Gå til hovedinnhold

Audit of major accident risk management and barrier handling in Odfjell Drilling

During the period 8-28 April 2010, the PSA conducted an audit of Odfjell Drilling AS’ (Odfjell’s) major accident risk management and barrier handling. Two regulatory non-conformities and six improvement items were identified during the audit.


This audit activity was linked to two of the PSA’s main priorities for 2010:

Management and major accident risk”, with the objective that management at all levels of the industry will work to reduce major accident risk, and ensure that this work is pursued in an integrated manner.

Technical and operational barriers”, with the objective that technical and operational barriers will be maintained in a unified and consistent manner to keep the risk of major accidents as low as possible.

The audit activity was opened with an all-day start-up meeting with the management on land on 8 April 2010. During the period 26-28 April 2010, we performed a verification on board the Songa Delta facility.

During the opening meeting on land, the PSA received a review of the company management’s understanding of strategies and principles forming the basis for design, use and maintenance of barriers. The roles of management and measures to ensure risk understanding in connection with major accidents were presented to the audit team. Odfjell is in the process of implementing a “bow-tie” methodology (loop diagrams) which will illustrate hazard situations and preventive barriers on one side, and consequence-reducing barriers following incidents on the other. This methodology was also reviewed during the meeting.

On board the Songa Delta, the audit team verified Odfjell’s management of major accident risk by reviewing two major accident scenarios in connection with the facility’s defined hazard and accident situations (DFUs). The verification was conducted further through interviews and reviews of the emergency preparedness plan and other documentation. The audit team paid special attention to risk factors in connection with well activities, fire and explosion hazards and the handling of emergency preparedness following serious incidents.

An inspection was also performed on the facility. A major accident scenario was practiced during a drill, attended by the entire emergency preparedness organisation. A meeting was also held with the safety delegate service.

The audit activity was well facilitated by Odfjell.

Background
The audit activity was linked to two of the PSA’s main priorities for 2010:

• “Management and major accident risk”, with the objective that management at all levels of the industry will work to reduce major accident risk, and ensure that this work is pursued in an integrated manner.
• “Technical and operational barriers”, with the objective that technical and operational barriers will be maintained in a unified and consistent manner to keep the risk of major accidents as low as possible.

The Songa Delta is a semi-submersible drilling facility. Odfjell is responsible for the operation of the facility, which is owned by Songa Offshore. Odfjell received an Acknowledgement of Compliance (AoC) for the facility (Deepsea Delta) in 2004, and a new AoC for the Songa Delta in 2009, following the change of name and owner.

Goal
The goal of the audit activity was to evaluate the company’s understanding, knowledge and competence in connection with major accident risk and barrier thinking, both in management and with employees. Further, we wanted to evaluate strategies and principles for management, design, use and maintenance of barriers – especially as regards major accident scenarios.

Result
Odfjell’s main management has defined, and seems to focus on, which DFUs can primarily trigger major accidents, in addition to the appurtenant operational and organisational barrier systems.
The introduction of the “bow-tie” methodology is intended to safeguard and visualise the connection between hazard situations and barriers. The PSA views the methodology as a useful tool with good potential for practical application through further development in the company.

The methodology has not yet been prepared for all major accident scenarios. Following the verification on board, it is our opinion that the implementation and use of the methodology on the Songa Delta has not progressed as far as it seems to have progressed on land. Among the personnel on board the Songa Delta, we found little knowledge of the bow-tie methodology and how this methodology was planned to be used. We also noted uncertainty in relation to the use of the bow-tie diagrams among management personnel on land. It was not evident to the audit team that there was an increased focus on training and exercises related to major accident risk. We also discovered varied knowledge and understanding of which DFUs have major accident potential.

The operator for the ongoing operation on Songa Delta, Det norske, has taken the initiative for greater focus on major accident risk. We were informed of this during the land meeting. This initiative was not particularly visible on board. The personnel were also not very familiar with the initiative.

The audit team received a positive impression of the working environment on board. The safety delegate service emphasised that there was good cooperation between management and employees in the company. We found it to be very tidy on board and the outdoor areas were cleared.

We identified two non-conformities and six improvement items during the audit.

Observations
The PSA’s observations can generally be divided into two categories:
• Non-conformity: Related to observations where we have identified breaches of the regulatory requirements.
• Improvement item: Related to observations where we find deficiencies, but do not have sufficient information to be able to ascertain breaches of the regulatory requirements.

Management of major accident risk

Improvement item: Major accidents – follow-up of barriers and performance requirements
A system was not established to safeguard the overall requirements for follow-up of barriers and performance requirements across the organisation.

Basis:

Other conditions

Non-conformity: Personnel transport basket as primary rescue appliance
The personnel transport basket was set up as the primary rescue appliance during evacuation.

Basis:
The following observations were made during the verification related to the personnel basket:

Non-conformity: Training – fire team
Deficient plan for training the fire team.

Basis:
During the interviews, we discovered that the fire team did not have regular training for every rotation, apart form adjusting personal equipment upon arrival. Upon request, a plan could not be produced for such training of the fire team.

Improvement item: emergency preparedness plan – systematics and performance requirements
The emergency preparedness plan did not appear to be user friendly, and was not updated.

Basis:

Improvement item: Fire fighting in the machine room
A permanent fire fighting system had not been installed in the machine room to fight larger fires in the room in a quick and efficient manner.

Basis:

Improvement item: Full evacuation with lifeboats in emergency situations
Unclear procedures for evacuation with lifeboats in a major accident situation

Basis:
During interviews, we discovered uncertainty and unclear criteria in connection with evacuation to the sea. There was particular uncertainty in connection with when lifeboats are to be lowered in the event of ships on collision course and in hazardous situations where time is critical and leaving the facility is considered. The evacuation of the emergency team takes place in the same lifeboat as for the ordinary personnel, forcing this lifeboat to wait for the emergency team.

Improvement item: Training and drills
Deficiencies in training and drills.

Basis:

Improvement item: Battery-operated emergency lighting
Deficient battery-operated emergency lighting in rooms for CO2 containers.

Basis:
During the inspection, we discovered that battery-operated emergency lighting was not present in the storage room for CO2 containers. During an emergency situation, there could be a need for access to the room for initiating the system directly on the container bank.

Other comments

An emergency preparedness drill with major accident potential (fire in machine room) was held involving the entire emergency preparedness organisation. Lessons learned from the drill were summarised later. The drill uncovered clear weaknesses and risks in connection with the use of CO2 as an extinguishant, and the importance of quality-assuring communication and information.

The audit team had prepared two table-top drills based on major accident risk; relating to Songa Delta’s DFUs “Fire in machine room” and “Loss of well control”. Two groups, each with six crew members, participated in a dialogue with the audit team concerning the hazardous situation, barrier elements, barrier functions and incident scaling on one side, and handling, fighting, evacuation and normalisation on the “consequence” side. There was also focus on decisions, decision authority and responsibilities in the discussion.

The two groups participating in the table-top drills, exhibited positive involvement and openness. Through these drills, we gained good insight into the crew’s awareness of hazardous situations and barriers that can prevent major accidents on board the Songa Delta.