USS John S. McCain: Touchscreen Steering Controls and a Fatal Loss of Situational Awareness
The USS John S. McCain collision showed how complex control-transfer modes, touchscreen throttles, inadequate training, and confused steering ownership could become fatal in congested waters.
A routine attempt to split bridge duties changed who actually controlled the ship
Before dawn on August 21, 2017, USS John S. McCain was approaching the Singapore Strait in dense traffic. The commanding officer directed the bridge team to split helm and throttle responsibilities so a lee helmsman could assist the main helmsman. During the transfer, steering control was unintentionally shifted to the lee helm station, and the throttles were also unganged so port and starboard propulsion could be controlled separately. The NTSB reconstruction shows that the crew’s understanding of control ownership diverged from the Integrated Bridge and Navigation System’s actual state.[1] A control-transfer feature intended to add flexibility instead created a hidden mode transition during a high-workload navigation period.
Control ownership must be visible to everyone who depends on it
When authority can move among consoles, the interface should make the current controller unmistakable at every station and require an explicit handoff acknowledgment. Silent or unilateral transfer turns mode state into a memory test.
The helmsman reported a loss of steering when steering had actually moved elsewhere
After control shifted, the original helmsman could no longer command the rudder and reported a steering casualty. The ship began turning to port while watchstanders tried to understand what had happened. Naval Safety Command material later summarized the central confusion as uncertainty about ‘who had control of what.’[2] This is a classic mode-error pattern: the system is functioning, but it is in a mode different from the one the user believes. Mode errors are especially dangerous because troubleshooting a nonexistent equipment failure can delay the simple action needed to restore control.
Mode changes need explicit acknowledgment
A control transfer should announce the new owner, require acknowledgment where practical, and leave the previous station with unmistakable feedback that authority has moved. Hidden reassignment is especially dangerous when a crew interprets nonresponse as equipment failure instead of a mode change.
Unganged throttles made the attempted slowdown increase the turn
When the bridge ordered a reduction in speed, the lee helmsman pulled back the port throttle, but because the throttles had become unganged the starboard shaft remained at higher power. The thrust imbalance increased the ship’s turn to port. The control action was locally reasonable—slow the ship—but the hidden mode changed its physical effect.[2] Continuous controls need strong feedback about coupling state. A user should be able to feel or immediately see whether two propulsion channels are linked, and a command that creates large asymmetric thrust should be conspicuous before it is applied in restricted waters.
The same gesture produced a different result after a hidden mode change
Interfaces become hazardous when familiar actions silently change meaning. Mode-dependent controls should either preserve consistent semantics or make the mode difference impossible to miss through layout, color, tactile feedback, and confirmation.
Touchscreen throttles removed tactile information available on mechanical controls
The NTSB found that the bridge system used touchscreens for thrust commands and specifically compared them with mechanical throttles, which provide position, direction, force, and coupling information through touch as well as sight. The board concluded that the design of the touch-screen steering and thrust system increased the likelihood of the operator errors that led to the collision.[1] Touchscreens are not inherently unsafe, but they are weakest when operators must control continuous physical variables without looking down or when they need to verify state by feel. In those tasks, tactile redundancy can be a safety feature rather than nostalgia.
Training and procedures did not match the complexity of the installed control system
The Navy’s own collision report said the accident was avoidable and cited complacency, procedural noncompliance, and a substandard level of knowledge about the ship control console.[3] That finding prevents a simplistic conclusion that the touchscreen alone caused the crash. A complex interface can be safe if crews are deeply trained and procedures control mode changes; a strong crew can also be defeated by ambiguous design under stress. Here, both layers were weak. Systems engineering has to align interface complexity with realistic training time, watch turnover, fatigue, and the number of rare modes a crew is expected to remember.
Training burden is a design cost
If safe use requires memorizing obscure transfer logic that is rarely practiced, the architecture is spending human memory as a reliability resource. That cost should be measured and reduced just like hardware complexity.
The Navy treated the collision as part of a broader surface-force problem
The McCain collision followed other serious mishaps, including the USS Fitzgerald collision two months earlier. The Navy’s Comprehensive Review of Surface Force Incidents looked beyond the individual bridge team to operational tempo, certification, training, maintenance, and organizational oversight.[4] This broader scope matters because local interface errors can be symptoms of systemic conditions. Crews need time to train on upgraded equipment, standards need to reflect actual hardware, and leaders need reliable indicators that watch teams are proficient before assigning demanding operations.
The NTSB recommended design as well as procedural changes
The NTSB issued recommendations covering steering modes, bridge procedures, training, and incorporation of human-engineering principles when complex systems are modernized. Its public summary specifically called for design guidance based on ASTM F1166 and better procedures for transferring steering and thrust control.[5] The recommendations recognize that accident prevention cannot rely on telling operators to be more careful. If a recurring action is easy to misunderstand, redesign should reduce the opportunity for confusion while training and procedures provide additional layers.
The engineering lesson is to simplify control transfer before adding more automation
The McCain case shows the danger of flexible, software-defined control without equally strong mode visibility. A safer bridge system would make control ownership explicit, require positive transfer acknowledgment, prevent accidental uncoupling during critical maneuvers, preserve tactile feedback for continuous thrust where practical, and provide an emergency steering mode whose behavior is simpler than the normal mode. Modernization should be judged by whether watchstanders can correctly predict the ship’s response after every transfer—not by the number of functions a touchscreen can consolidate. In safety-critical control, flexibility is valuable only when humans can see, feel, and verify the configuration that flexibility creates.
Works Cited
- 01
- 02Naval Safety Command — Lessons learned from USS John S. McCain steering casualty navalsafetycommand.navy.mil
- 03
- 04
- 05
CodeHistory is a living archive. Citations document the evidence used for this edition; later evidence may refine the account.