Warship 2026: Scaling the Fleet - Delivering Added Mass with Affordable Minor Warships

Data-driven Remaining Useful Life (RUL) prediction for maritime machinery usually assumes its sensor suite is given; the designer of an affordable minor warship, committing to instrumentation years before the first failure, needs to know how few channels suffice. This study asks that question, channel by channel, at the Marine Engineering Laboratory (MEL), a lab-scale electric ship machinery plant comprising six coupled systems, using 68 run-to-failure records under three operational profiles that stand in for mission states. Sequential forward selection and backward elimination draw sensor sufficiency curves; permutation importance and SHAP values check whether channel rankings are identifiable; profile-disjoint folds measure robustness across mission states; and a holdout scored exactly once tests whether the curves reproduce. The holdout verifies a five-channel cooling set, 0.611 RMSE against a no-sensor baseline of 1.107; fuel and electrical each support a single channel; and every change of mission state costs prognostic accuracy.

Development in autonomous technologies has driven a decade of research into automation transparency, emphasising the communication of a system's actions, reasoning, and projected outcomes to its operator. Although much of this evidence comes from commercial maritime, aviation and automobile sectors, the underlying Human Factors challenges align closely with those of naval operations, where automation increasingly supports propulsion, platform management, and mission critical decisions. This paper reviews the experimental literature on automation transparency and distils ten design principles spanning information content, system integration, uncertainty communication, and evaluation of transparent systems. Three overarching insights emerge for naval human machine integration: (1) transparency is not a quantity to be maximised but a set of design decisions concerning what information is communicated, when it is communicated, and how it is prioritised; (2) the purpose of transparency is to calibrate trust rather than maximise it, including enabling operators to identify and reject incorrect automated recommendations; and (3) transparency must be sustained throughout the system lifecycle, encompassing design, training, operation, and post operational review. These findings provide an evidence based foundation for the design and evaluation of transparent automation in future naval systems.

The article is divided into several distinct dynamic interface study parts.  The objective of this analysis is to describe improved air and sea vehicle launch and recovery methodologies using top-level multi-disciplinary dynamic interface descriptions.  Dynamic Interface (DI) as an academic discipline is defined as the study of the combination of two or more free bodies in motion along with their associated systems. Prior to sea trials, simulation is normally conducted.  The simulation must simultaneously compute the encountered inertial loads on the air vehicle from the simulated ship motions.  These are coupled with direct wind force or disturbed airflow over the deck as air wake.  The key is to define constantly the ship’s attitude and movement as a function of the of the air vehicle virtual centre of gravity.  The same methodology may be applied to other motion sensitive systems like the launch and recovery of RHIBs, for example, and even piloted air vehicles.  The elegance of this methodology is in its simplicity and its capacity to incorporate other operational scenarios.  The test equipment at sea is analysed using a similar process as in the simulated cases.  All the motion parameters are measured.  Having defined the ship environment, including a short horizon forecasting motion tool, the focus is to improve on the forecasting advanced length giving the launch and recovery events more time to prepare.  On recovery, air vehicle stability at the instant of touch-down or whilst sitting on the deck, is tracked as a function of the encountered deck forces coupled with air wake.  To extend the deck behaviour forecasts, it is necessary measure the seaway at remote locations ahead of the ship.  The long horizon forecasting tool is called Quiescent Period Prediction or QPP.  In practice, the Radar kit maps the seaway which is computed from its emission backscatter.  The backscatter is reduced to wave lines composed of wave height, direction and speed of propagation.  The backscatter image along with the shadowed waves are reduced to a sea spectrum containing all the frequencies from the converted wave form.  Switching to the long horizon forecast of airflow, the Quiescent Period Prediction air wake portion operates by measurements made by a LiDAR system.  The system measures arriving wind systems at a remote location of the ship to identify the safest, quieter intervals of air parcels.  Whilst the air and sea dimensions are physically independent, quiescent water patches are often shadowed by a quieter parcel of air.  QPP LiDAR can monitor various airflow components over the deck, including chimney exhaust.  The exhaust plume can have a significant impact on the safe launch, recovery and securing of the air vehicle.  With the deck motions defined for the robot, many more opportunities to land are uncovered.

Navies around the world are moving towards multi-vehicle maritime autonomy with a few operators expected to supervise multiple assets. While technical capabilities can be advanced through thousands of simulated iterations, the human factor challenges of such supervision cannot be accelerated in the same way, as iterative experiments with human is essential for system evaluation and improvement. This paper reviews the human factors lessons accumulated over decades in the aviation and army domains to gain design insights for naval maritime autonomy. Through a semi-systematic review, situation awareness, cognitive workload and trust emerged as the most recurrent challenges. For each challenge, the paper introduces its underlying mechanisms, measurement approaches and emerging solutions, including explainable artificial intelligence, machine learning based workload estimation, cognitive agents and large language models. The review concludes that closing the remaining validation gap calls for close cooperation between navies and researchers.

The Royal Navy is moving rapidly towards a hybrid navy in which crewed ships, autonomous platforms, sensors and digital networks operate as a distributed but digitally connected force. This transition introduces challenges in terms of fleet mix design, platform and fleet level requirement setting and operational analysis. This paper presents a combined top-down and bottom-up approach to assessing hybrid navy configurations, using well-established principles to assess the survivability of capability.  

The existing bottom-up platform orientated capability of the Purple Fire vulnerability/lethality tool is developed to enable it to assess hybrid fleets using a top-down system-of-systems survivability assessment approach. The utility of this new analysis technique is demonstrated by linking it with the Synthetic Environment MACE for operational assessments, allowing a dynamic understanding of capability resilience in realistic operating scenarios.  

The need for this type of analysis in assessing fleet mixes and in adopting new systems efficiently is discussed alongside the presentation of a real-world example analysis.  

Accurate understanding of landing craft behaviour during beaching, offloading and retracting operations is essential for the safe and effective execution of amphibious missions. Although computational tools for predicting shallow water hydrodynamics and hull–seabed interaction continue to improve, real world data to validate these models which in turn guide operational decision making, remain limited. To address these gaps, an extensive scale model test campaign was conducted to analyse the performance of three generic landing craft designs across a range of representative littoral conditions. The model-testing results demonstrate clear differences in beaching dynamics across various landing craft types. Preliminary numerical simulations show good correlations with experimental data for seakeeping characteristics, while there remain noticeable differences in other areas, particularly ship motions and beaching dynamics. Future work will focus on refining the fidelity of the simulation tools and establishing their general predictive capabilities through bespoke full-scale experimental validation activities, establishing appropriate limiting criteria and developing tactical decision aids to support safe operating envelopes for amphibious operations.

Pressure to reduce warship complement is legitimate: fewer people can reduce accommodation, volume, power and 
through-life cost. Yet lean-crewing experience shows that satisfying a workload requirement does not necessarily 
establish that capability can be sustained over time. 

This paper proposes recovery margin: the reserve capacity within the sociotechnical system to absorb demand, recover 
and remain effective across repeated demands. Drawing on the logic of naval design margins, it treats human-system 
reserve as something that can be consumed, restored and protected over time. Recovery-margin assessment is proposed as 
a complement to existing task-based validation, making visible whether sufficient reserve remains as demands accumulate. 
Fatigue, work design and autonomy are considered as mechanisms that shape how that reserve is depleted, restored and 
distributed across the wider capability system. 

Affordable distributed lethality is a balancing act across the warship life cycle cost including shipyard infrastructure investment, a batch building contract model and modular mission systems that match changing threats to installed mission systems. Affordable distributed lethality lives at the centre of the Venn Diagram for Upkeep, via maintenance and repair, Update, via obsolescence management, and Upgrade, via constant mission system capability integration. Affordable distributed lethality facilitates scaling the fleet by delivering added mass into minor warships. 

This paper examines the Human Factors (HF) challenges associated with deploying midsized torpedo-style Autonomous Underwater Vehicles (AUVs) and their Launch and Recovery (LAR) systems within modular frameworks. Mission modularity, including the containerisation of capability, enables navies to adapt expensive platforms to evolving roles and emerging needs. It supports rapid deployment of systems, such as AUVs, either through purpose-built modular spaces on naval vessels or by utilising Vehicles of Opportunity (VOO). While modularity enhances flexibility and operational agility, it introduces complexities related to confined working environments, system integration, and maintaining safe and effective operations in variable conditions. The study highlights the importance of aligning Human Readiness Levels (HRL) with Technical Readiness Levels (TRL) to mitigate risks and improve effectiveness. Key HF considerations include designing for maintenance within containerised systems and supporting efficient operations on diverse platforms. Addressing these challenges is crucial ensuring sustained operational capability across the fleet. 

Modernising sustainment practices through digital solutions is critical to delivering affordable naval capability at scale in Australia. As Defence shifts toward continuous naval shipbuilding and sustainment, the performance of the complete lifecycle ecosystem becomes the key outcome. This will require sustaining and upgrading the in-service fleet while simultaneously recapitalising with new platforms. The needed result is a build and sustainment environment where workforce, shipyard capability, and configuration control are managed as an integrated system. 

The paper examines how shipbuilding-specific Product Lifecycle Management (PLM) can provide Defence, shipyards, design agents, sustainment partners, and suppliers with an authoritative digital thread that connects engineering, procurement, construction, and in-service support. Rather than attempting to force every participant onto one authoring platform, the proposed approach separates CAD authoring from lifecycle information governance. A shipbuilding-specific PLM layer becomes a single harbour for controlled product structure, configuration state, asset condition, source lineage, change history, and downstream release.