Collision between US Navy Destroyer Fitzgerald and Philippine-Flag Container Ship ACX Sagami Nada Bay off Izu Peninsula, Honshu Island, Japan July 17, 2017 Marine Accident Report P32020-101007 . . Na?onal -- Satew Board PBZOZO-101007 Notation 65315 Adopted August 3, 2020 Marine Accident Report Collision between US Navy Destroyer Fitzgerald and Philippine-Flag Container Ship ACX Sagami Nada Bay off Izu Peninsula, Honshu Island, Japan June 17, 2017 National M, Transportation Salem Board 490 L?Enfant Plaza SW Washington, DC 20594 National Transportation Safety Board. 2020. Collision between US Navy Destroyer Fitzgerald and Philippine-Flag Container Ship ACX agami Nada Bay oijzu Peninsula, Honshu Island, Japan, June 1 7, 201 7. Marine Accident Report Washington, DC. Abstract: This report discusses the June 17, 2017, collision between the US Navy Destroyer Fitzgerald and the container ship ACX off Honshu Island, Japan. The Fitzgerald was heading in a southerly direction, bound for the Philippines, and crossing the track of the ACX east-northeast-bound for Tokyo Bay. Seconds before the collision, the watch of?cers on both vessels ordered course and speed changes, but the vessels ultimately collided. As a result of the accident, seven Fitzgerald sailors died, three were injured, and the destroyer sustained more than $300 million in damage. No one was injured on the ACX but the vessel sustained damage to its bow. No pollution was reported. This report identi?es the following safety issues: The Fitzgerald crew?s fatigue, the practice of US naval vessels not to broadcast automatic identi?cation system (AIS) information, failure of both vessels to follow required actions in accordance with the International Regulations for the Prevention of Collisions at Sea, the Fitzgerald commanding of?cer?s failure to adequately assess the hazard presented by the vessel?s intended transit, and insuf?cient oversight by the US Navy. As a result of this investigation, the National Transportation Safety Board makes new safety recommendations to the US Navy and Sea Quest Ship Management, Inc. The National Transportation Safety Board (N TSB) is an independent federal agency dedicated to promoting aviation, railroad, highway, marine, and pipeline safety. Established in 1967, the agency is mandated by Congress through the Independent Safety Board Act of 1974, to investigate transportation accidents, determine the probable causes of the accidents, issue safety recommendations, study transportation safety issues, and evaluate the safety effectiveness of government agencies involved in transportation. The NTSB makes public its actions and decisions through accident reports, safety studies, special investigation reports, safety recommendations, and statistical reviews. The NTSB does not assign fault or blame for an accident or incident; rather, as speci?ed by NTSB regulation, ?accident/incident investigations are fact??nding proceedings with no formal issues and no adverse parties and are not conducted for the purpose of determining the rights or liabilities of any person? (Title 49 Code of Federal Regulations section 831.4). Assignment of fault or legal liability is not relevant to the statutory mission to improve transportation safety by investigating accidents and incidents and issuing safety recommendations. In addition, statutory language prohibits the admission into evidence or use of any part of an NTSB report related to an accident in a civil action for damages resulting ?om a matter mentioned in the report (Title 49 United States Code section 1154(b)). For more detailed background information on this report, visit the NTSB investigations website and search for NTSB accident ID DCA17PM018. Recent publications are available in their entirety on the NTSB website. Other information about available publications also may be obtained from the website or by contacting? National Transportation Safety Board Records Management Division, CID-40 490 L?Enfant Plaza, SW Washington, DC 20594 (800) 877-6799 or (202) 314-6551 Copies of NTSB publications may be downloaded at no cost from the National Technical Information Service, at the National Technical Reports Library search page, using product number For additional assistance, contact? National Technical Information Service 5301 Shawnee Rd. Alexandria, VA 22312 (800) 553-6847 or (703) 605-6000 NTIS website NTSB Marine Accident Report Conte Contents i Figures Acronyms and Abbreviations iv Executive Summary vi 1 Factual Information 1 1.1 Background and Precipitating Events 1 1.2 The Accident 7 1.3 Navigation Equipment 12 1.3.1 Fitzgerald 12 1.3.2 ACX 13 1.4 Organizational Guidelines 13 1.4.1 Fitzgerald 13 1.4.2 ACX 14 1.5 Personnel 15 1.5.1 Fitzgerald 15 1.5.2 ACX 15 1.6 Manning 16 1.6.1 Fitzgerald 16 1.6.2 ACX 18 1.7 Work/Rest History 18 1.7.1 Fitzgerald 18 1.7.2 ACX 19 1.8 Waterway Information 19 1.9 Environmental Information 20 2 Analysis 21 2.1 Exclusions 21 2.2 Actions to Recognize and Avoid the Collision 21 2.2.1 Fitzgerald 21 2.2.2 ACX 25 2.3 Operating Procedures 28 2.3.1 Fatigue 28 2.3.2 Navy Oversight 30 2.3.3 Navy Actions since the Accident 31 3 Conclusions 32 3.1 Findings 32 3.2 Probable Cause 32 4 Recommendations 34 4.1 New Recommendations 34 NTSB Marine Accident Report Appendix A Investigation 35 Appendix Vessel Information 36 Appendix Pertinent Documents Reviewed by Investigators but Protected by the Navy ..37 References 38 ii NTSB Marine Accident Report I res Figure 1. US Navy Destroyer Fitzgerald. 1 Figure 2. Chart of the accident area off the south coast of Honshu Island, Japan 2 Figure 3. Key members of the Fitzgerald bridge watch team leading up to the accident. 3 Figure 4. Key members of the Fitzgerald CIC watch team leading up to the accident. 5 Figure 5. Container ship ACX 6 Figure 6. A comparison of radar screens from different vessels before the collision 9 Figure 7. NTSB reconstruction of the vessels? paths 10 Figure 8. Post collision damage to the Fitzgerald?s starboard side 11 Figure 9. Postaccident damage to the ACX bow. 12 Figure 10. NTSB path reconstruction for Fitzgerald and ACX 24 NTSB Marine Accident Report Acronyms and Abbreviations AB AIS ARPA BMOW CIC CO COLREGS DESRON ECDIS GPS hp IMO JOOD kW MoBoard NOAA NTSB OOD PQS SMS SOLAS SORM STCW Code VDR able-bodied seaman automatic identi?cation system automatic radar plotting aid boatswain mate of the watch combat information center commanding of?cer collision regulations (Convention on the International Regulations for Preventing Collisions at Sea) Destroyer Squadron (Navy unit) electronic chart display and information system global maritime distress and safety system global positioning system horsepower International Maritime Organization junior of?cer of the deck kilowatt maneuvering board National Oceanic and Atmospheric Administration National Transportation Safety Board of?cer of the deck personnel quali?cation standards (Navy) safety management system International Convention for the Safety of Life at Sea Standard Organization Regulations of the Navy Seafarers? Training, Certi?cation and Watchkeeping Code voyage data recorder iv NTSB Marine Accident Report VMS voyage management system VHF very high frequency X0 executive of?cer NTSB Marine Accident Report Executive Summary Accident Summary About 0130 (local time) on June 17, 2017, the US Navy Destroyer Fitzgerald with 315 persons on board was southbound at a speed of about 22.1 knots in the bay of Sagami Nada off Japan?s Honshu Island after departing the US Navy Base at Yokosuka, Japan, bound for the Philippines. The Philippine-?ag container ship ACX operated by Sea Quest Ship Management, Inc., with 20 crewmembers on board was east-northeast-bound at a speed of about 18.5 knots, headed to Tokyo, Japan, ?om Nagoya, Japan. As the distance between the two ships continuously decreased, neither vessel radioed the other. Seconds before the collision, the watch of?cers attempted to maneuver the vessels to avoid impact, but the actions were too late, and the ships collided. Seven Fitzgerald crewmembers died in the accident, and three crewmembers suffered serious injuries. The destroyer sustained extensive damage to its forward starboard side. The ACX sustained damage to its bow; no injuries were reported. Investigation The National Transportation Safety Board (N TSB) was the lead federal agency in this accident investigation and delegated its authority to the US Coast Guard to gather documents and perform interviews on behalf of the NTSB. The NTSB developed the analysis and probable cause based on the evidence gathered by the Coast Guard and additional documentation provided by the Navy. Probable Cause The National Transportation Safety Board determines that the probable cause of the collision between US Navy Destroyer Fitzgerald and container ship ACX was the Fitzgerald?s bridge team?s failure to take early and substantial action to avoid collision as the give-way vessel in a crossing situation. Contributing was ineffective communication and cooperation among the Fitzgerald crew on the bridge and in the combat information center (CIC), and the Fitzgerald commanding of?cer?s (CO) insuf?cient planning for the hazards of the vessel?s intended transit. Also contributing was the Navy?s ineffective oversight of the Fitzgerald in the areas of operations scheduling, crew training, and fatigue mitigation. Also contributing to the accident was the ACX watch of?cer?s lack of early detection of the Navy vessel and insuf?cient actions to avoid collision once in doubt as to the destroyer?s intentions. Safety Issues Safety issues identi?ed in this accident include the following: Fitzgerald crew?s insufficient training, The surface warfare supervisor missed several critical targets, and the officer of the deck (00D), who was in charge of the bridge personnel, made some poor navigational decisions and did not request support from the CIC. Further, the tactical action of?cer (in charge of the CIC) did not ensure that her personnel supported the bridge team. NTSB Marine Accident Report Fitzgerald crew?s fatigue. The ship scheduled multiple events on the day before the accident that required the participation of much of the crew, including key on the accident watch. All of these had little or no sleep before heading to watch. Further, the accident occurred just prior to a time period considered to be a circadian low (roughly 0200?0600), when the body is normally more fatigued and prone to diminished alertness and degraded performance. Practice of US naval vessels not to broadcast automatic identification system (AIS) signals. The Fitzgerald was equipped with AIS, a maritime navigation safety communications system that automatically transmits vessel information to other vessels, allowing early detection of a target. On the day of the accident, the Fitzgerald was not transmitting its data, although it was receiving information about other vessels in the area. To track the destroyer?s position electronically, other vessels had to rely on visual means or radar. The destroyer was built by design to present a smaller target on radar displays than other (non-military) vessels of similar size. The destroyer?s radar signature appeared signi?cantly smaller than that of a comparable merchant vessel of the same size on the radar on the ACX the container ship with which it later collided. Failure of both vessels to follow required actions in accordance with the Convention on the International Regglations for Preventing Collisions at Sea (COLREGS). As the Fitzgerald proceeded in a southbound direction on the evening of the accident, the COD picked up three contacts?the ACX the Maersk Evora, and the Wan Hai 266?off the destroyer?s starboard bow at approximately 12 miles. The three vessels were on parallel paths heading east-northeasterly. Per the COLREGS, when two vessels are crossing, the vessel that has the other on its starboard side shall keep out of the way of the other vessel. The Fitzgerald had all three vessels, the ACX the Maersk Evoraits starboard side, and therefore was the give-way vessel to all three ships. Instead, the bridge team on the Fitzgerald continued ahead, ultimately crossing ahead of the Wan Hai 266 and placing the destroyer in the path of the ACX which resulted in the collision. As a potential collision situation was developing, the second officer on board the ACX the stand-on vessel, did not take suf?cient action when it became apparent that the give-way vessel was not taking enough action to avoid collision. Fitzgerald commanding officer not augmenting bridge personnel w_ith a more experienced officer while the vessel was crossing busy coastal traffic routes. The Fitzgerald OOD was supported by 6 bridge and 20 CIC personnel, and was manned in accordance with the US Navy?s at-sea policy. Fitzgerald commandi_nq officer not ad?uatelv assessing the hazard presented by the vessel's intended transit. The vessel?s path leading up to the accident crossed major shipping routes off the coast of Japan; however, the risk of transiting through areas known for heavy traf?c was not addressed. Insufficient oversight and directive bv the US Navy. The Fitzgerald?s schedule leaving port to comply with certi?cation requirements and to return to its deployment schedule provided little rest for the crew on the day before the accident. The Navy had no fatigue mitigation program or standards for ensuring shipboard crews had adequate rest. The Navy was required to assess and certify that the operating procedures and vii NTSB Marine Accident Report quali?cation system aboard the Fitzgerald were effective, but the crew?s navigation decisions on the night of the accident indicate the Navy?s assessment and certi?cation process requires review. Findings Weather and the steering and propulsion systems on board the Fitzgerald and the ACX were not factors in the accident. The Fitzgerald bridge team, on the give-way vessel, did not take early and substantial action to avoid collision with the stand-on ACX The communication and cooperation among the Fitzgerald crew on the bridge and in the combat information center were ineffective leading up to the accident. The Fitzgerald?s unexplained small course change to starboard minutes before the collision put the vessel on a collision course with the ACX The Fitzgerald combat information center personnel did not effectively support the bridge team in tracking nearby surface targets. The Fitzgerald provided a detectable radar signature, and the second of?cer on board the ACX should have acquired the destroyer on the automatic radar plotting aid to determine risk of collision. When the second of?cer on the ACX did not receive a response to his signaling attempts and saw no noticeable change in approach from the destroyer, he did not take suf?cient action to avoid the collision. The absence of an automatic identi?cation system signature broadcast from the Fitzgerald likely contributed to the lack of its early detection by the ACX bridge team. The Fitzgerald commanding of?cer did not adequately assess the hazard presented by the vessel?s intended transit. which crossed busy coastal traf?c routes. and should have assigned a more experienced of?cer to augment the bridge team. Several Fitzgerald bridge and combat information center were likely acutely fatigued at the time of the accident. which impacted their situation awareness and ability to identify and respond to the approaching ACX The Navy failed to provide effective oversight of the Fitzgerald in the areas of operations scheduling, crew training, and fatigue mitigation. Recommendations New Recommendations As a result of its investigation, the National Transportation Safety Board makes the following new safety recommendations: To the US Navy NTSB Marine Accident Report Review and revise ?eetwide training and quali?cation requirements for of?cers of the deck related to the collision regulations. (M-20-10) Review and revise bridge resource management training in vour approved course curriculum to promote a cohesive team environment and improve communication within and between bridge and combat information center teams. (M-20-11) Instruct vour vessels to broadcast automatic identi?cation system information while operating in the vicinity of commercial vessel traf?c at all times unless such broadcast compromises tactical operations or strategic interest. (M-20-12) To Sea Quest Ship Management, Inc. Provide additional training for vour navigation of?cers on collision avoidance regulations, radar, and automatic radar plotting aid. (M-20-13) NTSB Marine Accident Report 1 Factual Information 1.1 Background and Precipitating Events At 1130 (local time) on the morning of June 16, 2017, the day before the accident, the US Navy Destroyer Fitzgerald departed the port of Yokosuka on Honshu Island, Japan, about 50 nautical miles northeast of the collision site, to conduct scheduled operations in the bay of Sagami Wan.1 The Fitzgerald ?rst went to an anchorage for about 3 hours, where it received ammunition. The vessel then proceeded to a location in Sagami Wan off the southwest coast of the Miura Peninsula where, between 1800 and 211 1, the crew conducted helicopter ?ight operations and then, between 2133 and 2303, small boat operations.2 Figure 1. US Navy Destroyer Fitzgerald. (Source: US Navy) After completing these operations, the Fitzgerald was scheduled to conduct engineering drills June 17?18 while en route to operations in the South China Sea, and then continue to Subic Bay, Philippines. 1 All miles in this report are nautical miles (1.15 statute miles) unless otherwise noted. Weather Visibility data may be reported in statute miles. Supporting documentation for information referenced in this report can be found in the public docket for this accident, which can be accessed from the National Transportation Safety Board? 5 (N TSB) Accident Dockets web page by searching DCA17PM018. NTSB Marine Accident Report 25 nm vuhmuka T?kr? - 1.3.- MI I'i?DI'lEi'llJ Island Saga mi Peninsuia Wan' izu a: . Peninsula Russia 3,me Japan Nagoya. Till] nm 3393"? Paci?c Due-a Nada Figure 2. Chart of the accident area off the south coast of Honshu Island, Japan. The accident site is marked with a red triangle. (For more detail, see section 1.8 Watenivay for chart: National Geospatial Intelligence Agency, with annotations by NTSB. Background source for inset map: Google Earth.) On June 16 at 2200, a watch turnover took place on board the Fitzgerald navigation bridge. It was preceded by a watch exchange between the offgoing (who had stood the 1700? 2200 navigation watch) and the oncoming 2200?0200 personnel. The new bridge complement included six persons: Officer of the deck (00D). Directly responsible for the safe navigation and general operation of the ship and in charge of the bridge personnel during her shift. She was commissioned in the Navy in August 2014, had served on board the Fitzgerald since May 2016, and had previously stood three or four bridge watches in this waterway. Junior officer of the deck (J 00D). Principal assistant to the COD. Her duties included recommending traf?c management based on information from radar (such as data by automatic radar plotting aid, or ARPA) and from manual plotting. She was commissioned in the Navy in June 2012, had served on board the Fitzgerald since September 2016, and had been in this waterway during that time, but had not previously stood a navigational watch. Conning officer. Assigned to conduct the movement of the vessel, as directed by the GOD, relaying steering and propeller thrust orders from the COD to the helmsman. The position was an entry-level watch position for new of?cers or for those who were new to the bridge, generally requiring no prior bridge quali?cations. He was NTSB Marine Accident Report commissioned in the Navy in January 2017, and the Fitzgerald was his ?rst ship. The accident watch was his ?rst bridge watch on this waterway. 0 Quartermaster of the watch. Assigned to plot the Fitzgerald?s movement and make recommendations to the COD. He joined the Navy in 2002 and the Fitzgerald in September 2016, having previously served on Norfolk, Virginia-based ships in the Mediterranean Sea and the Arabian Gulf. 0 Helmsman. Assigned to carry out steering and propeller-thrust orders from the conning of?cer and the COD. She joined the Navy in 2015 and the Fitzgerald in April 2017. Boatswain mate of the watch (BMOW). The BMOW was the enlisted person who, in addition to other duties, managed the enlisted personnel on the bridge watch including the helmsman and lee helmsman. He joined the Navy in July 2006 and the Fitzgerald in May 2015. The following graphic illustrates the key personnel and their hierarchy in the Fitzgerald bridge watch leading up to the accident: Officer of the deck Quartermaster Junior officer of of the watch the deck Conning officer Boatswain mate of the watch Helmsman Figure 3. Key members of the Fitzgerald bridge watch team leading up to the accident. The only reported equipment malfunction in the bridge or CIC was an inoperative radar repeater (the SPA-25) in the CIC.3 It was shut down completely. 3 A CIC is an internal room on board a military vessel that serves as a tactical center and provides processed information for command and control of the nearby area of operations. The CIC focuses on potential security threats and tracks air and surface contacts. CIC personnel coordinate with and support the navigation bridge crew. 3 NTSB Marine Accident Report The itzgerald?s ?rst- and second-in-command?the commanding of?cer (CO) and the executive of?cer (XO)?had been on the navigation bridge throughout the evening of June 16.4 About 2300, as the Fitzgerald small boat operations ended and the destroyer departed the Sagami Wan training area to begin the voyage south, the CO exited the bridge for the night. Before leaving, he issued the night orders for the transit and advised the GOD of the change. He doubled the standard allowable deviation from the predetermined trackline before the DOD was required to notify him. The CO told investigators he made this change to allow the GOD more leeway in contact avoidance so that he could ?get a little more sleep before the next day.?5 The XO left the bridge about 2330, approximately half an hour after the CO. Like the navigation bridge, the destroyer?s CIC also had a watch change for the 2200?0200 shift. The oncoming CIC complement included six persons: 0 Tactical action officer. Assigned to oversee the defensive and offensive posture of the vessel and to ensure CIC personnel supported the bridge watch. She joined the Navy in December 2008 and, after serving several seagoing and shoreside assignments, joined the Fitzgerald in January 2017, qualifying as tactical action of?cer the following March. She had stood about 10 bridge or CIC watches in the accident waters. 0 Watch officer. Assigned to supervise the CIC and its operations, ensure all contacts were detected and reported, and ensure all evaluated information was disseminated to pertinent entities, such as the bridge. The watch of?cer was commissioned in the Navy in 2015 and joined the Fitzgerald in December that year. He had made about ?ve transits in the accident waterway. 0 Watch su perviso r. Assigned to report directly to the CIC watch of?cer and support both him and the tactical action of?cer. The watch supervisor was also the primary person to plot via maneuvering board (MoBoard) and report the results to the bridge.6 The watch supervisor joined the Navy in 2006 and the Fitzgerald in November 2016, qualifying as CIC watch supervisor in March 2017. 0 Surface warfare coordinator. Assigned to develop the surface plot after obtaining target information from both the surface warfare supervisor and the optical sight system operator (see below) and evaluate surface contacts contained on the surface plot. Responsible for apprising the bridge of surface radar contacts that might impact the destroyer?s intended movement, and for populating the surface plot with other 4 See Section 1.5.1 of this report for information on the CO and experience. 5 The NTSB delegated its authority to the US Coast Guard to gather documents and perform interviews on behalf of the NTSB in this accident investigation. 6 Manual plotting is done by placing a mark on a sheet of paper called a maneuvering board (or ?MoBoard? for short), corresponding to the radar target?s true bearing and range from one?s vessel and then repeating this process at a set time interval (usually 3 minutes.) A line connecting the marks and extending in the direction in which the target is moving provides a vector whose closest point to one?s vessel is the closest point of approach and the location from one?s own vessel where the closest proximity will occur. After vectorially adding one?s vessel?s course and speed to the target ship?s vector, a third vector is obtained by connecting the dots to form a triangle. The length and direction of this third vector gives the target vessel?s course and speed. 4 NTSB Marine Accident Report ship information obtained via automatic identi?cation system The surface warfare coordinator joined the Navy in 1994 and the Fitzgerald in December 2015. He said that he had made ?a couple? of previous trips in the accident waterway. Surface warfare supervisor. Assigned to identify surface contacts by radar and populate the surface plot or report the surface contact to the surface warfare coordinator. The surface warfare supervisor joined the Navy in 2014 and the Fitzgerald in May 2016, and quali?ed for his current role in March 2017. He had not previously transited in the accident waterway during a CIC watch. Optical sight system operator. Assigned to visually locate surface contacts via a night-Vision-capable optical system. He joined the Navy in May 2015; the Fitzgerald was his ?rst ship. Although he stated that he had stood many CIC watches as optical sight system operator, investigators could not determine if any of those watches were in the accident waterway. The following graphic illustrates the key personnel and their hierarchy in the Fitzgerald CIC watch leading up to the accident: Watch officer Tactical action officer Surface warfare coordinator Optical sight system operator Surface warfare supervisor Watch supervisor Figure 4. Key members of the Fitzgerald CIC watch team leading up to the accident. 7 AIS is a maritime navigation safety communications system. At 2- to 12-second intervals on a moving vessel, the AIS automatically transmits vessel information, including the vessel?s name, type, position, course, speed, navigational status, and other safety-related information, to appropriately equipped shore stations, other vessels, and aircraft. The rate at which the AIS information is updated depends on vessel speed and whether the vessel is changing course. AIS also automatically receives information from similarly equipped vessels. Coast Guard regulations require AIS in waterways governed by vessel traf?c control (Title 33 CFR Part 164). 5 NTSB Marine Accident Report As the Fitzgerald proceeded southbound, it approached an area of the waterway where the destroyer would be crossing the path of vessels transiting the coast of Japan and heading toward and coming from Tokyo Bay. Per the standing orders, the GOD was instructed to call the CO if other vessels had a closest point of approach (CPA) of less than 3 miles. The report was to be made at 5 miles or 20 minutes prior to the calculated CPA, whichever occurred ?rst. According to the Fitzgerald?s voyage management system (VMS), between 2340 and 0025, the destroyer overtook contacts both to port and starboard.8 At 0033 on June 17, VMS showed ?ve vessels heading in a general northeasterly direction, crossing from starboard to port relative to the Fitzgerald, and two vessels heading in the same general direction as the Fitzgerald (on the starboard bow). The 00D called the CO to discuss these contacts. To negotiate around these vessels, the Fitzgerald deviated from the planned trackline by as much as 3,800 yards (1.8 nautical miles). The destroyer, at an overall speed of about 19?20 knots, proceeded to transit astem of the ?ve northeast-bound vessels, leaving them on its port side, with the closest separation at just under 1 mile. A?er 0033, although subsequent vessels had CPAs of less than 3 miles from the Fitzgerald, the COD placed no further calls to the CO. One of the many vessels transiting in the same area was the Philippine-?ag container ship ACX It had departed the port of Nagoya, Japan, about 70 miles west-northwest of the accident site, at 1648 the previous a?emoon (June 16) after discharging and loading containers. The ACX with about 15,330 metric tons of cargo on board, was scheduled to arrive at the Tokyo Bay pilot station at 0500 on the morning of June 17. It was transiting in an east-northeast direction. Figure 5. Container ship ACX post collision. (Source: Sea Quest Management) 8 A voyage management system is part of Navy vessels? electronic so?ware suite and is used for voyage planning and navigation. (Also see section 1.3.1.) NTSB Marine Accident Report The master of the ACX had been working on the bridge while the ship was in Nagoya and during its departure from there until 20 minutes past 2000, when he had transferred the conn to the third of?cer. He then completed paperwork, including sending an email at 2123, and shortly thereafter went below to rest. As the ACX continued its voyage toward Tokyo, about midnight, the third of?cer was relieved by the second of?cer, and an able-bodied seaman (AB) then on watch was relieved by the accident AB scheduled to stand the 0000?0400 watch. From midnight until the accident, the bridge was staffed by the vessel?s second of?cer in charge of the navigation watch, and the AB serving both as helmsman and, when the ship was in autopilot steering, lookout. About midnight on June 17, the ACX was eastbound near the southern tip of the Izu Peninsula, transiting at a speed of about 18 knots. Travelling nearly parallel to the ACX was the Singapore-?ag container ship Wan Hat 266, also headed to Tokyo Bay and about 2 miles north off the ACX port side. The Maersk Evora was also traveling nearly parallel to the ACX approximately 4 miles south and astem of the ACX The crew of the A CX reported no problems with the vessel?s steering and propulsion systems prior to the accident, and VDR data indicated no problems. 1.2 The Accident The Fitzgerald was equipped with AIS but was not transmitting data to other vessels. Per Navy practice at the time of the accident, naval vessels did not broadcast AIS information to other vessels but could receive information from other vessels. The itzgerald?s AIS was set to receive- only during the accident voyage. To track the destroyer?s position electronically, other vessels had to rely on visual means, radar, and automatic radar plotting aid (ARPA). 9 The destroyer was built by design to present a smaller target on radar displays than other (non-military) vessels of similar size. According to voyage data recorder (VDR) information obtained from the ACX about 0108, the Fitzgerald was about 12 miles away from the container ship. The Fitzgerald 00D told investigators that at that distance, she ?rst noticed two vessels on the bridge SPS-73 radar. She said she tried to ?hook? or electronically acquire the vessels but had trouble doing so. 1? According to the Coast Guard, ?The OOD stated the closest vessel was ACX which was being overtaken on her starboard side by a second vessel. Based on diagrams created by the GOD during the interview, along with information obtained ?om Voyage Data Recorders from ACX and Wan Hai 266, the vessel identi?ed by the DOD as ACX may have instead been Wan Hai 266.? However, the review of data from the three ships indicates that the DOD was likely acquiring the ACX and the Maersk Evora, with only intermittent identi?cation of the Wan Hat 266. The OOD told investigators that when the Fitzgerald was about 10 miles away from the ACX (about 0112, according to VDR data), she could Visually see the ship?s lights through the bridge windows, off the destroyer?s starboard bow. According to the CIC surface warfare 9 An ARPA is a computer?assisted radar data processing system that generates predictive vectors and other ship movement information. ARPAs calculate and display collision?related information and generate potential maneuver scenarios for operators to take to avoid collision with other vessels. Only acquired radar images will provide computed and informational data (such as AIS) on an ARPA. Acquiring a radar image on the ARPA can be accomplished automatically or manually. Automatic acquisition is accomplished, by the ARPA and without human interaction, after the operator sets time and distance parameters into the unit. Manual acquisition requires the operator to place an electronic cursor over the radar image and press a button. NTSB Marine Accident Report coordinator?the person monitoring and coordinating the overall surface picture, including all surface radar targets?he scanned his scope for contacts and then went to the restroom sometime between 0105 and 0110 and returned sometime between 0115 and 0120. He said that on his return, he again scanned his radar for targets but initially saw nothing, nor did his support personnel (the surface warfare supervisor and the optical sight system operator) report any targets to him. About 01 15, the second of?cer on the ACX began a scheduled course change per the vessel?s voyage plan. The ACX gradually turned from a mainly eastbound course of 089? to an east-northeast-bound course of about 069?. At 0119, when the course change was completed, the Fitzgerald was about 6.5 miles away, still approaching from the north off the container ship?s port bow. The second of?cer said he visually spotted a green light at a distance of about 3 miles, which later was determined to be the Fitzgerald. 11 He continued watching the vessel (both visually and on radar) as the Fitzgerald drew closer. Both he and the AB stated that the green [starboard] light was the only light they saw illuminated on the target. The Fitzgerald OOD told investigators that when the closer vessel was about 4 miles away, the destroyer?s ARPA provided a closest point of approach of 0.75 mile, with that vessel crossing astern of the Fitzgerald. A time-aligned overlay of Wan Hai 266 radar images, ACX radar images, and Fitzgerald VMS images between 0121 and 0125 indicated that the DOD used ARPA to acquire and track two eastbound vessels to starboard in this period. The closest vessel was the ACX and the second vessel was the Maersk Evora. The OOD told investigators that, about this time, she discussed the Fitzgerald?s distance to the two eastbound vessels with the JOOD, who until this time in the transit had predominantly been perfonning lookout duties and training the new conning of?cer in how to stand lookout. The JOOD told investigators that she went to the starboard bridge wing and looked at the ACX through a large set of binoculars called ?big eyes.? She said that there was a second vessel behind the ACX and that she urged the GOD to come look through the big eyes, which she did. The JOOD said she told the GOD to slow the destroyer?s speed, but that the DOD replied that a slowdown would make the situation worse. The OOD told investigators that she thought about turning to starboard and going astern of both vessels but decided against this maneuver because that course would take the destroyer closer toward land. At this time, the vessel was 8.2 miles offshore from the Izu Peninsula. According to the Fitzgerald?s deck log, at 0122, a course change was ordered to 200? (from 190?). About a minute and a half later, at 0124, ARPA information from the Wan Hai 266 showed that the destroyer had come right, to the new course. The reason for this course change was not revealed during post accident interviews. A study performed by the NTSB determined that the Fitzgerald would have passed 0.5 nautical miles ahead of the ACX if the Fitzgerald had remained on a course of 190? instead of changing to 200?. 12 11 According to international regulations, as a power driven vessel >50 meters in length, the Fitzgerald was required to display green and red navigational sidelights Visible at a minimum distance of 3 miles, a masthead light forward and a second masthead light abaft of and higher than the forward one, visible at 6 miles, and a stemlight visible at 3 miles. 12 The Vehicle Performance Study used GPS position data from the ACX and the Wan Hai 266 together with Fitzgerald course and bearing information from the Wan Hai 266 to reconstruct and evaluate the pre-collision paths of the ACX and the Fitzgerald. NTSB Fitzgerald Wanhai 266 I 4' ACX Marine Accident Report 4 Fitzgerald Wanhai 266 I '7 ACX 3 Maersk Evora mm Figure 6. A comparison of radar screens from different vessels before the collision. Left, screen capture from the Wan Hai 266 electronic navigation software, about 6 minutes before the collision. Right, screen capture from the ACX electronic navigation software (radar and ARPA), at the same time. (Vessel names have been added to screen images.) About the same time, the second of?cer on the ACX watched the Fitzgerald closely and when the destroyer was about 3 miles away, he checked his radar, looked at the destroyer?s green light, checked the radar again, and asked the AB to stand by the wheel. 13 The second of?cer told investigators he then went to the port side of the bridge and ?ashed a signal light in the direction of the Fitzgerald. According to the Coast Guard Report, this occurred at 0127:35, or about 2.5 minutes after the green light was ?rst observed, and about 3 minutes before the collision. The second of?cer told investigators that there was no reply to his signaling light. Shortly thereafter, he returned to the radar but then walked back to the port side of the bridge and ?ashed the signal light twice at the destroyer. He said he was expecting the Fitzgerald to turn because the ACX was the stand- on vessel in this crossing situation with the Fitzgerald to port, stating: ?you are the green light, you are supposed to alter course.? The distance between the Fitzgerald and the ACX was now closing at about 50 feet per second as the two vessels were moving toward each other at a speed of relative motion of more than 30 knots. The Fitzgerald?s surface warfare coordinator told investigators that after returning ?om the restroom to the CIC, he continued checking the radar and communicating with the optical sight system operator to ensure no surface contacts were in the destroyer?s immediate vicinity when suddenly he noticed a ?pop-up? radar contact on AIS. He said he directed the optical sight system operator to look in the direction of the contact. According to the optical sight system operator, who told investigators that he had visually seen about 80?100 targets in the general area during his watch, he spotted a ship on the given bearing and estimated it to be about 4?5 miles away (investigators later determined that this vessel was the Maersk Evora). The optical sight system operator said he 13 The second of?cer told investigators that he used the X-band, 3?cm radar, set to a 12?mile range. Its images were recorded by the VDR. NTSB Marine Accident Report continued turning his camera to the right and then saw the ACX He said he froze for about 8 seconds because he was stunned at how close the ACX was. The tactical action of?cer also saw the image of the ACX on the optical sight system operator?s camera but could do nothing in the short time between recognition and the collision. When the vessels were about 1,000 meters apart, the second of?cer on the ACX told the AB to apply 10? starboard rudder, then 15? starboard rudder, and ?nally hard to starboard. Fitzgerald (southbound) l. 001 1500 1&1 kt 7011800 1910 i 9012100 1920 I 2 01 2.100010 ,o'013000 19 kl Toshlma A CX {eastbound} Udone~shlma /'01 30 00 19.510 Maersk Evora (northeastbound) 01 27:00, 196 kt 6.00mi . :1 Google Earth Figure 7. NTSB reconstruction of the paths of the Fitzgerald, Wan Hai 266, ACX and Maersk Evora. (Background source: Google Earth) On board the Fitzgerald, the DOD had believed during the approach that the destroyer would clear both crossing vessels (the ACX and the Maersk Evora, or possibly the Wan Hai 266 and the ACX The OOD told investigators that she saw the ACX superstructure about a minute before the collision and she realized the Fitzgerald was not going to clear the containership. According to the GOD, she initially ordered the conning of?cer to come hard right, however, she cancelled that order before the conning of?cer could relay the order to the helmsman. The 00D said she then ordered hard left rudder and ahead ?ank speed. The BMOW, who was standing near the helmsman, stated that the GOD ?gave the order all ahead full for 25 knots, and right after that, all ahead ?ank.? He went on to say that when the OOD gave the order for ?hard le? rudder? he ?just grabbed the wheel? (from the helmsman) and ?put it over.? A review of Fitzgerald engine parametric data indicated that both the port and starboard engine throttle settings were simultaneously advanced by increments at 0130:06, 0130:14, and 0130222. Engine monitoring data indicated that both engines responded to the requested commands. At 0130:32, with the Fitzgerald traveling at 22.1 knots and the ACX at 18.4 knots, the vessels collided. Neither the Fitzgerald nor the ACX bridge teams sounded any alarms or made any announcements to warn their crews of the impending collision. 10 NTSB Marine Accident Report The ACX bow penetrated the itzgerald?s hull and superstructure, trapping numerous crewmembers aboard the destroyer, and the vessel took on a 7? starboard list. The itzgerald?s crew went to work to control the ?ooding, assist injured crew, and search for the missing. Seven crewmembers, trapped in their berthing compartment, perished. The destroyer sustained damage in excess of $300 million. Figure 8. Post collision damage to the Fitzgerald?s starboard side. (Source: US Navy) No one was injured on board the A CX but the container ship sustained damage to its bow and forward compartments. The vessel?s forepeak tank sustained a large gash, and a smaller gash was noted in the boatswain?s stores compartment. Penetrations to the hull were above the waterline. The cost to repair the ACX was not reported to investigators. ll NTSB Marine Accident Report ACX Figure 9. Postaccident damage to the ACX bow. (Source: Japan Transport Safety Board) The ACX crewmembers involved in the incident were not required to be toxicologically tested after the accident, nor was postaccident alcohol testing conducted on the Fitzgerald crewmembers. The Fitzgerald crew were tested for drugs on June 21, and the results were negative. 1.3 Navigation Equipment 1.3.1 Fitzgerald The Fitzgerald had two surface search radars. The SPS-73 radar was primarily used by bridge personnel, and the SPS-67 radar was mainly used in the CIC, but the bridge and the CIC had the capability to use both types of radar.14 The Fitzgerald had a voyage management system (VMS), which was used for voyage planning and navigation. VMS received inputs from numerous sensors 14 SPS-67 and radar are both short-range radar systems used for surface searches with minor technical capability differences. 12 NTSB Marine Accident Report and computers including the SPS-73 radar, GPS, and ARPA. This information was overlaid on the ECDIS-N VMS displays located on the starboard side of the bridge and in the CIC.15 The Fitzgerald was ?tted with AIS, which is a maritime navigation safety communications system that automatically transmits vessel information, including the vessel?s name, type, position, course, speed, navigational status, and other safety-related information, to appropriately equipped shore stations, other vessels, and aircraft. The AIS information is broadcast on a VHF frequency as dictated by international regulation. AIS also automatically receives information ?om similarly equipped vessels. Other vessel AIS information can be integrated and displayed on commercially available ECDIS and ARPA systems. The itzgerald?s AIS could be con?gured in broadcast or receive-only modes. In receive-only mode, the destroyer received AIS information about other vessels in the area on a dedicated laptop but did not transmit its own data. The laptop displayed information about other vessels and was capable of computing closest points of approach. According to the GOD, there was also a readout for the AIS laptop available, but the only information provided on the readout was the Maritime Mobile Service Identity (MMSI) number, course, and speed of the other ship. In addition, Fitzgerald crew members reported that the propulsion and steering systems of the destroyer were operating properly prior to the collision. 1.3.2 ACX Bridge equipment on the ACX included ECDIS and ARPA systems that received data from GPS, S-band and X-band radars, and other sources; a global maritime distress and safety system (GMDS S) suite with VHF radios; and an AIS. AIS is required on board all commercial vessels 300 gross tons or more operating on international voyages, and the ACX AIS was transmitting and receiving information at the time of the accident. ACX crew members reported that the propulsion and steering systems of the containership were operating properly prior to the collision. 1.4 Organizational Guidelines 1.4.1 Fitzgerald The Fitzgerald standing orders were derived primarily from two sources: general COLREGS and the Standard Organization Regulations of the Navy (SORM). The purpose of the SORM is to provide regulations and guidance governing the conduct of Navy personnel, and it applies to both Navy and Marine Corps shipboard personnel. Portions of the guidance could be tailored or ampli?ed to meet a particular vessel?s mission, task, and function. This included required reporting, conducting and relieving a watch, organization of at-sea watches, duties of and the relationships between them. The standing orders required the CIC to: 1) track and report (to the DOD) all surface contacts passing within a certain distance of the destroyer; and 2) for contacts with a CPA less than a speci?ed 15 is the Navy?s version of ECDIS. VMS is the software suite/sub?system of ECDIS-N. l3 NTSB Marine Accident Report distance, maneuvering board and Digital Dead Reckoning Tracer were to be used to provide course, speed, CPA, and time of CPA to the OOD. The standing orders also required the OOD to make sure that CIC performed these functions and that the OOD was to ensure that electronic aids and maneuvering boards were used, on the bridge, to independently calculate the same information that CIC was calculating. The standing orders also required the OOD to call the CO if the vessel deviated by a predetermined distance from the track line in the voyage plan. The night orders were supplemental to the standing orders and were issued each evening when the vessel was at sea. Night orders were tailored for the particular event/evening and could temporarily modify the standing orders to suit the current operations. In the night orders in effect at the time of the accident, the CO doubled the distance the vessel could deviate from the voyage plan track before the COD was required to notify him. Operational Risk Management (ORM) is a Navy?wide instruction instituted to identify and assess hazards associated with performing tasks and then managing, controlling, or mitigating risks associated with the speci?c tasks. 16 Operational risk management was part of Fitzgerald?s navigation brief for leaving Tokyo Bay. According to Fitzgerald ?s operations of?cer (who was also the accident tactical action of?cer) the coordination and scheduling of events for the destroyer, such as the night ?ight operation certi?cation held on the evening of June 16 and the engineering certi?cation scheduled for June 17? 18, was planned in conjunction with DESRON 15 personnel.? The ?nal schedule was ultimately approved by the Fitzgerald CO. 1.4.2 ACX The master?s standing orders called for the watch of?cer to maintain no less than a 1-mile closest point of approach to all vessels and to closely monitor vessels if the time to the closest point of approach was 6 minutes or less. If the watch of?cer could not maintain the l-mile distance, the master was to be contacted. The Master?s night orders for the run between Nagoya and Tokyo included requirements to apply the COLREGS when dealing with other vessels, allow 15 minutes for slowing down the vessel, and call the master if in doubt. According to the safety management manual of Sea Quest Management, the operator of the ACX the company was to ensure the safe and reliable operation of its ships and protect the environment. Speci?c sections of the manual included procedures to be used in special navigation situations and for risk management, and rules for safe navigation of the vessel. Per the rules for safe navigation, bridge management teams were listed for varying environmental and traf?c conditions, and there was a table documenting the personnel associated with manning Watch Levels 1, 2, and 3. The bridge manning on board the ACX at the time of the accident was Watch Level 1. 16 OPNAV Instruction 3500.39C, Operational Risk Management. 17 Destroyer Squadron 15 (DESRON 15) was the forward deployed squadron command, headquartered in Yokosuka, of a group of destroyers including the Fitzgerald. 14 NTSB Marine Accident Report According to the company safety management system (SMS), under Watch Level 1, the watch of?cer and AB were responsible for the safe navigation of the vessel. 18 1.5 Personnel 1.5.1 Fitzgerald When the Fitzgerald completed boat operations and started its outbound voyage on the evening prior to the accident, 293 crewmembers and 22 riders were aboard the destroyer. 19 On the morning of the accident, six bridge and seven CIC crewmembers had responsibilities associated with developing the surface contact plot or picture. All assigned personnel were quali?ed aboard the Fitzgerald in accordance with the Navy?s personnel quali?cation standards (PQS) for the positions they were assigned. In addition to the crewmembers already mentioned in the accident narrative, key personnel were: Commanding officer (CO). The CO was in overall charge of the vessel. He ?rst served as executive of?cer aboard the Fitzgerald from November 2015 through March 2017. He then went to various schools until May 2017, when he took over as C0 of the Fitzgerald. He had extensive experience in the accident waterway. Executive officer (X0). The XO was second in overall charge of the ship and was responsible to the CO for ensuring the vessel?s readiness. He was commissioned in the Navy in October 1999. After serving at shoreside commands for about 7 years, he joined the Fitzgerald in March 2017 and had made two trips in the accident waterway. 1.5.2 ACX As speci?ed on its safe manning certi?cate, the ACX was required to be manned by a total of 12 of?cers and ratings. On the accident voyage, the container ship carried a crew of 8 of?cers, 10 ratings, and 2 cadets. All were credentialed, with endorsements, by the Republic of the Philippines, in accordance with the Seafarer?s Training, Certi?cation, and Watchkeeping (STCW) Code and other International Maritime Organization (IMO) regulations. At the time of the accident, the second of?cer and an AB were on the bridge. Master. The master was in overall charge of the vessel. He held a certi?cate of competence to sail as master on vessels over 500 gross tons (indicating an unlimited credential), with the necessary endorsements ?om the Philippines. He had been a credentialed master for the past 9 years and said that he had made at least 5 trips in the accident waterway. His ?rst contract aboard the ACX was for 9 months. He started his second contract in April 2017. Second officer. The of?cer in charge of the navigational watch at the time of the collision, he held a certi?cate of competence to sail as of?cer in charge of a navigational watch on vessels over 500 gross tons, with the necessary endorsements. He joined the ACX in April 2017, and this 18 Watch Levels refer to the level of manning on the bridge of merchant vessels. Under Watch Levels 2 and 3, the master was in charge of the navigation of the vessel and the of?cer of the watch or extra of?cer were responsible for monitoring traf?c through the use of ARPA and radar. 19 A rider is a person not considered part of the vessel?s regular crew but who is aboard to perform a speci?c function. 15 NTSB Marine Accident Report was his ?rst contract aboard the vessel. He said that he had made many trips on the accident waterway. Able-bodied seaman (AB). The AB on watch during the accident provided investigators a Seafarer?s Registration Certi?cate for able-bodied seaman and had sailed in this position for about 10 years. As the AB on watch, he generally steered the vessel and performed lookout duties. He joined the ACX in August 2016, and this was his ?rst contract aboard the container ship. 1.6 Manning The manning of a Navy ship is designed to support varying Readiness Conditions (Conditions I-IV), as described in OPNAV Instruction 3501-31 1B. At the time of the Fitzgerald accident, the ship was in Readiness Condition OPNAV Instruction F35 01-3 1 1B describes this condition as Wartime, Increased Tension and Forward Deployed Cruising Readiness, wherein the ship shall be capable of meeting the following criteria: I) able to keep installed systems manned and operating as necessary to conform with prescribed Required Operational Capabilities; and 2) able to accomplish all normal underway maintenance, support and administrative functions. Additionally, the maximum expected crew endurance for Condition is 60 days, with opportunity for 8 hours of rest each day. ?Fill? is a Navy personnel term and is a measure of the number of personnel onboard who have the correct paygrade and occupational specialty (known as a rating) for the positions in which they are ?lling. ?Fit? is a higher measure and represents the number of personnel onboard with speci?c experience, training, and quali?cations required for the positions they are ?lling. Because Fit and Fill are normally less than 100 percent due to transitions, attrition, or other shortfalls, the Navy sets manning targets to ensure an even distribution of personnel across the force. According to the Fitzgerald ?s Manning Brief, dated June 19, 2017, the manning targets for the entire ship were 92 percent Fit and 95 percent Fill. According to the Navy?s Manning Control Authority Fleet (MCAF) Directive 15-1, ships outside of the continental United States (such as the Fitzgerald) are expected to be continuously manned to approved manning targets, but their actual manning levels on the accident voyage were 88.1 percent Fit and 93.6 percent Fill. 1.6.1 Fitzgerald Navy of?cers and enlisted crew were trained by shipboard and shoreside personnel and quali?ed by the CO (or a crewmember with delegated authority) under the Navy-wide PQS system developed and monitored by the Naval Personnel Development Command. The PQS mandated minimum standards and pro?ciencies for the personnel performing those functions. To qualify for a a crewmember must demonstrate knowledge and pro?ciency by completing tasks listed in a PQS booklet developed for that Individual units may tailor the Navy-wide PQS by adding new tasks to re?ect equipment carried on board but not covered in the booklet and deleting tasks that do not apply to the vessel. Investigators reviewed the PQS booklets for the GOD, the JOOD, the surface warfare coordinator, the CIC watch of?cer, and the CIC watch supervisor. Each person was required to understand the fundamentals and operation of radar equipment, including how to tune it, and to know the roles and responsibilities of the various members of the bridge and CIC teams. The 00D, the CIC watch of?cer, the CIC watch supervisor, and the surface warfare supervisor were required to 16 NTSB Marine Accident Report ?discuss? navigation rules (Inland and International Rules of the Road) and how they applied to collision avoidance situations. They were also required to pass a written examination. The 00D, the surface warfare coordinator, the CIC watch supervisor, and the surface warfare supervisor were also required to demonstrate the use of a MoBoard (maneuvering board) to determine risk of collision per their respective PQS processes. When a PQS quali?cation was obtained by an individual, it was valid for the platform/ship on which the person was serving. When the individual transferred to a different ship, he or she had to requalify aboard that vessel. According to the surface warfare coordinator, the previous CO required weekly training on rules of the road for the entire wardroom;20 however, if one attained a 100 percent score on the exam, there was no requirement to attend future training sessions or take additional exams. So weekly rules training fell mostly to those junior of?cers who were qualifying to become 00Ds. The navigator told investigators that all quali?ed 00Ds were required to attend rules of the road training on a quarterly basis. The new C0 had only been on the Fitzgerald in the role of commanding of?cer a few weeks, so it was not known if he also had a plan for rules of the road training. The Fitzgerald had spent the majority of 2016 in the shipyard, and so, the C0 told investigators that when he came into command, he knew that his bridge and CIC teams would be relatively less experienced when they got under way. He and his team spent ?no less than 6 provisions on the watch bill? prior to getting underway the day before the accident because the C0 wanted to ?place seniority in controlling watch stations. . .and ensure that there was suf?cient levels of force?il backup.? He explained that he had a lot of young ?rst-tour division of?cers going through the quali?cations and that he had not personally observed their performance, nor had he been able to assess their abilities as a whole. Although the C0 had con?dence in the crewmembers? quali?cations, he put measures into place to combat their inexperience in order to ?increase navigation and seamanship, level of knowledge, and watch team performance? while in the shipyard. Examples of these mitigating factors included having the teams focus on basic ship handling fundamentals, preparing navigation briefs, and using paper charts. The C0 said he would then provide scenarios in which the crewmembers would use local nautical charts in restricted water transits, implementing skills they were learning. He said that he considered the performance of his bridge watch to be ?satisfactory? and noted no areas of de?ciency that required improvement. When the X0 described how he built the watch teams, he also said that they went over the watch schedule several times. The X0 stated in his interview that he did not trust the 00D but felt that, given the surface warfare coordinator?s experience, he was comfortable with the watch team because the coordinator could back up the bridge team although he was not physically on the bridge. In contrast, when the surface warfare coordinator was asked about his involvement with the bridge, he said, ?my involvement with the bridge is somewhat limited.? The X0 described the 00D as ?below average? and said that she had been reprimanded for actions in the CIC in which he found her behavior indicative of overcon?dence. When asked why he did not express his concern to the C0, he stated, ?because I didn?t provide support and backup, I just didn?t. I think back and I don?t know why I 20 Wardroom is the ship?s dining area where commissioned of?cers eat. When used as a description for crew, it generally means all commissioned of?cers of the vessel and not the enlisted crew. 17 NTSB Marine Accident Report didn?t. I think that I trusted the Captain?s judgment too, more than my judgment on her performance. You know, I thought that maybe there is an error of what I was seeing compared to what he was seeing.? When asked why, if he didn?t trust the DOD on the bridge, he did not volunteer to stay up, he replied, would say that I was tired at that point and I probably wasn?t thinking about what was going to happen next. I was thinking more of ?what are we going to do at 0600 when we start the engineering drills? and not getting from midnight 2330 until 0600.? 1.6.2 ACX As a commercial vessel engaged in international trade, the ACX was subject to the STCW Code and other IMO regulations. Among other provisions, the STCW Code stipulates requirements pertaining to crew experience, training, and assessment. For deck of?cers who held licenses for unlimited tonnages, the training included bridge resource management (BRM), radar and ARPA operations, ECDIS, and rules of the road. Further, when a licensed of?cer renewed his mariner credential at the mandatory 5-year interval, radar and ARPA pro?ciency needed to be re- tested. In addition, all merchant personnel are required, by regulation, to obtain familiarization training for the equipment they will be operating prior to assuming the duties they are tasked to perform. 1.7 Work/Rest History 1.7.1 Fitzgerald Investigators reviewed the number of rest hours for the Fitzgerald bridge team during the 24 hours leading up to the accident. The 00D awoke at 0400 on the morning before the accident, prepared for getting under way, stood watch leaving the dock, was the safety of?cer during the ammunition loadout, did some administrative work, and took a 1-hour nap before heading to her 2200?0200 OOD watch. The JOOD, who had been working since 0700 on the morning before the accident, said she also took a 1-hour nap before heading to the bridge for the 2200?0200 watch. According to the 96-hour work/rest record for the surface warfare coordinator, he slept for about 7 hours on the evening of June 15 and about 3.5 hours prior to standing watch on the evening of June 16. The surface warfare coordinator stated that he became fatigued during the last hour of his watch. The optical sight system operator was reportedly so tired that he was falling asleep during the watch, which prompted his supervisor to insist that he stand up while working. The Fitzgerald crewmembers were placed, by position, into watch sections to man bridge, engine, and CIC watch positions (?stations?). The number of crew that stood watch, at any given time, was dependent on the watch condition of the vessel. Watch conditions were set for various operations or threats to the vessel. The number of watch sections for each position depended on the number of quali?ed crewmembers for that position. Each position had at least three watch sections. If a suf?cient number of crewmembers were available, presumably well rested and quali?ed, a fourth watch would be added to ease the burden on the other watch sections. During the accident voyage, Fitzgerald crewmembers were assigned to the bridge covering ?ve periods: 0200? 0700, 0700?1200, 1200?1700, 1700?2200, and 2200?0200. In this watch organization, the period for each watch team would shift with each cycle of the watch rotation. For example, the watch team that had the 0200?07 00 watch would next have the watch from 2200?0200 that evening, and then have the watch from 1700?2200 on the following day. Following this accident 18 NTSB Marine Accident Report and the McCain/Alnic MC collision,21 which occurred about 2 months later in August 2017, the Navy mandated ?circadian watch bill? schedules that followed set watch times each day?"23 1.7.2 ACX The crew of the ACX was required to adhere to STCW Code regulations for rest, which mandated a minimum of 10 hours in any 24-hour period and a total of 77 hours in any 7-day period. The 10 hours of rest could be divided into two periods, as long as one of the periods was at least 6 hours long. According to records reviewed by investigators, the ACX ?s second of?cer had 49.5 hours of rest in the 96 hours before the accident and 12.5 hours in the preceding 24 hours. His rest period just before the accident was 9 hours long. No 96-hour report was provided for the AB or the master. Investigators reviewed the ACX ?s schedule and noted that the container ship visited three ports on a daily basis commencing on June 13, and the vessel was scheduled to call at Tokyo on the morning of June 17 . The ACX had three bridge watch teams assigned to the six traditional periods: 0000?0400, 0400?0800, 0800?1200, 1200?1600, 1600?2000, and 2000?2400. Under this organization, each watch team stood the same two watches each 24-hour period. 1.8 WatenNay Information Sagami Nada bay, where the collision took place, lies off the Izu Peninsula on the southeast coast of Honshu Island. At the northeast corner of Sagami Nada lies the Uraga Suido strait, which leads directly into Tokyo Bay. The overall length of Sagami Nada is about 55 miles and its narrowest point is about 20 miles. The accident occurred near the southern terminus of Sagami Nada, about 5 miles off the southeast tip of the Izu Peninsula. National Geospatial-Intelligence Agency Publication 158, Sailing Directions (Enroute) for Japan, Vol. 1, contains general information about the accident area ports, waterways, traf?c separation schemes, and traf?c encountered within the ports. It also directs the reader to consult Sailing Directions 120 (Planning Guide) for additional information about the waterways treated in Publication 158. According to the National Geospatial?Intelligence Agency 2015 Planning Guide to the US. Sailing Directions for the Paci?c Ocean and Southeast Asia, the accident waterway is noted for its heavy traf?c and frequent collisions.24 21 Collision between US Navy Destroyer John McCain and Tanker Alnic MC, Singapore Strait, 5 Miles Northeast of Horsburgh Lighthouse, August 21, 201 7. Marine Accident Report Washington, DC. 22 Commander, Naval Surface Forces messages to surface ?eet units, ?Circadian Watch bill,? May 3, 2013, and ?Force-Wide Circadian Implementation,? September 20, 2017. 23 See section 2.3.3: Navy Actions since the Accident, for more information on postaccident actions taken by the Navy. 24 National Geospatial-Intelligence Agency Pub. 120, 2015, page 210. This publication is free and available on the intemet site. This publication was referenced in NGA Pub 158, the Sailing Directions (Enroute) Japan, Vol. 1. l9 NTSB Marine Accident Report 1.9 Environmental Information According to numerous crewmembers of both vessels, the visibility on the morning of the accident was good. The Fitzgerald OOD estimated the Visibility as between 10 and 11 miles and the sea state as about 1?2 .25 The ACX crew logged the weather as cloudy skies with about 10 miles of visibility. The ACX 0200 deck log entry noted north winds at 11?16 knots, air temperature of and sea temperature of Data retrieved from the Naval Oceanographic Of?ce Global Hybrid Coordinate Ocean Model and the NOAA Wavewatch 111 model for the accident site around the time of the collision indicated air temperature near sea temperature at north-northeast winds at 16.67 knots, and a wave height of 2.4 feet. Moonrise was at 2332 on June 16; the phase of the moon was waning, and it was in its last quarter. 25 According to the World Meteorological Organization, a sea state of was for seas between 0 and .33 feet and a sea state of 2 was for seas between .33 feet and 1.64 feet. 20 NTSB Marine Accident Report 2 Analysis 2.1 Exclusions At the time of the accident, the sea state was 1?2 and visibility was about 10 miles. The crews from both the ACX and the Fitzgerald reported no concerns related to the weather. In addition, the crew of the ACX reported no problems with the vessel? steering and propulsion systems prior to the accident, and VDR data indicated no problems. Similarly, records from the Fitzgerald indicated that steering and propulsion were tested prior to the accident and were found in good working order. The only reported equipment mal?inction on the Fitzgerald bridge or CIC was an inoperative radar repeater (the SPA-25) in the CIC. It was shut down completely and played no role in the accident. The NTSB therefore concludes that weather and the steering and propulsion systems on board the Fitzgerald and the ACX were not factors in the accident. Because investigators were provided only limited toxicological results for the Fitzgerald crew and no test results for alcohol or other drugs for the ACX crew, the evidence was insuf?cient to determine conclusively whether alcohol or other drug use was a factor in this accident. 2.2 Actions to Recognize and Avoid the Collision 2.2.1 Fitzgerald As the Fitzgerald proceeded in a southbound direction on the evening of the accident, the DOD picked up two radar contacts (the ACX and the Maersk Evora, or possibly the ACX and Wan Hai 266) off the itzgerald?s starboard bow at approximately 12 miles. The two vessels were on a parallel path on an east-northeasterly heading following normal traf?c patterns toward Tokyo. Per the COLREGS, when two vessels are crossing, the vessel that has the other on its starboard side shall keep out of the way of the other vessel. The Fitzgerald had the ACX the Maersk Evoraits starboard side and therefore was the give-way vessel to all three ships. The regulations required the Fitzgerald, as the give-way vessel, to take early and substantial action to keep clear of the stand-on vessels, and to avoid crossing ahead of the other vessels. The advisable course of action in this circumstance was for the Fitzgerald to alter course to starboard and pass well clear of all three vessels. Rather than taking early and appropriate action, such as turning or slowing to avoid the three vessels as required by the navigation rules (and the standing orders), the bridge team on the Fitzgerald continued ahead and ultimately crossed ahead of the Wan Hai 266 and placed the vessel in the path of the ACX resulting in the collision. Although there was room to pass ahead of the Wan Hai 266, doing so was not prudent given the presence of the ACX The actions of the bridge team on the Fitzgerald did not conform to the collision regulations as they approached the Wan Hai 266 and ACX Therefore, the NTSB concludes that the Fitzgerald bridge team, on the give-way vessel, did not take early and substantial action to avoid collision with the stand-on ACX Navy OODs are required to pass a rules of the road test, and each vessel typically has its own requirements for carrying out recurring rules of the road training. Based on the circumstances of this accident, there is a clearly demonstrated need for review of the Navy?s training for its 21 NTSB Marine Accident Report related to the collision regulations. Therefore, the NTSB recommends that the US Navy review and revise ?eetwide training and quali?cation requirements for OODs related to the collision regulations. Given that the DOD on the Fitzgerald identi?ed the two vessels at 12 miles out and about 22 minutes before the collision, investigators sought to understand how the accident could have occurred given the numerous resources available to the Fitzgerald bridge team and the time available to take appropriate action. The bridge team was made up of six personnel. The bridge team was operating under the standing orders, which, among other things, provided requirements for notifying the CO regarding vessel traf?c. Further, the bridge team was supported by in the vessel?s CIC who were charged with tracking surface contacts and reporting to the bridge. The bridge team is expected to capture and prioritize critical information, analyze that information, and take appropriate action. Training, experience, and standard operating procedures are developed to mitigate errors, particularly in dynamic situations. Although the Fitzgerald OOD identi?ed the two crossing vessels on the vessel?s starboard bow at 12 miles? distance on the bridge radar (about 0108), she told investigators she was unable to acquire them at that time using ARPA to determine CPA. When the contacts were ?rst identi?ed, the GOD had suf?cient time and could have made use of the resources available to her, including the other bridge and CIC. Bridge resource management (BRM) makes use of all available resources, including equipment and information and human resources, to achieve safe operation. Good communication amongst a bridge team is key to successful BRM. The OOD should have communicated with the JOOD about the vessels and asked her to monitor them. The 00D should also have requested assistance from CIC, where numerous were available to track the vessels using various means. According to the standing orders, the CIC was tasked to support the bridge team, and the tactical action of?cer, as head of the CIC, was to ensure that this occurred. The 00D, per standing orders, was to ensure that the CIC provided the bridge team with information so that the vessel could be safely navigated. During interviews of several of the bridge and CIC investigators learned that little communication took place between members of the bridge team and CIC even though the bridge team had identi?ed vessel traf?c and CIC members saw surface contacts on the optical sight system or were aware of the vessel changing course to avoid traf?c. According to Crouch (2013), 60 percent of all marine accidents that fault the human operator are related to ineffective communication. The incorporation of BRM encourages junior of?cers to be an integral part of a team, putting forth their opinions and, when safety is of concern, challenging their superior of?cers. Senior of?cers, in turn, should effectively be more open to gathering feedback from the bridge team. Successful implementation of BRM would ensure that the team is communicating about traf?c in the area and should provide a means for anyone to speak up if they believed there to be an issue. It is unclear whether the Fitzgerald crew, at the outset of their watch, completed a comprehensive turnover to ensure they all understood the intended plan, the night orders, or the high-traf?c area in which they were about to transit. The tactical action of?cer stated that, after leaving the Tokyo Bay area, no discussion took place about the navigation transit, tracks, or likely high-density marine traf?c, and that these discussions were not normally held. Discussion amongst a group allows for different viewpoints and different experiences to be shared. Additionally, it allows the group to collaboratively identify risks, discuss mitigation of those risks, and prepare for the upcoming activities. The lack of communication and failure to use available resources is indicative of a poorly functioning bridge and CIC team. The NTSB concludes that the 22 NTSB Marine Accident Report communication and cooperation among the Fitzgerald crew on the bridge and in the CIC were ineffective leading up to the accident. The NTSB notes that all bridge of?cers on seagoing merchant vessels are required to satisfactorily complete formal BRM training, provided by quali?ed trainers, as part of their licensing requirements. Instilling the concepts of BRM on a regular basis is instrumental in ensuring that the crew understand the challenges of communicating effectively in the bridge and CIC environments. Accidents such as this could serve as examples of how to apply these concepts more appropriately. Given the lack of communication among the bridge team and CIC the NTSB recommends that the US Navy review and revise BRM training in its approved course curriculum to promote a cohesive team environment and improve communication within and between bridge and CIC teams. When the 00D reported the contacts to starboard, the GOD had already acquired the ACX using ARPA and determined the CPA to be 1,500 yards (0.75 nautical mile), with the ACX expected to pass astem of the Fitzgerald. The 00D told investigators she was comfortable with this distance and began looking at the next vessel (Maersk Evora) to determine CPA. Despite the CPA for the ACX being within the reporting requirements so. Instead, she said she planned to do an ?after the fact report.? This deliberate contravention of the explicit instructions is troubling. Had she followed the standing orders and alerted the CO, it is likely that he would have required the GOD to take action and the accident could have been averted. At 0120, when the OOD identi?ed the vessels to starboard, she recommended that they slow the vessel. The OOD did not agree and said she thought it would make the situation worse, although she did not provide any substantive reason she believed this to be true. The OOD also told investigators they discussed turning to starboard but that she was concerned that they would be heading toward land. At this time, the vessel was 8.2 miles from the Izu Peninsula and therefore any temporary course change toward land would not have placed the vessel at risk. Further, there were no contacts to starboard that would have posed a problem had they turned. With the maneuverability of the Fitzgerald, a substantial turn to starboard or a reduction of speed would have allowed the Fitzgerald to pass safely behind the ACX 23 NTSB Marine Accident Report Fitzgerald/ACX Path Reconstruction: Positions at 0130:00 2 Fitzgerald course change 0124 1.5 ill/alt Hal 265 0130. 166- kt 1 Fitzgerald actual path Fitzgerald path holding previous course 0.5 2 Fitzgerald 8 0130, 19ACX 0130, 17.8 kt 0-1 . 415 Time: 0130:30 0 -1 Fitzgerald ACX -0?1 Wan Hai 266 0 0.1 -1.5 -1 -0.5 0.5 1 1.5 Distance East, nm Figure 10. NTSB path reconstruction for Fitzgerald and ACX from data compiled from NTSB Vehicle Performance Study. The Fitzgerald?s deck log shows that a course change was ordered at 0122, and the vessel came right from 190? to 200?. They had steadied on the new course by 0124. This course change proved to be a critical error, and investigators were unable to determine the reason for it. The NTSB conducted a Vehicle Performance Study using GPS data as well as course and speed data for the Fitzgerald, ACX and Wan Hat 266, to recreate the tracks of the vessels leading up to the accident. The study found that if the Fitzgerald had not made the course change from 190? to 200? 8 minutes before the collision, the destroyer would have passed ahead of the ACX with a CPA of about 1,000 yards, or about 0.5 nautical mile. Therefore, the NTSB concludes that the 24 NTSB Marine Accident Report itzgerald?s unexplained small course change to starboard minutes before the collision put the vessel on a collision course with the ACX Seven in CIC were tasked with monitoring the surface picture and supporting the COD and bridge team by providing radar contact information to the bridge watch. Numerous CIC personnel told investigators that no contacts were close enough to the Fitzgerald to warrant plotting; however, when investigators reviewed ECDIS screenshots from the Fitzgerald? VMS, they noted numerous contacts within the tracking and reporting requirements. On the night of the accident, this same information was available in real-time both on the bridge and in the CIC. At least some personnel in CIC did see contacts but did not apparently share this information with other CIC The optical sight system operator, who reported to the surface warfare coordinator as part of the surface watch, told investigators that he visually observed at least 80 contacts using his line of sight camera. The surface warfare coordinator stated that pop-up targets would appear on his radar scope and the surface warfare supervisor?s scope had a lot of clutter. Despite the clutter, however, he did not advise any of his supervisors so a technician, who was part of the watch bill, could look at the scope to determine if the issue was related to equipment tuning or operator error. The COLREGS call for every vessel to use all available means appropriate to determine whether risk of collision exists. The Fitzgerald had both radar and SPS-73 radar, but only the radar sent data to the VMS. Neither the surface warfare coordinator nor the surface warfare supervisor switched from the reportedly cluttered radar to the SPS-73 radar or looked at the VMS console, which was overlaid with targets acquired by the SP S-73 radar. Had they done so, they would have seen the targets that were populating the VMS. Not using the SPS-73 radar deprived the surface warfare coordinator from the use of another tool to determine if risk of collision existed. If the tactical action of?cer or surface warfare coordinator had used the SP S-73 radar, which provided contact information to the VMS and con?rmed the presence of vessels near the Fitzgerald, the surface watch team could have tracked and provided collision avoidance information to the GOD. In addition, the AIS information, which would have assisted in the tracking and identi?cation of traf?c, was available to the CIC In addition, although the tactical action of?cer, who was in charge of the CIC, told investigators that role was to support the bridge team, she did not require the surface warfare supervisor or the surface warfare coordinator to track and report contacts closer than a certain distance to the Fitzgerald, despite the standing orders to do so. The NTSB therefore concludes that the Fitzgerald CIC personnel did not effectively support the bridge team in tracking nearby surface targets. 2.2.2 ACX A little after 0100, the ACX was on an easterly heading off Japan?s southeast coast, heading for an 0500 pilot station arrival near Tokyo Bay and bound for Yokohama after departing Nagoya earlier on the evening of June 16. Review of the ACX VDR capture of the X-band radar showed that about 0106 on the morning of June 17, a faint radar return was visible of what was later determined to be the Fitzgerald. It ?rst appeared when it was approximately 13 miles away and along with the faint target 25 NTSB Marine Accident Report was a following echo or radar trail.26 As the target (Fitzgerald) and the ACX continued to close, the target signature could be seen appearing and disappearing on the radar. Other vessels? signatures at the same distance and further away had been acquired on the ACX ARPA and were being tracked, but the ARPA did not display any information about any of them, such as course or speed. By 01 15, the target signature was steady on each radar capture recorded by the ACX 3 VDR (about every 15 seconds). About 0119, the signature displayed radar trails as the ACX completed its course change to port to 069? per the ship?s voyage plan. After the ACX course change, the Fitzgerald was about 6.5 miles away off the container ship?s port bow. The container ship Wan Hai 266 was travelling east on a parallel path to the ACX and about 2 miles north (closer to the Fitzgerald). Investigators reviewed Wan Hai 266?s captured ARPA and VDR information and compared it to the ACX At 0114, the Wan Hai 266 acquired the Fitzgerald on its ARPA and held the target along with displayed information up to and through the collision with the ACX When the Wan Hai 266 acquired the Fitzgerald, the destroyer was 7.3 miles away on a course of 188? true and making 19.8 knots. Although the Wan Hai 2 66 was about 2 miles closer to the Fitzgerald, the fact that the container ship could acquire the Navy vessel, which was built purpose?illy to minimize its radar signature and thus produce a smaller radar return than similar-size merchant vessels, meant that the second of?cer on the ACX could also have acquired the Fitzgerald at a similar distance on the ACX ARPA. The steady target that the Fitzgerald displayed on the ACX radar, and the radar trails that were emanating from the target, show what should have been a clear target of concern as it presented a risk of collision based on the track of the container ship. Prudent seamanship dictates that the target should have been acquired to more accurately determine risk of collision. Acquiring the target on the ARPA would have provided the Fitzgerald?s true course, speed, CPA, and TCPA (Closest Point of Approach and Time to CPA respectively). Although the second of?cer on ACX indicated that he was aware of the vessel (Fitzgerald) after he turned the container ship onto the 069 track, he never acquired it. The NTSB concludes that the Fitzgerald provided a detectable radar signature, and the second of?cer on board the ACX should have acquired the destroyer on the ARPA to determine risk of collision. When the second of?cer and the AB on the ACX ?rst visually spotted the green (starboard running) light of the Fitzgerald, they consulted the radar and determined that the approaching vessel was about 3 miles away. Investigators reviewed the radar display, which showed that the target (Fitzgerald) was only about 5 minutes away from collision. The second of?cer stated the ?rst course of action he took was to grab the signaling lamp and ?ash it at the green light. According to the Coast Guard Report, this occurred at 012735, or about 2.5 minutes after the light was ?rst observed, and about 3 minutes before the collision. The second of?cer told investigators that there was no reply to his signaling light. The second of?cer observed the approach of the vessel by visually checking its green light and looking at the radar. The distance between the vessels was closing rapidly due to the close 26 A radar echo or trail is an afterglow of a contact?s past positions that follows a contact displayed on a vessel?s radar screen. The direction and length of the trail gives the operator an indication of the relative motion of the vessel being tracked by radar. 26 NTSB Marine Accident Report proximity and the speed of approach of both vessels (at almost 19 knots each). Flashing the lights indicated that the second of?cer was in doubt. Left with a potential collision situation developing, although as the stand-on vessel he was supposed to hold course and speed, the second of?cer on the ACX could take action when it became apparent that the give-way vessel was not taking enough action to avoid collision. He could have tried to radio the oncoming vessel, but with no name or way to identify the Fitzgerald, several vessels (or no vessels) might have responded to the broadcast. Further, in addition to signaling with a lamp, he should have sounded the whistle signal with ?ve short blasts to indicate doubt, per the COLREGS. No one aboard the Fitzgerald recalled seeing the ?ashing light. It is possible that a whistle signal from the ACX may have alerted someone on the destroyer of the danger and to take evasive action. Although the second of?cer did order a series of starboard rudder orders which began changing the vessel?s heading to starboard at 0129:25 (about 1 minute before the collision), his effort was not early or aggressive enough to avoid the collision. Therefore, the NTSB concludes that when the second of?cer on the ACX did not receive a response to his signaling attempts and saw no noticeable change in approach from the destroyer, he did not take suf?cient action to avoid the collision. The ACX deck and engine of?cers were trained and credentialed for the capacity in which they were working. However, as noted above, the NTSB found that the second of?cer did not adequately use all means to assess risk of collision, and did not take suf?cient action to avert the collision. Because a well-trained navigation team is essential for safe operations, the NTSB recommends that Sea Quest Ship Management, Inc., provide additional training for its navigation of?cers on collision avoidance regulations, radar, and ARPA. Assessing risk of collision can be and has been successfully accomplished without AIS. However, the worldwide, mandatory use and acceptance of AIS as an added tool to navigators? ?all available means? can assist in assessing risk of collision. AIS is required on most commercial vessels. Both the Wan Hai 266?s radar and the ACX radar depicted the itzgerald?s radar signature as signi?cantly smaller than that of a comparable merchant vessel of the same size. The destroyer?s design included features meant to reduce its radar signature. Investigators? review of the ACX ?s VDR radar images showed more than two dozen other larger-size targets acquired on the radar. In addition, there were several smaller targets, displaying the same size signature as the Fitzgerald, that were not acquired. It could not be determined why the second of?cer chose to acquire some targets and not others. Although the second of?cer was using the radar to plot and acquire even smaller targets so that a full appraisal of the situation and of the risk of collision could be made, it is likely that an AIS tag attached to the signature of the destroyer would have helped the second of?cer realize that the small radar contact was a US Navy destroyer. AIS would have identi?ed the vessel so the second of?cer could have called the destroyer by name on bridge-to-bridge radio and given the second of?cer more information about the true size of the approaching vessel. One of the key purposes of AIS is to identify other vessels. By not broadcasting AIS, the Fitzgerald took away one of the primary means of the ACX to identify and detect the destroyer. Therefore, the NTSB concludes that the absence of an AIS signature broadcast from the Fitzgerald likely contributed to the lack of its early detection by the ACX bridge team. The NTSB report on the McCain/Alnic MC collision also identi?ed that the transmission of AIS information from the McCain would have improved the situation awareness of on surrounding vessels. After that accident (which occurred after the itzgerald/ACX 27 NTSB Marine Accident Report collision), the Navy directed all of its ships to transmit AIS data when transiting Traf?c Separation Schemes and other high-density traf?c areas. However, the Fitzgerald standing orders already required that they transmit ?when we are transiting a Traf?c Separation Scheme, high traf?c density areas,? yet the Fitzgerald was not transmitting prior to the accident, which indicates they may not have recognized the area as high traf?c. The Navy?s guidance does not de?ne high traf?c, requiring individual COs to subjectively determine when to broadcast. Further, the NTSB believes that AIS can assist in preventing collisions because it provides use?il information (such as the name, course, speed, CPA, and range of the vessel) that is displayed on another vessel?s radar and/or ECDIS display. This information is bene?cial, regardless of the amount of traf?c or the waterway the vessel is operating in. The NTSB therefore recommends that the Navy instruct its vessels to broadcast AIS information while operating in the vicinity of commercial vessel traf?c at all times unless such broadcast compromises tactical operations or strategic interest. 2.3 Operating Procedures According to Navy risk management policy, ?all activities, commands, personnel, and contractors under the direct supervision of government personnel shall apply operational risk management principles . . . Investigators noted that the predeparture navigation brief for the Fitzgerald in Yokosuka contained the treatment of operational risk management consistent with Navy policy and that the brief was discussed by wardroom personnel before getting under way. However, investigators did not ?nd operational risk management plans for the voyage plan, outbound to sea, after completing boating operations and numerous bridge turnovers. The XO stated that he and the navigator discussed the transit south and then the navigator (separately) discussed the transit with the CO, but no brief was held, similar to the navigation brief for leaving Tokyo Wan, to discuss the transit south. The tactical action of?cer stated that there was no discussion of the navigation transit, tracks, or likely high-traf?c areas and that these discussions were not normally held. Discussion during a brie?ng among crewmembers, as a group, allows for different viewpoints and different experiences to be shared about the upcoming transit. Additionally, it allows the group to collaboratively identify risks, discuss mitigation of those risks, and prepare for the voyage. Had they held this brief, they may have recognized that their intended path would be crossing the ?ow of traf?c in multiple areas. A likely mitigation for this risk would be assigning a senior of?cer to remain on the bridge to assist with navigating traf?c. There were multiple of?cers available, including the X0 and CO, with more experience. Despite the C0 doubling the allowed deviation from their track line, the DOD still had to call the CO for traf?c with a CPA of less than 3 miles, and should have called at least three other times (but did not). The NTSB concludes that the Fitzgerald CO did not adequately assess the hazard presented by the vessel?s intended transit, which crossed busy coastal traf?c routes, and should have assigned a more experienced of?cer to augment the bridge team. 2.3.1 Fatigue Numerous watch personnel on board the Fitzgerald had collateral duties on the day before the accident. Though this type of workload is not uncommon, it is worth noting, given the high tempo of operations leading up to the accident. The OOD was the boat operations safety of?cer; the tactical action of?cer was the operations of?cer and, after completing work in those roles, both went on watch. With the high workload, the crew on watch at the time of the accident had minimal time to rest. The tactical action of?cer stated she had no rest. The 00D stated that she woke at 0400 the day before the accident and took about an hour-long nap before going on watch at 2200. In the CIC, the 28 NTSB Marine Accident Report optical sight system operator was reportedly falling asleep, which prompted his supervisor to insist that he stand up while working. Most people will experience fatigue with less than 7?8 hours of sleep in any 24-hour period; sleeping less than 7?8 hours in any 24-hour period leads to acute fatigue, whereas habitually obtaining less than 7?8 hours of sleep results in accumulated sleep debt leading to chronic fatigue. In addition, the accident occurred just prior to a time period considered to be a circadian low (roughly 0200?0600), when the body is normally more fatigued and prone to diminished alertness and degraded performance. Altering sleep periods (such as sleeping primarily during the day for a period of time and then switching to sleeping at night and vice versa) impacts the quality of sleep, which compounds fatigue related to circadian lows. The surface warfare supervisor missed several critical targets, and the GOD made some poor navigational decisions and did not request support from the CIC. Further, the tactical action of?cer (in charge of the CIC) did not ensure that her personnel supported the bridge team. All of these had little or no sleep before heading to watch. The NTSB concludes that several Fitzgerald bridge and CIC were likely acutely fatigued at the time of the accident, which impacted their situation awareness and ability to identify and respond to the approaching ACX As stated in the NTSB report about the 2017 McCain/Alm'c MC collision: Prior to the accident, the Navy did not actively address fatigue among crews on Navy vessels. (The Navy has longstanding requirements addressing fatigue and crew rest on Navy aircraft?) In a 2013 message, the Navy recommended that its surface force ships implement circadian watch bills that allow to establish sleep patterns resulting in adequate rest. Following this and other accidents in 2017, the Navy directed that surface force ships implement the circadian watch bills recommended in 2013. While this directive is a step in the right direction, the primary focus of this initiative is on the watch schedule. It does not fully consider overall work and rest, which is the most critical factor in addressing the fatigue faced by Most notably, the 2013 and 2017 messages do not address nor prescribe mandatory rest periods similar to the STCW Code requirements of commercial mariners. STCW Code requirements were the result of regulator and shipping industry efforts to develop a work/rest balance that addresses human needs and safety while meeting the demands of the maritime environment. Considering activities such as weapons handling and underway replenishments, Navy operations are as dangerous, if not more dangerous, than commercial operations. While combat and other non-standard operations may require crewmembers to forego adequate rest for short periods, relying on fatigued crewmembers to accomplish normal, daily tasking is unnecessarily dangerous. Given the impact that fatigue had in the McCain collision, the NTSB issued Safety Recommendation M-l9-l4, recommending that the US Navy institute Seafarer?s Training, Certification, and Watchkeeping Code rest standards for all crewmembers aboard its vessels. In 27 Of?ce of the Chief of Naval Operations, NATOPS General Flight and Operating Instructions, OPNAV Instruction 3710.7U, Washington DC: Department of the Navy, 2009. 29 NTSB Marine Accident Report response to this recommendation, the Navy stated they were taking further actions to address crewmember fatigue. Building on their implementation of circadian watchbills provided in the Navy?s Crew Endurance Policy, commanding of?cers are now required to account for fatigue as part of their deliberate planning and risk analysis. In their response to the NTSB recommendation, the Navy stated that they planned to implement crew rest standards that reasonably compare with Seafarers? Training, Certi?cation and Watchkeeping Code (STCW) standards for a 7-day period. STCW crew member requirements re?ect a minimum of 77 hours rest within a 7-day period. Assuming a 7-day underway schedule, the Navy now speci?es 87 hours of ?non-available? (off- duty) time within a 7-day period. The Navy?s Fatigue Management Policy requires enhanced training and planning. According to the Navy, commanding of?cers review crew watchbills to ensure they meet circadian and watch length/crew day requirements of the Fatigue Management Policy, review the ship?s published routines, and observe their actual routines in execution to ensure the ship?s routine supports achieving the crew rest standards. Further, the Navy has established an ongoing collaboration between the Surface Force, the Naval Postgraduate School (NPS) and the Navy Health Research Laboratory to develop a basis for measurement and a data collection method that will ensure fatigue policy contents are in place across the Fleet. As part of their fatigue management policy, surveys are to be conducted to measure fatigue as a risk factor by each ship across a 36-month cycle. In its response to the NTSB recommendation, the Navy also stated that the Surface Force remains in close collaboration with NPS in support of leveraging ongoing fatigue management research and tailoring Surface Force policies and associated metrics. The Navy?s proposed policies and plans to address crewmember fatigue, if implemented appropriately, should result in fatigue mitigation and, therefore, the NTSB has reclassi?ed Safety Recommendation M-19-14 as ?Closed?Acceptable Alternate Action? on April 27, 2020. 2.3.2 Navy Oversight The Fitzgerald?s aggressive schedule leaving port to comply with certi?cation requirements and to return to its deployment schedule provided little rest on the day before the accident. The schedule for the Fitzgerald was developed with the assistance and input of the ship?s shoreside command and this type of schedule appeared to be the norm for this area of Navy operations (Forward Deployed Naval Forces). The bridge team and CIC did not function effectively as a team, and the OOD on watch displayed a lack of familiarity in the application of the rules of the road and did not follow the standing orders. Although the CO had a responsibility to ensure that were properly trained and quali?ed, the Navy? training organization and the superiors were required to assess and certify that the operating procedures and the quali?cation system were effective. The Fitzgerald crew?s inability to effectively navigate in traf?c on the night of the accident calls into question the Navy?s assessment and certi?cation process. Based on watchbills reviewed by the NTSB, the ship appeared to be manning and operating equipment in accordance with Condition requirements. However, based on work/rest records reviewed by the investigators, it was apparent that the ship was unable to meet the criteria of ensuring the crew was able get 8 hours of rest each day, given their schedule of operations. It is conceivable 30 NTSB Marine Accident Report that this level of manning degraded the capability, as well as other operational duties among critical personnel. Gaps in manning, combined with a lack of experience and training as well as an aggressive schedule, can lead to crew errors. A ?can do attitude,? which the Navy discussed in its own comprehensive review, can often lead to positive thinking and increased morale. However, without the necessary resources to keep that momentum going, over a period of time, it can lead to erosion of training, morale, and degradation in performance and can desensitize senior leaders to the operational risks associated with this high-tempo work environment. The NTSB concludes that the Navy failed to provide effective oversight of the Fitzgerald in the areas of operations scheduling, crew training, and fatigue mitigation. 2.3.3 Navy Actions since the Accident The US Navy conducted a comprehensive review of this collision and other recent Navy accidents that occurred in the western Paci?c region. Based on that assessment, the Navy established a Readiness Reform and Oversight Committee to address the recommendations obtained from the comprehensive review. The committee developed three levels of actions centered around operational safety, effectiveness, and a culture of excellence. The ?rst two levels of action relating to operational safety and effectiveness have been completed. These actions include performing competency checks and assessments, instituting fatigue management policies and tools, improving training, enhancing and prioritizing manning, implementing BRM workshops, standardizing standing orders, and promulgating AIS guidance. The third level of action relates to a culture of excellence. These actions remain in progress and include continuing enhancements for training, clarifying responsibilities, and improving ?eet technologies. 31 NTSB Marine Accident Report 3 Conclusions 3.1 Findings 1. Weather and the steering and propulsion systems on board the Fitzgerald and the ACX were not factors in the accident. 2. The Fitzgerald bridge team, on the give-way vessel, did not take early and substantial action to avoid collision with the stand-on ACX 3. The communication and cooperation among the Fitzgerald crew on the bridge and in the combat information center were ineffective leading up to the accident. 4. The Fitzgerald?s unexplained small course change to starboard minutes before the collision put the vessel on a collision course with the ACX 5. The Fitzgerald combat information center personnel did not effectively support the bridge team in tracking nearby surface targets. 6. The Fitzgerald provided a detectable radar signature, and the second of?cer on board the ACX should have acquired the destroyer on the automatic radar plotting aid to determine risk of collision. 7. When the second of?cer on the ACX did not receive a response to his signaling attempts and saw no noticeable change in approach from the destroyer, he did not take suf?cient action to avoid the collision. 8. The absence of an automatic identi?cation system signature broadcast ?'om the Fitzgerald likely contributed to the lack of its early detection by the ACX bridge team. 9. The Fitzgerald commanding of?cer did not adequately assess the hazard presented by the vessel?s intended transit, which crossed busy coastal traf?c routes, and should have assigned a more experienced of?cer to augment the bridge team. 10. Several Fitzgerald bridge and combat information center were likely acutely fatigued at the time of the accident, which impacted their situation awareness and ability to identify and respond to the approaching ACX 11. The Navy failed to provide effective oversight of the Fitzgerald in the areas of operations scheduling, crew training, and fatigue mitigation. 3.2 Probable Cause The National Transportation Safety Board determines that the probable cause of the collision between US Navy Destroyer Fitzgerald and container ship ACX was the Fitzgerald?s bridge team?s failure to take early and substantial action to avoid collision as the give-way vessel in a crossing situation. Contributing was ineffective communication and cooperation among the Fitzgerald crew on the bridge and in the combat information center (CIC), and the Fitzgerald commanding of?cer?s (CO) insuf?cient planning for the hazards of the vessel?s intended transit. Also contributing was the Navy?s ineffective oversight of the Fitzgerald in the areas of operations scheduling, crew training, and fatigue mitigation. Also contributing to the accident was the ACX 32 NTSB Marine Accident Report watch of?cer?s lack of early detection of the Navy vessel and insuf?cient actions to avoid collision once in doubt as to the destroyer?s intentions. 33 NTSB Marine Accident Report 4 Recommendations 4.1 New Recommendations As a result of its investigation, the National Transportation Safety Board makes the following new safety recommendations: To the US Navy Review and revise ?eetwide training and quali?cation requirements for of?cers of the deck related to the collision regulations. (M-20-10) Review and revise bridge resource management training in your approved course curriculum to promote a cohesive team environment and improve communication within and between bridge and combat information center teams. (M-20-11) Instruct your vessels to broadcast automatic identi?cation system information while operating in the vicinity of commercial vessel traf?c at all times unless such broadcast compromises tactical operations or strategic interest. (M-20-12) To Sea Quest Ship Management, Inc Provide additional training for your navigation of?cers on collision avoidance regulations, radar, and automatic radar plotting aid. (M-20-13) BY THE NATIONAL TRANSPORTATION SAFETY BOARD ROBERT L. SUMWALT, Ill JENNIFER HOMENDY Chairman Member BRUCE LANDSBERG MICHAEL GRAHAM Vice Chairman Member THOMAS B. CHAPMAN Member Report Date: August 3, 2020 34 NTSB Marine Accident Report Appendix A Investigation The NTSB Of?ce of Marine Safety investigated this accident because it involved a public vessel and a nonpublic vessel, in accordance with Title 49 United States Code, section 1131(b)(1), and the joint Guard regulations of Title 49 Code of Federal Regulations, section The NTSB was the lead federal agency in this accident investigation. The NTSB delegated its authority to the US Coast Guard to gather documents and perform interviews on behalf of the NTSB. Coast Guard Activities Far East led these efforts with support from various other Coast Guard units. The NTSB developed the analysis and probable cause based on the evidence gathered by the Coast Guard and additional documentation provided by the Navy. 35 NTSB Marine Accident Report Appe nd ix Vessel Information Vessels Fitzgerald ACX Owner/Operator US Navy Sea Quest Ship Management, Inc. Homeporthorl: of registry Yokosuka, Japan (home port) Manilla, Philippines Flag United States Philippines Type Destroyer Container ship Year built 1994 2008 IMO number 9360611 Classification Society Nippon Kaji Kyokai (NK) Construction Steel Steel Length 504.5 ft (153.8 m) 730.3 ft (222.6 m) Draft 32.5 ft (9.9 m) 39.4 ft (12.0 m) Beam/width 66.4 ft (20.2 m) 98.8 ft (30.1 m) Displacement Engine power; manufacturer Persons on board 8,261 long tons (8,394 metric tons) Four General Electric LM2500 gas turbine engines, 30,000 hp (22,371 kilowatts each; twin controllable-pitch propellers 315 50,944 long tons (51,762 metric tons) One MAN 8K80MC-C diesel engine, 39,200 hp (29,231 kW); one ?xed-pitch propeller 21 36 NTSB Marine Accident Report Appendix Pertinent Documents Reviewed by Investigators but Protected by the Navy Certain Navy documents relevant to the accident are not releasable to the public. However, the Navy provided NTSB investigators access to these documents to aid the investigation. A brief description of these documents follows. Fitzgerald Voyage Management System (VMS) screenshots. These are protected pictures of the VMS screen displayed on the bridge and in CIC. The pictures show the path of the vessel projected onto an electronic chart of the area. The pictures also show radar targets acquired from the bridge radar?s ARPA unit. Fitzgerald forward- and aft-looking Optical Sight System (088) camera videos. These protected videos captured images of surrounding lights and shapes prior to and after the collision. Fitzgerald Deck Log. This protected log book contains a chronological record of various bridge and other shipboard evolutions. Fitzgerald At Sea Condition Watch Bill. This protected document contains the names, ranks/ratings, and watch positions held by Fitzgerald crewmembers at the time of the accident. Crewmember Fitness Reports and Training Jackets. These protected documents contain the evaluations and completed trainings for the accident of?cer of the deck, junior of?cer of the deck, tactical action of?cer, CIC watch of?cer, CIC watch supervisor, surface warfare coordinator, and the surface warfare supervisor. Personnel Qualification Standard (PQS). Published by the Naval Personnel Development Command, these protected booklets contain a compilation of ?the minimum skills that an individual must demonstrate in order to qualify to stand watches or perform other speci?c duties necessary for the safety, security or proper operation of a ship, aircraft or support system.?28 Combat Systems Departmental Eight O?clock Reports. These reports for June 13, 2017, contained information on combat systems equipment that was either operational but needed repair, or equipment that was totally non-operational. These reports are considered protected documents. 28 Introduction to the PQS system, NAVEDTRA page 4. 37 NTSB Marine Accident Report References Coast Guard. 2015. Navigation Rules: International?Inland. Washington, DC: Department of Homeland Security. Crouch, T. 2013. Navigating the Human Element. An Introduction to Human Factors for Professional Mariners. Navy. 2017. The US Navy Arleigh Burke Class Destroyer. Commander, Naval Surface Forces. 2013. Message to surface forces: ?Circadian Watch bill.? 032157ZMAY13. Commander, Naval Surface Forces. 2017. Message to surface forces: ?Force-Wide Circadian Implementation.? 200042ZSEP17. Endsley, Mica and Daniel Garland. 2000. Situation Awareness: Analysis and Measurement. Mahwah, New Jersey: Lawrence Erlbaum Associates. National Geospatial Intelligence Administration (NGA). 2017. Pub. 158, Sailing Directions (Enroute) Japan Volume]. National Geospatial Intelligence Administration (NGA). 2015. Pub. 120, Sailing Directions (Planning Guide) Pacific Ocean and Southeast Asia. UK Hydrographic Of?ce. 2013. NP42A, Admiralty Sailing Directions, Japan Pilot Volume 2, The south coast of Shikoku from Komo Saki to Kanda Saki; the south coast of Honshu from Hi- no-Misaki to aito Saki; the islands of Napo Shoto. PortVision, Vessel Position Report for ACX DYUG, 06-12-201 7 through 06-] 7-201 7. 38