Chapter 17

Vessel Safety and Accident Analysis

Wayne K. Talley*

1. Introduction

Vessel safety consists of the safety of a vessel and the safety in the operation of a vessel. The safety of a vessel is concerned with whether a vessel adheres to construction, design or technical standards and therefore is seaworthy. Since safety in the operation of a vessel is difficult to observe, input and outcome proxies for vessel operation safety are often used. Input proxies represent actions by shipping lines, ports and government to decrease the probability of unsafe vessel operation; outcome proxies represent unsafe outcomes of vessel operation. Vessel maintenance and operator training expenditures are examples of input proxies; ship accidents and fatalities and injuries from these accidents are examples of outcome proxies. Since little is known of the extent to which safety inputs translate into vessel operation safety, outcome rather than input proxies are often used as proxies for vessel operation safety.

A vessel accident is an unintended happening and may or may not result in damage to the vessel and injuries to individuals on board. The probability of a vessel sustaining damage in an accident is the product of two probabilities: (1) the probability of involvement in an accident (event probability), and (2) the probability of vessel damage given that an accident has occurred (damage conditional probability). Similarly, the probability of an injury in a vessel accident is the product of the event probability and the probability of an injury given that an accident has occurred (injury conditional probability). The severity of an accident may vary from: no vessel damage to the loss of the vessel, no injuries to fatalities, and no cargo damage to loss of cargo. A vessel may remain seaworthy following an accident or may be non-seaworthy.

The remainder of the chapter is structured as follows: Section 2 presents a discussion of vessel safety concerns. Causes of vessel accidents and related injuries are presented in Section 3 and vessel accident costs are discussed in Section 4. Determinants of vessel accident property damage costs, injuries and injury severity, and seaworthiness are discussed in Sections 5, 6 and 7, respectively. Statistics for world accident vessel losses are presented in Section 8. A summary of the above discussion is found in Section 9.

2. Vessel Safety Concerns

Prior to World War I the need for uniform international vessel safety rules and their enforcement became evident. Since vessels involved in international commerce must fly the flags of their countries of registry and obey laws of this registry, the flag states (or countries) were asked to adopt and enforce internationally agreed upon rules. The multilateral enforcement of multilateral vessel safety rules worked well until “flags of convenience” (FOCs) or open registries were adopted, i.e. the registration of vessels in countries other than those of its citizen owners (Goss, 1994). Prior to the 1950s, FOCs were insignificant in number. Today, over one-half of the world’s shipping fleet is registered with open registers – raising concern for enforcement of international safety rules, given that some FOCs confine their interests to collecting registration dues, having no interest in adopting or enforcing rules, and can not be compelled to do so as sovereign powers. The major open-registry flag countries include Panama, Liberia, Cyprus and the Bahamas.

What determines whether a vessel will fly a foreign flag (FOC) or a national flag (the flag of the country of its citizen owner)? An investigation into the determinants of vessel flag (for vessels that trade internationally) is found in a study by Hoffmann, Sanchez and Talley (2005). The likelihood that the operator of a vessel will choose a foreign flag decreases with vessel age and if the vessel was built in the vessel operator’s country, but increases with the size of the vessel, if the vessel is a container vessel, if the vessel operator’s country is a developed country, and if the vessel is classed by a member of the International Association of Classification Societies (IACS).

The ineffective FOC enforcement of international vessel safety rules has been addressed by some countries establishing port state control (PSC) systems, i.e. systems that unilaterally enforce such rules (Payoyo, 1994). In 1982 12 European countries signed the Paris PSC Memorandum of Understanding, arranging to inspect safety and other certificates carried by vessels of all flags (including each other’s) visiting their ports, and to insist, by detention if necessary, on deficiencies being rectified. The International Maritime Organization (IMO) defines PSC as “the inspection of foreign ships in national ports to verify that the condition of the ship and its equipment comply with the requirements of international regulations and that the ship is manned and operated in compliance with these rules” (IMO, 2009). In 1995 member countries inspected 8,834 vessels, of which almost half had deficiencies; vessels were detained in port when deficiencies were regarded as so serious that the vessel or those on board were in danger, or where the marine environment could be threatened (Porter, 1996).

During the years 2002 to 2006, member countries of the Indian MoU for PSC undertook 26,515 PSC vessel inspections, ¹ 7.7% of the vessels involved in these inspections were detained. Determinants of the probability that a vessel would be detained based upon these data are found in a study by Cariou, Mejia and Wolff (2008). A strong positive relationship exists between this probability and the vessel’s age at the time of the inspection. Also, the inspecting authorities in Iran, India and Australia have a higher probability of vessel detentions than inspecting authorities of other member countries. Further, general cargo/multi-purpose and chemical cargo vessels are more likely to be detained.

In addition to FOCs, doubts also exist about the vessel safety enforcement performance of classification societies. Classification societies which are generally privately owned inspect vessels to ensure that they are seaworthy, meet national-flag requirements and conform to international safety standards. Classification societies also produce vessel specification rules and supervise the construction and design of vessels to insure that these rules are followed. There are more than 50 classification societies worldwide with some having been in existence for more than 200 years. The five largest classification societies (based upon the number of vessels that they classify) are the Bureau Veritas (France), Det Norske Veritas (Norway), American Bureau of Shipping, Lloyd’s Register of Shipping and Nippon Kaiji Kyokei.

Both existing and new vessels are classified. Although vessels are not required to be classified, vessel insurers must be confident that vessels are seaworthy and thereby will insure only vessels that have been classified. Further, charterers of vessels require that vessels be classed and owners of non-classified vessels cannot obtain the necessary trading certificates required by ports of call (Talley, 2005).

Since classification societies have no legal authority, they compete for vessel-owner clients. Consequently, an insoluble conflict of interest arises between themselves and vessel owners, since the latter hire classification societies to class vessels. In a competitive environment for vessel-owner clients and when vessel owners themselves are facing stiff competition, societies are under pressure to reduce their safety demands, possibly classing non-seaworthy vessels. By the end of the 1970s, the UK P&I Club, responding to the concern that classification societies could no longer provide an accurate evaluation of vessel quality, established its own vessel appraisal system. Protection and indemnity (P&I) clubs are vessel owners’ organisations that provide liability insurance for the same vessel owners. The criticisms of classification societies include: (1) extreme variations recorded in the quality of services provided; (2) difficulty in obtaining vessel inspection reports given the contractual links between societies and vessel owner clients; (3) unwarranted extensions of the classification of older vessels; and (4) safety rules that do not consider the operational aspects of safety on board (e.g. crew quality and operating standards (Boisson, 1994). The major classification societies responded to their critics by establishing the International Association of Classification Societies (IACS), having the objectives of promoting the highest standards in vessel safety and preventing marine pollution. IACS members are bound to satisfy Quality System Certification Scheme (QSCS) standards.

Shrinking crew sizes are also a safety concern, since fewer crew members may be available for watch duties and on-board maintenance chores. Opponents of smaller crew sizes (e.g. labour organisations) argue that safety has deteriorated with smaller crews, (e.g. from increased fatigue from longer working hours, poor vessel maintenance practices and less time for on-the-job training). They further note that crew fatigue was cited as a major contributing factor to the Exxon Valdez oil spill in Alaska. Proponents (e.g. vessel operators) argue that smaller crews are more safety conscious and are better trained to operate automated systems. One study (National Research Council, 1990) that investigated the issue concludes that available information shows no link between crew size and commercial vessel safety. However, if crew sizes continue to shrink, they may fall below safety threshold levels; if so, vessel operation safety will deteriorate.

The construction and maintenance of vessels are also vessel safety concerns. New vessels are often constructed with lightweight high-tensile steel, which is thinner than plain steel and thus more likely to crack and suffer dangerous stresses. In a competitive environment, vessel operators are under pressure to decrease the time that their vessels are in port. Consequently, the maintenance of vessels in port is expected to decline as vessel time in port declines.

Nearly 80% of vessel accidents are caused by human error – a human action or omission identifiable as the immediate cause of the event from which the liability arises, including blame worthy behavior from simple mistakes in arithmetic, judgment, and deliberate risk taking (Goss, 1994). In the past, the focus of vessel safety regulation has been the vessel rather than human actions aboard the vessel. However, this focus has shifted: The International Safety Management Code for vessels became mandatory in 1998, requiring shipping lines to document their vessel management procedures for detecting and eliminating unsafe human behavior. “This code is at the heart of the industry’s plan to switch toward regulating human factors instead of physical ones” (Abrams, 1996, p. 8B). The code was motivated by the fact that vessel accident insurance claims are often attributed to human error and it is less expensive to change human behavior than it is to redesign vessels for safety.

Older vessels are also a safety concern, especially older dry-bulk vessels. The overloading and the use of 30-tonne buckets, pneumatic hammers, and bulldozers to unload dry-bulk cargoes weakened the structures of older dry-bulk vessels. Inspections of older coal vessels, for example, reveal that corrosion of side shells is common; moisture in coal vaporises and re-condenses against side shells. The availability of experienced crew to man and repairmen to maintain and repair older vessels is also a safety concern.

Safety concerns for ferry vessels include insufficient fire protection and their instability. Roll-on roll-off ferries have giant holes that allow for the loading (roll-on) and the unloading (roll-off) of automobiles and other cargoes and preclude vertical watertight bulkheads that are standard features on most vessels. If water gets in and causes a pronounced list, the vessel will capsize and sink. If loading doors are breached, ferry vessels can sink without warning – approximately 60% of ro-ro ferries involved in accidents sink within 10 minutes (Talley, 2002). In September 1994 lock design flaws and a slow response by the crew were instrumental in the deaths of 852 people from the sinking of the Estonia ferry during a Baltic Sea storm. It was Europe’s worst maritime passenger disaster since World War II.

3. Vessel Accident and Injury Causes

The US Coast Guard classifies the causes of vessel accidents into human, environmental and vessel causes. Human causes include stress, fatigue, carelessness, operator error, calculated risk, improper loading, lack of training, error in judgment, lack of knowledge, physical impairment, improper cargo stowage, inadequate supervision, improper mooring/towing, design criteria exceeded, psychological impairment, intoxication, failed to yield right of way, improper safety precautions, failed to keep proper lookout, and failed to proceed at safe speed. Environmental causes include debris, shoaling, lightning, adverse weather, submerged object, channel not maintained, unmarked channel hazard, hazardous bridge/dock/pier, and adverse current/sea conditions. Vessel causes include corrosion, cargo shift, dragging anchor, stress fracture, brittle fracture, fouled propeller, improper welding, steering failure, propulsion failure, static electricity, temperature stress, inadequate controls/displays/lighting, inadequate horsepower, inadequate lubrication, and auxiliary power failure.

A vessel accident seldom has a single unambiguous cause. Causes are often a sequence of causes (or events). For example, adverse weather, the initial (environmental) cause of an accident, may in turn contribute to operator error, a secondary (human) cause of accident. However, if a single accident cause is to be selected, the suggestions include either the initial cause or the last cause (beyond the initial cause) in the sequence of causes at which the accident could have been prevented (Oster and Zorn, 1989).

The market environment in which shipping lines operate can also affect vessel safety. The profit-safety argument (Loeb, Talley and Zlatoper, 1994) states that there is a positive relationship between profitability and safety in the shipping line industry. That is to say, adverse financial conditions in the industry are expected to lead to an increase in vessel accidents. The profit–safety argument opposes market forces for promoting vessel safety and therefore favors regulation and more stringent public-sector enforcement for such promotion.

An underlying argument of the profit–safety argument is that a positive relationship exists between shipping line profits and safety expenditures (e.g. vessel maintenance) and these expenditures, in turn, have a positive influence on vessel safety – i.e. a decrease in profits will lead to a decrease in safety expenditures which, in turn, will reduce vessel safety, thereby resulting in an increase in vessel accidents. An investigation of this linkage for the airline industry is found in Talley (1993). A highly statistically significant positive relationship was found between airline operating margins, i.e. one minus operating costs/operating revenue, and relative maintenance expenditures (i.e. the ratio of maintenance expenditures to total operating costs) – thus implying that a decrease in profits will result in a decrease in maintenance expenditures. However, a significant statistical relationship was not found between relative maintenance expenditures and aircraft accidents. Thus, the profit–safety argument is only supported in part. A libewral interpretation of these results is that lower profits lower the safety margin (e.g. maintenance expenditures) of airlines but the lower safety margin may not lead to more aircraft accidents.

Market forces may also promote vessel safety. The market-response argument (Loeb, Talley and Zlatoper, 1994) states that shipping lines anticipating a deteriorating financial condition following a vessel accident will take safety precautions in a market environment. Shipping lines that are near bankruptcy might choose to reduce safety expenditures, thereby reducing costs and avoiding bankruptcy, but increasing the risk of vessel accidents. If vessel accidents increase as a consequence, the goodwill, however, of these lines will erode and their value to potential acquirers is likely to be lower. The market-response argument thus favors less regulation and public-sector enforcement for promoting vessel safety.

A test of the market-response argument for the airline industry is found in Mitchell and Maloney (1989). Specifically, Mitchell and Maloney (1989, p. 329) address the following question: “Are consumers reluctant to fly with airlines that have poor safety records or do they treat crashes merely as random events that bear no reflection on the quality of the airline?” If the former is true, the goodwill (or the value of the brand name) of the airline will decline, having an adverse effect on the performance of the airline’s stock; if the latter is true, a crash will not affect the performance of the stock. The authors investigated the abnormal stock market performance of airlines immediately following a crash. Two groups of crashes were considered – those caused by pilot error and those in which the airline was judged not to be at fault. Fifty-six such crashes between 1964 and 1987 were examined. For crashes caused by pilot error, the airline experienced statistically significant negative stock returns; for crashes for which the airline was not at fault, there was no stock market reaction. Mitchell and Maloney (1989, p. 355) conclude: “since our results suggest the market is quite efficient at punishing airlines for at-fault crashes, the need for increased airline safety regulation is not apparent.”

Alcohol consumption and intoxication of a vessel’s crew may not only cause the vessel to have an accident, resulting in crew injuries and deaths, but may also result in crew injuries and deaths without a vessel accident occurring. For passenger vessels, passenger intoxication can have similar results. It is well known that alcohol affects human performance. Specifically, alcohol affects human balance, increases risk-taking behavior, increases choice reaction time (the time a person needs to decide which of two responses is correct), has a detrimental effect on hand-eye coordination, and reduces one’s ability to make precise positioning movements of limbs (US Department of Transportation, 1988). Since alcohol affects balance, intoxicated crewmen and passengers are more likely to fall overboard than when sober. Further, given that water compounds the effects of alcohol on human performance, intoxicated crewmen and passengers are thus less likely to recover from falling overboard than when sober.

When crew or passengers fall overboard, water may interact with alcohol and compound alcohol’s effects on human performance. Specifically, alcohol can magnify the effects of caloric labyrinthitis – becoming disoriented, nauseous, or both, when water different from normal body temperature enters one’s ears. An intoxicated person whose head is immersed may become so disoriented as to swim down to death instead of up to safety. Cold water can affect muscle control (peripheral hypothermia) and thus compound alcohol’s effects on physical coordination, further impairing a swimmer’s abilities. Also, cold water may further impair an intoxicated swimmer’s air supply. The combination of inhalation (or gasp) response when suddenly placed in cold water and alcohol induced hyperventilation can result in aspiration of water and rapid drowning.

Safety regulation itself may cause a vessel accident – i.e. safety regulation may affect the allocation of resources by increasing the frequency of safety diminishing behavior (the safety offsetting behavior hypothesis). For example, the government regulation that automobiles must contain air bags may result in the drivers of such automobiles to be willing to undertake unsafe driving practices (e.g. driving aggressively), thereby resulting in automobile accidents (Peterson, Hoffer and Millner, 1995). Alternatively, as is often expected, safety regulation may have a safety compensating behavior effect – i.e. safety regulation will affect the allocation of resources by increasing the frequency of safety enhancing behavior. For example, in a study by McCarthy and Talley (1999) analysing recreational motorboat boating accidents, increases in boating safety training of boat operators increases the probability of the operators wearing safety floatation devices while boating.²