In the news last week was the removal of a submarine CO whose surfaced boat struck a channel buoy and then run aground. While I don’t want to talk about this specific event, I want to ask what many who have served as an officer-of-the-deck wonder: are our ships hitting things (ships, buoys, seafloor, etc.) more or less frequently? What direction are we trending and why? With the wide array of sensors, computers, and operator aids available to OODs these days, are we any better at not hitting things? The answer may surprise you (or may not!).

I have been struggling to answer it, but how would you try to figure this out?

We could go back and search backissues of the NavyTimes for keywords such as “collision” or “grounding” and see how many such events have occurred per year. I don’t find this very satisfying though because it doesn’t account for the size of the Fleet or the robustness of its activities. For instance, maybe there were more collisions during the 80’s, but our Fleet was much larger then. Having more ships probably leads to a greater number of overall collisions. Maybe there were less collisions during the early 90’s, but maybe our ships were out to sea less. Less time at sea gives a ship less opportunity to hit something.

I think the metric I would most want to get my hands on would be “number of collisions per day at sea.” Take a given year, count all the collisions and groundings, and then divided by the sum of the days at sea of all our ships. What do you think we would see over the past 30 years?

Unfortunately, I haven’t gotten my hands on this data, but I have found 3 things which are of interest :

1) Loss rates in the commercial maritime community continue to fallIHS Fairplay’s World Casualty Statistics showed that from 1997-2011 vessel loss rates have gone down. This doesn’t really say they’re hitting things less frequently–it just means they’re being lost at sea less often. I think it’s as good a proxy as we can get to say the world is probably sailing the seas more safely [1].

Yearly Number of Vessel Losses from 1997-2011GrowthinWorldFleet














2) Loss rates have not always been falling. Lloyd’s List and the International Union of Marine Insurers have been tracking an improvement in the rate of accidents and casualties since the 1980s. Before then there was no real movement in these numbers.

3) Technology is probably of limited (or no?) impact. You might think that radar, GPS, and AIS have driven down accident rates. After all, if you know where your ship is and another vessel’s bearing, course, and speed, there’s no way the two of you should hit, right? Researcher Charles Perrow in his book Normal Accidents studied maritime casualties from post-WWII to the 1980s and found that having radar and other collision avoidance tools had no impact on the probability of collisions or groundings. For a time, having radar increased a ship’s likelihood of hitting something. Perrow surmised that crews were overconfident in their radar’s ability and were less likely to take strong risk mitigations during periods of reduced visbilities. He and others documented the birth of “radar-assisted collisions” where two ships, both of which had radar and weren’t on collision courses, made erratic course/speed changes at the last moment and hit each other. [2]

Perrow hypothesized that ship captains had tremendous pressure on them to meet schedules and would take extreme risks to stick to the plan. As a result, they were just using these tools to be more aggressive without changing their margin to hazard. I’m not convinced this is the full story since loss rates have come down; I doubt captains are under significantly less pressures today.

For the larger maritime community, why have loss rates fallen? Is it better use of technology? Better governance structures? Better trained mariners? Do you think the US Navy has followed these trends?

If anyone has any data or thoughts on these questions, I’d love to hear it.

[1] Charts from “15 Years of Shipping Accidents: A Review for the WWF,” Nickie Butt, David Johnson, Katie Pike, Nicola Pryce-Roberts, Natalie Vigar, Southampton Solent University

[2] Perrow, Charles. 1999. Normal Accidents: Living with High Risk Technologies (updated). Princeton University Press.

Posted by Jeffrey Withington in Uncategorized

You can leave a response, or trackback from your own site.

  • J. Perlmutter

    If you were to search the databases you’d need to use the cutesy navy term of art for colliding with stuff. The navy says allision and allided. They can’t bring themselves to confess even in an OPREP 3.

    • Jeff Withington

      Good point. Have to find that database though!

    • CAPT Mongo

      Pretty sure that terminology came from the Coast Guard, actually. “Allision” refers to running into a non mobile object, whereas “Collision” means two moving objects running into each other.

      I think it’s pretty silly too. Lawyers perhaps involved.

  • James B.

    So lousy ship-drivers continue to be lousy with higher-tech ships? Given what I’ve seen of car drivers, until the computer is actually replacing the driver, you can’t fix stupid.

    • Jeff Withington

      Exactly. Or at least stupid isn’t fixed by an easy tech add-on.

  • Michael Junge

    Google this: Electronic Chart Display and Information System-Navy: analysis and recommendations. Schweighardt, Edward Joseph.

    I posted a link to it yesterday but the comment still hasn’t cleared moderation.

    • Jeff Withington

      Thanks Michael! Just saw this!