Dec 11

Dec 11 Post 9y Collision at Sea-Conclusion Part2

Introduction

The Navy report on the collision (ref (a)) divided its findings into three categories: Training, Seamanship and Navigation, and Leadership and Culture. As I noted in my previous post (9x), most of the findings either lack supporting evidence (How would the presence of the Officer of the Deck (OOD) and Conning Officer at the Navigation Brief have changed anything?), are vague (“as would have been appropriate,” “[a] basic level of system knowledge,” “sufficient experienced officers”), or are contaminated by hindsight bias (“a failure in risk management,” “If JOHN S MCCAIN had …,” “If ALNIC had …”). The word appropriate was used four times in the Findings section of the Navy report. There is no reference to the IBNS technical documentation or existing system performance problems and casualty reports (ref (c)) in the Navy report. It is impossible to know why these facts were omitted. Without consulting the NTSB investigation report (ref (b)) and Admiralty court opinion, it is difficult to learn from the Navy report.

Accepting references (a), (b), and (c) as bounded descriptions of events and assignments of responsibility in accordance based on the distinct logics driving the actions of the agencies that produced them, I used them to understand what happened and look for what operators could learn from the tragic collision between the USS JOHN S MCCAIN and M/V ALNIC 17 August 2017.

References

(a) Chief of Naval Operations. (2017). Memorandum for distribution, Enclosure (2) report on the collision between USS JOHN S MCCAIN (DDG 56) and motor vessel Alnic MC, retrieved from https://www.doncio.navy.mil/FileHandler.ashx?id=12011.

(b) National Transportation Safety Board. (2019). Maritime accident report: Collision between US Navy destroyer John S McCain and tanker Alnic MC Singapore Strait, 5 miles northeast of Horsburgh Lighthouse august 21, 2017. NTSB/MAR-19/01 PB2019-100970, retrieved from https://www.ntsb.gov/investigations/accidentreports/reports/mar1901.pdf

(c) Energetic Tank, Inc v. Unknown Defendant, 18-cv-1359 (PAC) (RWL) (United States District Court, Southern District (S.D.) New York, 2022), retrieved from https://casetext.com/case/in-re-energetic-tank-inc-2

(d) Reason, J. (1997). Managing the risks of organizational accidents. Routledge.

Untangling Collision Causality

In an effort to unpack the findings in the Navy’s report, improve the potential for learning and use concrete language (not meaningless terms like “leadership and culture”) for improved understanding, I separated causality for the collision into four categories:

key issues,
proximate causes,
latent conditions, and
knowable problems.

Key Issues

The key issues involved in the JSM-ALNIC collision were (attributions of responsibility follow each in parentheses):

Planning to station the Sea and Anchor Detail after entering the Traffic Separation Scheme (TSS) (CO)1
Entering the TSS with the ship’s most inexperienced OOD and Conning Officer (CO, XO, Senior Watch Officer, the OOD’s Department Head)2
Planning to conduct the TSS transit in the dark (CO)
The unplanned change of steering system operating mode despite everything else the Bridge team was managing (CO)3
Not using available indications to verify conning orders (Bridge team)4
Watchstander fatigue (CO, Senior Watch Officer)5

Proximate Causes

The CO’s order to change the ship control configuration, splitting helm control between the Helmsman and a Lee Helmsman who didn’t expect to have that watch (he probably didn’t attend the Navigation brief either), was the trigger “for the situation in which they found themselves.”6 The proximate causes for the collision were:

Cause 1: the improper transfer of helm control by an unknown actor. The CO didn’t order for the transfer to be done badly, but that’s what happened. “From the time when the CO ordered the Helm and Lee Helm split, to moments just before the collision, four different Sailors were involved in manipulating the controls at the SCC [Ship Control Console].”7 “No bridge watchstander in any supervisory position ordered steering control shifted from the Helm to the Lee Helm station”8. Nuclear operators call this “lack of formality.” I discussed this at length in Post 9t: Collision at Sea-What to do? Pt 3 Department Heads
Cause 2: the flaming garbage barge that was the watchstander response to the perceived loss of steering casualty in the 3 minutes and 28 seconds they had to respond before the collision.9

Latent Conditions (What you don’t know can hurt you)

There were conditions at the time of the collision, unknown to the ship’s leaders (or not mentioned in the Navy or NTSB reports). In his model of accident causality, James Reason called these latent conditions (Reason, 1997). Latent conditions always play a role in systems accidents like collisions at sea, but they don’t cause the accident. We only learn about them when operators stumble into them. Sometimes, but not always, this occurs just before disaster. The following problems were unknown to or unrecognized by the ship’s leaders before the collision and thus impossible to manage:

Routinely operating the steering system in a mode that made it possible for any station to take control of steering unilaterally. This was a consequence of poor system reliability (CO, the responsible Department Head, and many others inside and outside the lifelines of the ship)10, 11
Poor understanding of steering system operating modes and indications (many personnel)12
Poor technical documentation provided to the crew of the JSM for the IBNS system. There was no procedure in the IBNS Engineering Operational Sequencing System (EOSS) for transferring steering control between stations and there were deficiencies in the IBNS technical manual (Naval Sea Systems Command, NAVSEA).13,14. There is no mention of the IBNS technical documentation in ref(a), but it is specifically called out in refs (b) and (c).

Knowable Problems

There were three problems associated with steering control on the JSM that one has to consider knowable by experienced Navy operators. I point them out here not as a way of averring what the crew “should have known” because the fact is they didn’t. They are important because they give readers an opportunity to think about threats to reliability in their own contexts.

One problem that any mariner would consider knowable was the lack of proficiency with steering casualty response. This is primarily a Department Head (lead) and Commanding Officer (backup) responsibility. Even without drills that require lots of coordination, planning, and CO permission, an Officer of the Deck can improve the readiness of individual watchstanders for a steering casualty through level of knowledge checks (asking “What would you do if?” and “Explain how this works.”). They can also do casualty response walkthroughs with the Bridge team when not much is going on (about 90% of the time). Those are backups, however, for formal training and regular drills. I would not expect superior performance during a loss of steering casualty from the ship’s most inexperienced OOD and her watch team. The fact that only 3 minutes and 28 seconds elapsed between the perceived loss of steering and the collision is an illustration that there is little margin for poor performance in restricted maneuvering situations.

The CO’s deposition for the Admiralty court liability case stated that he considered “navigating the initial entrance to the Singapore Strait … more precarious than transiting in the Strait afterwards”15. This leads to the second knowable problem: the low experience of the Bridge watch team. This was a problem for entering a precarious high-traffic area in the dark.

The third knowable problem was the lack of proficiency and formality in operating the IBNS. The Personal Qualification System books for OOD and Lee Helmsman did not require demonstrations of the ability to transfer steering system control between stations.16 The crew did not make this adjustment to the qualification cards after the installation of IBNS a year earlier.

Final Thoughts

The two biggest issues contributing to the collision were the “knowable” lack of proficiency for a steering casualty on a glitchy steering control system AND poor understanding, from the Commanding Officer down to enlisted operators, of operations on that same system that also had deficient technical documentation. Being able to respond effectively to a loss of steering isn’t an exceptional skill any more than combating flooding. It is what professional mariners DO. Recognizing and taking effective action for poor system documentation are much harder than running lots of loss of steering drills. Running more drills might have identified problems and created more motivation to improve the technical documentation, but we’ll never know.

I have written extensively in posts 9r-9w about what I think can be learned at four levels of the ship’s crew: any operator, junior officers, department heads, and Commanding Officers. The recommendations aren’t boxes with solid walls. Junior officers should do all the things any operator should do and can take the lead on anything I recommended for department heads.

There's no "fixing" a senior officer who doesn’t seriously consider risk management recommendations from subordinates. Thinking you are the best risk manager on the ship without backup works until it doesn’t (see USS GREENEVILLE collision with the Japanese fishing training vessel). If the senior man in the organization doesn’t seriously consider safety recommendations from his direct reports, it is career decision time for those direct reports. This is easy to assert, but the biases involved in ignoring the recommendation of subordinates are very hard to overcome. We all have them and we are all affected by them. Using the STICC process I described in posts on CO Decision 2 and Decision Zero isn’t a cure, but it could help.

The nature of a military chain of command and practices for assigning responsibility and exercising authority do not make it easy to raise questions about a superior officer’s decision-making. Even worse, if a superior officer feels like a subordinate is questioning their judgment, it is unlikely that a fruitful review of risk management will ensue. This is an ethical dilemma with very high stakes and could be a speed line for ending a subordinate’s career. This is why it is important to recognize ethical dilemmas and options for managing them before they arise, something seldom taught in formal military training.

In the end, it is up to the individual to decide how much career risk to take by challenging a superior officer. I faced this choice more than once in my career and usually chose to speak up when I felt the risk of failure was high. When I thought my only options were: speak up and possibly get fired, or be lucky and keep quiet, but get fired if I wasn’t lucky, I preferred speaking up. I considered the heat it always generated preferable to getting my picture on the cover of Navy Times or facing years of legal hold with criminal negligence charges pending.

In my next and final post (I promise!) in this series, I will review some aspects of the mindset I tried to adopt for analyzing the events, context, and decisions in the collision between USS JOHN S MCCAIN (DDG 56) and motor vessel Alnic MC.