Analysis of Human Error in the Maritime Industry

The navigation of ships is usually dependent on the use of technology and a lot of calculation by the captain, once it is in the high seas (Kuwata, Wolf, Zarzhitsky and Huntsberger, 2014, p.119). Some of the supporting features which aid captains are benched on the design of the ships, which offer the needed positioning to overcome the murky waters and its impacts. In order to delve into how the ship and its management operate, Duffey and Saull have had a substantial contribution to the field, trying to help locate the intrigues that enable ships to sail in the waters. The use of technology has been met with mixed reactions, as it has both negative and positive impact based on the knowledge and its practical implementation. Modern ships are fitted with infallible technology which guarantees safety for the users. The maritime world, just like any sector of the economy has witnessed lots of changes in means of manufacturing the ships from the traditional products and designs to modern designs that are bespoke. However, there are gaps resulting from the use of technology which makes the chance of accidents not to be fully unavoidable. Hence, this still causes Marine accidents that lead to loss of lives and property, as well as endangering the aquatic life. The costs associated with fatalities are imaginable both to the investors or owners of the ships and those who lose their belongings and loved ones whenever an accident occurs ((Duffey and Saull 2003, p.284).

On a positive note, technology has more benefits that accrue to its existence. For instance, Duffey and Saull have attested that the risks in the high seas have been averted due to the existence of technology which is useful in predicting the navigation path of ships. Thus issues such as death traps and damages of the ship have been averted from human errors, through looking at the level of experience in using technology to build, operate and even manage ships (Duffey and Saull 2005, p.8).

Human error according to Duffey and Saull

Human error can be looked at from a variety of causes to maritime accidents. When looking at human error, a number of reasons have to be factored in (Grech, Horberry and Smith 2002, p.1718). For example, a decision by a pharmacist to offer a diagnosis without looking at the disease leads to misdiagnosis, the same applies to a marine industry where making a wrong decision may be detrimental to achieving safety of passengers.

The factors can be multiple in nature and range from technology, work practice such as multiple crews who bring different interpretation, the experience of the captain, communication barriers among other identifiable reasons. Even trained and experienced sailors often make costly mistakes, just as the technology that depends on the knowledge of the user. Thus, it becomes hard to erase the impact of human involvement in the running of ships (Grech, Horberry and Smith 2002, p.). Duffey and Saull have worked on means of managing risks trough coming with concepts, which revolve around human influence in the safety of the ships, both on-shore and offshore.

The use of the universal learning curves has been instrumental in determining means and way of increasing the level of precision, through predicting tragedy, accidents, and failures in the maritime industry. According to Duffey and Saull (2005, p.8), the weight training involves using Backpropagation learning method to determine the rate of error.

E (w) = ∑ (p= 1 to PT) ∑ (i= 1 to l) [di(p) – yi(p)]2 ,

where

E (w) = error function to be minimized,

w = weight vector,

PT = number of training patterns,

l = number of output neurons,

di

(p) = desired output of neuron I when pattern p is introduced to the MLP, and yi(p) = actual output of the neuron I when pattern p is introduced to the MLP

Through using data mining techniques, the rates of accidents are reduced as it helps in informing the chances of occurrence. The need to engage the relevant bodies on means of reducing fatalities calls for urgent measures, all which points at means of averting future fatalities from a multi-sectoral approach. The database of Fatalities Analysis Reporting System (FARS) is essential in looking at means and ways of reducing chances of collision among cars in the motor-vehicle industry (Duffey and Saull 2008, p.256).

The risk assessment that Duffey and Saull identified are closely linked to the level of concentration needed in the maritime world. People who engage in risky business, for example, tend to be more cautious in their trade dealings. The risk assessment is essential in areas where lives are involved. Hence, there should be a calculated move when dealing with machines, which are technology driven (Duffey and Saull 2002. p.67). As a conscious person, no one wishes to be part of the death statistics. Hence, the human error usually occurs without the knowledge of one committing the error.

With the invention of technology, the expected error rate has reduced significantly as the use of technology aids in the navigation process. The engineers who are involved with checking on the probability of the risk occurring usually have to be accountable to the companies that manufacture ships Manufacturing & Service Industries, 13(4), pp.279-291. (Duffey and Saull 2005, p.45) Duffey notes that the risk assessment is not an individual person fault, but a combination of different factors, all which plays out to the company. Thus, the possibility of a risk occurring has to go through the quality checks with reliability engineers always being called to task to check out the need to have measures deemed fit for achieving a final guarantee of passengers safety.

Reducing Errors during Navigation

In navigation within ships, engineers, as well as managers, often create a force to reckon with in terms of laying the foundation for eliminating errors that are associated with navigations at different levels. In respect to Technology, the General Accident Theory has been used to determine the level of human interaction with technology and the probability of reducing errors associated with human beings. The learning curve needs to be identified when dealing with ships navigation, factors which help in utilizing possible outcomes of minimizing errors during ships navigations (Duffey and Saull, 2008, p.249). Hence, human errors are reduced, with the use of a learning curve which predetermines past experiences to help come up with reliable solutions.

According to Duffey and Saull’s (2005, p.48), the perspective of reducing the incidents of accidents, the learning environment needs to be conducive for the captains, pilots and other navigators to achieve the best navigation. Duffey and Saull view the probability of reducing accident is dependent on the accumulated experience of one using the vessels. Hence, in areas with higher prospects of an accident happening such as on air, rail, shipping, industrial accidents, at hospitals all use the theory to reduce the likelihood of human errors since the design, manufacture, and maintenance of various machines is usually pegged on the experience garnered in use of respective technology. Hence, according to Duffey and Saull (2008, p.247), the universal learning curve offers the most reliable source of looking at experience and probability of accidents.

The same is applicable while looking at means of reducing the risks, through the universal learning curves. The past crash of the NASA Space Shuttle that led to deaths is a sharp reminder that there are levels of risk which may be hard to determine the causes, but it forms part of the statistics of determining the error rates. Hence, the pointers towards such incidences have shown that there are some incidences within human intervention that can be diagnosed to offer a solution at the initial stages, through looking into means and ways of reducing the possibility of a repeat of the same (Juang, Hou and Lee 1997, p.257). The worldwide trend in errors associated with human element has sharp pointers to areas that can be addressed through interventions in areas

Managing Maritime Risk

Through including human error, it is possible to manage risks associated with the human element. Incorporating the human element makes predictability of errors to be enhanced, hence ease in creating avoidance techniques. According to Duffey and Saull (2008, p.248), the universal learning curve helps in identifying the human errors through creating models of identifying the human error from the past recorded incidences.

p(ε) ≡ F(ε) = 1 - e-∫ λdε

Where, from the Learning Hypothesis above, at any given experience, ε, the failure rate, λ(ε) naturally includes the human element which is given by:

λ(ε) = λm + (λ0 - λm) exp - k(ε-ε0)

Looking at the homo-technological systems (HTS), it becomes easy to lay the basis of reasoning as it is almost impossible to describe all gradations, variations, and possibilities of the risk happening (Duffey and Saull 2003, p.284). Using the above formula, it is possible to calculate the desired route that a ship captain can take while on high seas to minimize the chance of error that may lead to fatality (Celik and Er, 2007,p. 3). Based on the past experience of the navigators, the probability of past failure re-occurring all becomes factored in in the analysis of the research, to provide a comprehensive analysis. Hence, the essential bits of moving across different locations help in reducing the chances of risks. The crew that is tasked with sailing the ship needs to have a higher knowledge of interpreting various signals from the ship in unity, since any differing opinions may jeopardize the chances of creating a safety net for the passengers. Issues such as the depth of the sea at different points, the route map to avoid collision with other ships and water users, the ability to predict problem ahead and make corrective measures on time are just but few mentions of the most probable moments when the ship crew can take advantage of.

Through using the Minimum Error Rate Equation (MERE) formula, it is possible to iron out major gaps in the management of ships (Duffey and Saull 2008, p.247). In order to gain ground on reducing chances of accidents happening, the use of non-dimensional error rate is essential of providing the overview of how errors can be minimised to the lowest level possible. The component is useful in outlining the design, operation, and management of ships can be achieved through incorporating:

E* =A*/A*o= exp (-kT)

NB: A*o = (1-Am/Ao) whereas A* = (1-Am/A) and k is learning rate constant. From the formula, it is possible to determine how the ship’s captain can reduce the level of errors during navigation, through identifying the possibility of having such outcomes should the ship’s captain use technologically driven ships to minimize the error and accident rates.

The design

Ships have a unique design, which emanates from the technology used to navigate such as the compass to the direction of the ship. The use of a liquid magnetic compass is an essential component for taking bearings of the route map of the ship Patraiko et.al. 2009, p.219). Hence, a captain has to be on the look-out for signals which may depict signs of danger during navigation. The purpose of using ship navigators is to enhance berth to berth navigation, something that has been instrumental in defining the maritime industry. The outcome of having better navigational know-how among captains is a reduction in the percentage of ship losses, the guarantee of maintaining ship balance in the much dreaded high waters among passengers.

The engineers who are tasked with the responsibility of designing ships often tend to rely on the past experiences to curve out ultra-modern features which are able to circumvent the past challenges. In order to do so effectively, the resulting outcome of the study ranging from a span of 200years becomes essential, which is used to make the present design. Managers in charge of the management of ship have to understand this concept as it affects their business operations in a way or another (Duffey and Saull 2005, p.5). Hence, having wider knowledge about the design of the ship helps in making some of the informed decision such as the route not to use based on the design of the ship.

The learning environment usually has an impact in determining which direction that a ship takes (Li, Lee, Jun, and Lim, 2008, p.3205). The people who with technology often learn from mistakes, that they try to minimise chances of reoccurrence. Hence, it is necessary that the learning outcome from errors committed by others defines future room for a study to help identify the causes of errors and corrective measures for any future error occurring in relation to the past experience. Irrespective of the interval of years that one has experienced in the line of duty, chances of an error, either by the commission is likely to become hard to avoid.

Conclusion

The need to have safety measures in the maritime industry is a call that cannot be overemphasized. Through modeling the ships learning outcomes from past experiences, it becomes imperative to have all measures in place. The human errors that have been witnessed in the past can form the tuning point of making the waters safe places once more. Duffey and Saull’s approach to means of defining human factors has been essential in the field of automobile, aviation, maritime among other fields. Hence, it becomes of great importance to calculate the margin of error while navigating to avoid chances of the accident occurring. Thus, engineer’s works are shaped by past experiences and the universal learning curves. The level of homo-technology in the wake up of fatal accidents needs to have a paradigm shift, as approaches to safety need to be advanced to eliminate chances of past errors reoccurring again.

References

Celik, M. and Er, I.D., 2007. Identifying the potential roles of design-based failures on human errors in shipboard operations. TransNav, International Journal on Marine Navigation and Safety od Sea Transportation, 1(3).

Duffey, R.B. and Saull, J.W., 2002. Know the Risk. Butterworth and Heinemann, Boston, USA.

Duffey, R.B. and Saull, J.W., 2003. Errors in technological systems. Human Factors and Ergonomics in Manufacturing & Service Industries, 13(4), pp.279-291.

Duffey, R.B. and Saull, J.W., 2005. The Probability of System Failure and Human Error and the Influence of Depth of Experience. In Proceedings of International Topical Meeting on Probabilistic Safety Analysis (PSA’05), San Francisco, CA, September (pp. 11-15).

Duffey, R.B. and Saull, J.W., 2008. Managing and predicting technological risk and human reliability: a new learning curve theory. Proceedings of the Institution of Mechanical Engineers, Part O: Journal of Risk and Reliability, 222(2), pp.245-254.

Grech, M.R., Horberry, T. and Smith, A., 2002, September. Human error in maritime operations: Analyses of accident reports using the Leximancer tool. In Proceedings of the human factors and ergonomics society annual meeting (Vol. 46, No. 19, pp. 1718-1721). Sage CA: Los Angeles, CA: Sage Publications.

Juang, B.H., Hou, W. and Lee, C.H., 1997. Minimum classification error rate methods for speech recognition. IEEE Transactions on Speech and Audio processing, 5(3), pp.257-265.

Kuwata, Y., Wolf, M.T., Zarzhitsky, D. and Huntsberger, T.L., 2014. Safe maritime autonomous navigation with COLREGS, using velocity obstacles. IEEE Journal of Oceanic Engineering, 39(1), pp.110-119.

Li, J.H., Lee, P.M., Jun, B.H. and Lim, Y.K., 2008. Point-to-point navigation of underactuated ships. Automatica, 44(12), pp.3201-3205.

Patraiko, D., Wake, P., Weintrit, A., Guan, K., Shi, C., Wu, S., Xu, T., Zalewski, P., Im, N., Seo, J.H. and Vejražka, F., 2009. e-Navigation and the Human Element. Mar. Navig. Saf. Sea Transp, 4, p.29.