Can unmanned ships really be safer than manned ones?

Unmanned vessels will lead to a drop in the number of collisions and groundings, but non-navigational related incidents, such as fires and structural failures are likely to be exacerbated, according to recent research.

This conclusion was made by three academics from the Gdynia Maritime University in Poland, the Aalto University in Finland, and the Finnish Geospatial Research Institute, as a result of research submitted to and published in the journal of Reliability Engineering and System Safety in September this year.

Relocating crew to offshore bases comes with risk. While it may reduce the direct injury to crew onboard and reduce the risk of human error, it also opens up vessels to an array of hazards that are yet to be identified.

A lack of scientific study on these safety issues only highlights that there is still so much uncertainty around this.

In the research carried out, 100 accidents involving 119 vessels were analysed, concentrated in areas of vessels where unmanned operations would be likely to take place in the future. In total, 63 people lost their lives and 28 were injured in these 100 incidents.

A ‘What-if’ framework augmented by the Human Factors Analysis and Classification System (see information box below) for Marine Accidents and a consequences check that asked two questions were carried out.

  1. If the ship were unmanned, how would that fact affect the likelihood of particular accident?
  2. If the accident occurred anyway, would its consequences be more or less serious if there were no crew on board?

Based on the analysis, the authors found that the likelihood for groundings and collisions to take place would be much less on an unmanned ship, whereas fires, explosions and flooding would more severe to the damage on the ship on an unmanned vessel.

In the case of grounding, the researchers note that 24 out of 32 cases were found to be a direct result of nautical offers being distracted or following improper passages plans.

How likely will different types of accidents occur on both manned and unmanned vessels?

The academics estimate that in the cases analysed, 16 out of 19 collisions could be avoided using unmanned ships. This is because a lot of the incidents were a result of the bridge team’s non-compliance with the COLREG Rule 5 – Look out. If unmanned ships were deployed, it would be even safer to install infra-red cameras and other supporting equipment to help anticipate potential collisions.

However, something that would need to be considered in the result of a collision of an unmanned ship, is the search and rescue operations using manned ships that would take place to reach survivors. How to ensure efficiency of global search and rescue operations alongside a transition period into unmanned ships must be addressed.

Helicopter rescue on a burning ship (left) and the sinking of the Costa Concordia in April 2012 (right).

In terms of fire, self-heating is a common cause but also crew were found to disregard fire safety precautions. In the 100 incidents analysed, seven cases of fire were a result of technical failures, which is an increased vulnerability on unmanned ships.

In terms of stability risks, the authors expect that unmanned ships would be more likely to suffer because there is no crew onboard to mitigate the effects of heavy storms and severe weather conditions. In heavy weather, autopilot is generally discouraged from being used, but in the case of an unmanned vessel, there would be no option to switch to crew control.

In general, flooding events are likely to have a greater impact on cargo on an unmanned ship as there would be no crew onboard to adjust pipe lines or other components that may lessen the severity of the flood. However, taking improper action due to stress in the emergency or lack of training can contribute to higher levels of damage with crew onboard.

The authors believe that autonomy can reduce the probability of an accident, particularly when it comes to navigation-related ones, providing a systematic approach and reliable safety system is applied. However, this research did not look into the role that crew has upon preventing accidents, only their contribution to them and to dealing with them once they have occurred. Some incidents can be significantly reduced in severity thanks to the actions of crew taken immediately following, but the stress the crew face in an emergency can also mean that they do not always take best actions to mitigate the impacts of a situation.

The more experienced crew members are, the more likely they are to react to an unusual or extreme situation in the best possible way. With unmanned and autonomous ships, onshore staff will have to make rapid decisions based on the data they have available, which may be incomplete and carry some risk as the complete picture will not be available.

This presents some further risk in the event of a situation. However, as the research shows, the likelihood of emergency cases such as collisions and groundings will reduce with unmanned vessels, leading to a lower chance of rapid emergency intervention by crew and reactive decision-making.



The Human Factors Analysis and Classification System (HFACS) identifies the human causes of an accident and provides a tool to assist in the investigation process and target training and prevention efforts. It was developed by Dr Scott Shappell and Dr Doug Wiegmann in response to a trend that showed some form of human error was a primary causal factor in 80% of all flight accidents in the Navy and Marine Corps.

Leave a comment here

Comments are closed.