A disorientated pilot caused the collision between the "City of Rotterdam" and the "Primula Seaways" in the River Humber, an MAIB report stated. An error of navigational judgement by the pilot of the "City of Rotterdam" was the reason for the accident in December 2015. The report report detailing what caused the two vessels to crash was released on Feb 8, 2017.On 3 December 2015, the Panama registered pure car carrier City of Rotterdam collided with the Danish registered ro-ro ferry Primula Seaways on the River Humber, UK. Both vessels were damaged but made their way to Immingham without assistance. There was no pollution and there were no serious injuries.The MAIB investigation identi ed that the outbound City of Rotterdam had been set to the northern side of the navigable channel and into the path of the inbound ferry, but this had not been corrected because the pilot on board had become disoriented after looking through an off-axis window on the semi-circular shaped bridge. The car carrier was of an unconventional design and his disorientation was due to ‘relative motion illusion’, which caused the pilot to think that the vessel was travelling in the direction in which he was looking. Consequently, the pilot’s actions, which were designed to manoeuvre the car carrier towards the south side of the channel, were ineffective.That the pilot’s error was allowed to escalate the developing close quarters situation to the point of collision was due to: intervention by City of Rotterdam’s master was too late, and the challenges to the pilot’s actions by Primula Seaways’ bridge team and the Humber Vessel Traf c Service being insuf ciently robust. Although Primula Seaways started to reduce speed about 2 minutes before the collision, a more substantial reduction in speed was warranted.Following the accident, and an early MAIB recommendation, action has been taken by Fairmont Shipping (Canada) Limited, City of Rotterdam’s managers, to reduce the likelihood of relative motion illusion and to improve the bridge resource management of its deck of cers. Action has also been taken by Associated British Ports, the harbour authority
for the River Humber, to con rm the competency of the pilot and the suitability of Primula Seaways’ master to hold a pilotage exemption certi cate.Bureau Veritas, City of Rotterdam’s classi cation society, has been recommended to propose measures to the International Association of Classi cation Societies that are aimed at raising the awareness of relative motion illusion and promoting the need for naval architects and shipbuilders to adhere to internationally accepted ergonomic principles for bridge design. THE COLLISIONThe collision between Primula Seaways and City of Rotterdam stemmed from the latter being set to the northern side of the Bull Channel by the wind and the tidal stream, followed by the distortion of its pilot’s spatial awareness due to a ‘relative motion illusion’. The pilot was aware that the car carrier was to the north of its intended track but he perceived the action he was taking to head to the southern side of the buoyed channel was as substantial as the navigational constraints allowed. The pilot was under the impression that the vessel was heading to the south, whereas its heading was not altered signi cantly beyond the axis of the channel (Figure 12) until collision was imminent. Consequently, City of Rotterdam remained on the northern side of the buoyed channel and in the path of the inbound Primula Seaways.The bridge teams on board both City of Rotterdam and Primula Seaways and
the VTS operators were appropriately quali ed and experienced. However, the interventions made by VTS were not suf ciently robust to make the pilot appreciate that more aggressive action was required to avoid the developing collision situation. Moreover, the absence of any challenge or intervention by City of Rotterdam’s bridge team until collision was imminent indicates an over-reliance on the pilot and a breakdown in the bridge resource management on that vessel. In addition, although Primula Seaways’ bridge team had identi ed the risk of collision and had taken action to clarify the pilot’s intentions, substantial action to avoid the collision was taken too late to be effective. CONCLUSIONSSAFETY ISSUES DIRECTLY CONTRIBUTING TO THE ACCIDENT THAT HAVE BEEN ADDRESSED OR RESULTED IN RECOMMENDATIONSThe collision stemmed from City of Rotterdam being set to the northern side of the Bull Channel by the wind and the tidal stream followed by the distortion of its pilot’s spatial awareness due to a ‘relative motion illusion’. [2.2]City of Rotterdam’s pilot’s relative motion illusion deceived him into thinking that his view from the window above the starboard VHF radio, which was 33° off the vessel’s centreline axis, was the vessel’s direction of travel. [2.3]As it was dark, the inward slope of the window removed all objects in the pilot's periphery, and there were no visual clues such as a forward structure or bow tip, the illusion would have been compelling. [2.3]The pilot’s ability to reconcile the headings he had ordered with his perceived direction of travel was probably hindered by further psychological effects of the relative motion illusion, such as the cognitive costs of transferring between different frames of reference. [2.3]Soon after City of Rotterdam entered the main navigation channel, the master and the third of cer left the responsibility for the vessel’s safe passage predominantly to the pilot. [2.4]City of Rotterdam’s master and third of cer did not challenge the pilot’s actions despite concern at the vessel’s position being expressed by Primula Seaways and the VTS. The master’s intervention, 14 seconds before the collision, was far too late to be effective. [2.4]City of Rotterdam’s bridge team’s over reliance on the pilot, and its lack of effective monitoring of the vessel’s progress, were evidence of ineffective bridge resource management. [2.4]Although Primula Seaways’ engines were reduced to ‘half ahead’ 2 minutes before the collision, a more substantial reduction of speed was warranted in view of the doubt concerning City of Rotterdam’s movement. [2.5]The VTS intervention at 2038 could have been more effective in alerting the bridge teams on board both vessels to its concerns had it been pre xed with a ‘warning’ message marker and it had not referred to the pilot by name. [2.6]The location of the VHF radios by the off-axis windows on board City of Rotterdam increased the potential for relative motion illusion. [2.7]The potential for relative motion illusion was unforeseen and therefore not taken into account during the design of City of St Petersburg and City of Rotterdam. [2.7]Stricter adherence to the ergonomic principles of bridge design detailed in SOLAS V/15 would reduce the likelihood of human error. Therefore, the need for an IACS UI on the interpretation of the ergonomic principles of bridge design warrants reconsideration. [2.8]ACTION TAKEN MAIB ACTIONSOn 11 February 2016, the MAIB recommended Fairmont Shipping (Canada) Ltd to:2016/104Take action to reduce the likelihood of distorted spatial awareness on the bridges of City of Rotterdam and City of St Petersburg, taking into account, inter alia:● The importance of emphasising to crew and embarked pilots the risk of spatial distortion occurring on these bridges.● The increased risk of distorted spatial awareness when standing away from the centreline or a navigation station, including when using the xed VHF radios.● The need to monitor pilots’ actions at all times and to challenge when appropriate.4.2 ACTIONS TAKEN BY OTHER ORGANISATIONS 4.2.1 Fairmont Shipping (Canada) LtdFairmont Shipping (Canada) Ltd in response to MAIB recommendation 2016/04 has:● Installed a bow tip marker on the centreline immediately ahead of the centre bridge window to provide a reference point from any position on the bridge.● Increased the length of the VHF handset wires to enable the radios to be used from the forward centreline conning position.● Posted notices by the forward VHF radios, on the bridge deck head and in the chart room warning of relative motion illusion. The notices state:● CAUTION● ERRORS IN JUDGEMENT FROM ‘RELATIVE MOTION ILLUSION’ MAY OCCUR IF OBJECTS ARE VIEWED THROUGH SIDE WINDOWS ON THE CURVED SECTION OF THIS WHEELHOUSE.● ‘RELATIVE MOTION ILLUSION’ IS A PHENOMENON IN WHICH OBJECTS APPEAR TO MOVE AS THOUGH THE SHIP WAS HEADING IN THE DIRECTION OF VIEW THROUGH THE WINDOW. IT IS MORE LIKELY TO OCCUR DURING PERIODS OF DARKNESS ● Incorporated into the safety management system and the pilot information card references to spatial awareness and relative motion illusion. The ship manager also conducted an investigation into the collision. The investigation report (Annex C) identi ed measures to prevent a similar accident in the future. These included additional internal audits to monitor bridge teams during pilotage and coastal navigation, and refresher bridge resource management training for all masters and deck of cers. DFDS A/SDFDS A/S has:● Re-af rmed to its masters and navigating of cers the importance of good bridge resource management and continued to enrol them on its MCA approved ‘Maritime Resource Management Course’. Associated British PortsABP has:● Reassessed the pilot’s competency and re-authorised him for pilotage duties.● Reassessed and re-authorised Primula Seaways’ master’s PEC.● Reviewed its risk assessment for traf c in the vicinity of the Clee Ness VTS reporting point.● Reinforced the importance of using message markers when conducting VTS communications.SECTION 5 - RECOMMENDATIONS Bureau Veritas is recommended to:2017/104Propose to the International Association of Classi cation Societies that Recommendation 95 “Recommendation for the Application of SOLAS Regulation V/15 Bridge Design, Equipment Arrangement and Procedures (BDEAP)” is revised to:● Improve the de nition of conning position(s), taking into account the equipment that is required to be at, viewable from, and convenient to the position.● Raise the awareness of the dangers of navigating from off-axis windows and the effect of relative motion illusion.Propose to the International Association of Classi cation Societies that the status of Recommendation 95 is raised to a Uni ed Interpretation. Full report: https://assets.publishing.service.gov.uk/media/58984f60ed915d06e1000025/MAIBInvReport3_2017.pdf