDT) equipment, defect- detecO on models, augmented reality (AR) and mixed reality (MR) systems

Source: author’s analysis www.marinelink.com 21

MR #11 (18-33).indd 21 MR #11 (18-33).indd 21 11/4/2026 12:33:56 PM11/4/2026 12:33:56 PM