Various newspapers and electronic media in New Zealand and all over the world reported the disaster of The Rena on Oct 5, 2011. The ship with a speed of about 17 knots (20 mph), ran aground on the Astrolabe reef, about 17 kms away from Tauranga coast in New Zealand.
The ship was carrying 1,368 containers (8 of them hazardous materials), 1,700 tons of heavy fuel oil and 200 tons of marine diesel. Initially the ship listed 11 degrees to the port, with the front stuck on the reef.
The reasons for this incident were not fully known. As feared and expected, a 5 km oil slick threatened the wildlife and the rich fishing waters of the region.
Oil from the ship began washing ashore at Mt. Maunganui beach which attracts a huge number of surfers and beach lovers. Bad weather caused the ship to shift on the reef, and the crew was evacuated. Shifting of the ship caused further damage and leakage of about 130-350 tons of oil.
On Oct 11, 2011, the spill was declared New Zealandâs worst ever environmental disaster by the environment minister Nick Smith.
Strong winds and bad weather caused the ship to list 19 degrees to starboard, and 30 to 70 containers were washed overboard. None of them contained hazardous cargo, and they began washing ashore on Motiti island (see map). A vertical crack in the hull became visible on the port side. The captain of the ship appeared in a court charged with causing unnecessary risk to person and property.
Risks from the Contents
Aerial footage showed a large crack in the hull of the ship, increasing fears that it could break into two and sink. It also showed a container floating in the water surrounded by smoke, suggesting some chemical reaction was taking place. Some of the containers contained chemicals, but overall, the primary risk assessment was low. Recovery crew started to empty the ship of oil before the weather could make salvage efforts much more difficult. Many people started suffering health effects probably from the dispersant sprayed on the oil at sea, and mixing with the heavy oil that volunteers were recovering from beaches. In other areas of the east coast, residents were advised to disconnect rainwater pipes as they could be contaminated.
The main objectives behind this article are to: (a) identify the potential causes behind the Rena disaster using external sources of information (electronic and print media), personal interviews, and individualâs comments; (b) list the important events during the recovery operations; (c) compare the different alternative actions and approaches using decision tree analysis; d) study the effects of this disaster on supply chain management and customers businesses; and (e) make recommendations that might be useful in dealing with such a challenging disaster recovery operations in future.
So What Could Have Possibly Gone Wrong?
It is not yet clear how the ship really got grounded in the first place. The disaster could have occurred due to a range of reasons such as poor visibility due to bad weather conditions, failure of equipment or navigation instruments, erroneous information from the control room or wrong weather forecasting, and of course the negligence of crew members possibly due to lack of rest or the hangovers from long journey or drinks. According to the environment minister, the container ship may have been taking a shortcut when the accident occurred. âIt appears from the charts that they were in a rush to get to the port, went full bore, cut the corner, and hit the reef,â said Smith. In order to collate all the possible reasons, an Ishikawa (or fishbone) diagram has been used based on the media reports (Figure 1).
Table 1: Possible alternative responses, risks and payoffs
Alternative 1: Empty oil first from the Rena using efficient pumping system and then remove all the containers one by one using ship crane (s). Risks: The ship could split into two due to the weights of containers, load fatigue, wind pressure, and onslaught of tidal waves. It is a very slow and time taking process and can take almost 40 days given the constraints with pumping system and accessibility to the oil tanks. Containers being precariously balanced can fall over as they are constantly bombarded by tidal waves. The ship and cargo can catch fire in case of oil spill during the emptying process and can lead to huge losses (millions of $ depending on the contents of the containers). Payoff: This action can save the aquatic lives which can be destroyed if the oil remains in the ship and the ship splits apart leading to huge amount of spillage. |
Alternative 2: Remove containers first using ship crane(s), and then empty oil from the ship using pumping systems. It is advised that removing containers should be from the bow, and the stern sides as these loads can create more bending stresses and shear forces on the line of split. Removing these loads can reduce the splitting process very significantly (follow the theory of cracks initiation and cracks propagation in fracture mechanics). Risk: Removal of containers can be hampered by the unpredictable wind and weather conditions Payoff : Reduced possibility of the ship splitting into two parts or more due to reduced lateral and longitudinal pressures on the ship hull which may give more time and perhaps easy access to the ship for emptying the oil using more pumps. |
Alternative 3: Remove the containers using the ship crane, and empty the oil from the ship using pumping systems simultaneously. Risks: Increased possibility of fire hazards due to oil spills while containers being unloaded simultaneously plus risks of two processes interfering into each other due to space constraints. Payoff: It is a very quick process, and it can reduce the opportunity costs. |
Alternative 4: Do nothing and let them be at the mercy of natural forces (e.g. wind pressure, load fatigue, or tidal waves). Risks: This inaction will lead to huge losses, immense damage to natural environment, aquatic animals, clean and green image of New Zealand, tourism and business, fishing industry, etc. Payoff: There are no costs involved and no resources required, but there will be long-term consequences of this inaction. |
Probability of good weather = 0.5 (due to the summer season); bad weather = 0.3 (due to low pressure developed from time to time); ugly weather = 0.2 (due to storm, cyclone, hurricane, typhoon, etc). |
Decision Tree Diagram
The model takes into account three possible weather conditions (good, bad, and ugly) with their assumed probabilities (0.5, 0.3, and 0.2) to draw a decision tree diagram (Figure 2). There can be other rare possibilities during the recovery operations such as a volcanic eruption, an earthquake, a tsunami, or a combination of these; however, the probability could be something like 0.0000125 which has been kept out of consideration for the sake of simplicity.
Let Ai = set of possible alternatives that could be initiated by the recovery team to recover the ship and its cargo (Ai = A1, A2, A3, A4)
A1 = (OE1 + CR2 ) means oil emptying to be done 1st and the containersâ removal 2nd
A2 = (CR1 + OE2 ) means containersâ removal to be done 1st and the oil emptying 2nd
A3 = (OE3 + CR3) means oil emptying and containersâ removal to be done simultaneously
A4 = just do nothing and let everything be left to the act of nature.
Now each action or inaction will have its own associated cost and benefit (payoff or expected monetary value) as stated below.
Let EMVi = Expected monetary value or the payoff at ith node (where i = 1, 2, 3, 4)
Xij = net benefit at ith node (i = 1, 2, 3, 4) with jth conditions (j = g, b, u or good, bad, ugly)
Table 2: Estimated values of various payoffs
Node | Payoff in good weather | Payoff in bad weather | Payoff in ugly weather |
1 | X1g = -$40 million* | X1b = -$50 million | X1u = -$80 million |
2 | X2g = – $50 million | X2b = – $60 million | X2u = – $90 million |
3 | X3g = – $70 million | X3b = – $80 million | X3u = – $100 million |
*Net Payoff = {different benefitsâ different costs} = {(benefits to environment, aquatic animals, brand image of New Zealand, tourism, health, and safety, etc) â (cost of pumping system, cranes for handling containers, barge, helicopters for air surveillance, motorboats for inspection around the ship + labor cost + opportunity cost + cost of lost sales)}. The costs of A2 will be higher than the costs involved in A1 but less than A3 (i.e., CA1<CA2<CA3). Therefore, the net payoff will be in the reverse orders as shown in Table 2. Various payoffs in Table 2 are estimated values due to the lack of real data available. The EMV for each alternative is computed as follows:
EMV1 = 0.5X1g + 0.3X1b + 0.2X1u = 0.5(-40) + 0.3(-50) + 0.2(-80) = – $51 m
EMV2 = 0.5X2g + 0.3X2b + 0.2X2u = 0.5(-50) + 0.3(-60) + 0.2(-90) = – $61 m
EMV3 = 0.5X3g + 0.3X3b + 0.2X3u = 0.5(-70) + 0.3(-80) + 0.2(-100) = – $79 m
Based on the EMV values, it is obvious that alternative 1 suffers the least loss (i.e. $51 million loss as compared to the other two alternatives); therefore, the recovery team should follow alternative 1.
Figure 2 Decision Tree Diagram
Disaster and Insurance Issues
The Rena disaster has brought the issues of marine insurance in focus, as reported in the Risk Management Newsletter. International convention limits the shipâs liability to approximately NZ$ 14M which is based on the shipâs gross registered tonnage. It also comments on whether Resource Management Act liabilities can be imposed either in lieu of, or in addition to, the international convention (but this is a legal matter and is not insurance related). Many businesses have indicated their income has been or will be affected. Business interruption policies (BIP) would not include the losses arising from this incident because it is likely the businesses have no property on the Rena, and most of the BIPs do not include maritime perils in the coverage.
According to a Vero Marineâs circular to customers, âIf you have goods on board and are obliged to maintain supply to overseas customers and you are insured with Vero Marine, there may be some assistance in the form of reimbursement of some additional freight costs you incur should your cargo be lost or damaged in this casualty.â Other insurers have taken up similar steps. Virtually all marine cargo policies exclude delay and abandonment.
Conclusions and Recommendations
Enhanced visibility: Initial investigation suggests that human errors have caused the Rena disaster. Although Astrolabe reef is well marked on the maritime map, there is scope for making it more visible especially during the night or poor visibility conditions to prevent any disasters in the future. Modern technology like satellite, GPS, and RFID may be used to track such sea based barriers or obstructions to prevent such incidents.
Disruption in supply chain: The approach suggested by Ken Schroeder in Disaster Recovery Journal (âWhat If That Doesnât Work?â) could be used by suppliers while sending items to overseas customers by shipping to avoid disruption in supply chain. Suppliers sending their items to New Zealand customers by The Rena could have done better by asking this question to themselves: What if the shipping option doesnât work? Well, they can split their orders and supply the items on two ships rather than one. But, what if that doesnât work due to a strong storm which can disrupt all the shipping modes of transport? In that case they could have used other sources of supply such as using their strategic partners overseas who are not affected by this storm. That sounds a very plausible alternative. But what if that too doesnât work? Then the supply can be made by air cargo perhaps in small volumes to maintain supply in critical areas for the time during which the recovery operations on ship is in progress.
Although it is very difficult to benchmark the speed of recovery process carried out by the team, it appeared slow at times to the public. The forces of nature have provided the team with enough time and opportunity to apply more equipment and resources to remove oil and containers from the ship. Removal process could have been carried out even during the night using lighting arrangement and the night vision technology because it was summer season. Things could have been much worse and slower in the winter season.
Post-Incident Review: It is essential to carry out a post-incident review (PIR) to evaluate responses of the recovery team to identify gaps or weaknesses and determine ways and means to strengthen them. Marc Morgan (Disaster Recovery Journal, âThe Post-Incident Review Process: Can You Correct The Weaknessâ) suggests that âPIR should be carried out by a team of individuals that are not part of the organization or were not involved with the response to the incident.â However, the responders and managers should be given opportunity to provide their input in the process. PIR should evaluate the recovery operations in terms of performance measures such as speed, cost, effectiveness, and flexibility. It should also address the issues of planning, prevention, mitigation, and implementation of disaster response. The PIR process clearly provides an opportunity to learn from disasters and crises.
There seems to be huge ambiguity in terms of who should pay for the cost of recovery operations. Insurance clause in the international maritime regulations should be modified to express clearly the role of ship-owners in terms of the expenses of disaster recovery operations.
Dr. Ram Roy is working as a senior lecturer in management in the School of Business at Eastern Institute of Technology, Napier, New Zealand. Dr. Roy has published many papers in various journals in the domains of operations management, supply chain management, logistics, supply chain mapping, and disaster management. Dr. Roy may be contacted at rroy@eit.ac.nz.
Dr. Roy acknowledges the support of Maritime, New Zealand, Wildlife volunteers, Tauranga Community, and Eastern Institute of Technology, Hawkeâs Bay, New Zealand for their help during the research paper. The author also thanks Vineeta Roy for her help in proofreading.
