APPLICATION OF THE SYSTEM DYNAMICS TO THE WRECK OF THE TORREY CANYON

Transcription

1 APPLICATION OF THE SYSTEM DYNAMICS TO THE WRECK OF THE TORREY CANYON Shinji SHIMA 1, Kenji ISHIDA 2, Masao FURUSHO 3, Masaki FUCHI 4 1 A senior student Faculty of Maritime Sciences, Kobe University, JAPAN Tel: shimashin2@hotmail.com 2 Prof. Ph.D. Graduate School of Maritime Sciences, Kobe University, JAPAN k-ishida@maritime.kobe-u.ac.jp 3 Prof., Ph.D., Capt. Graduate School of Maritime Sciences, Kobe University, JAPAN Tel: Mobile furusho@maritime.kobe-u.ac.jp 4 Lecturer Capt. Graduate School of Maritime Sciences, Kobe University, JAPAN mfuchi@maritime.kobe-u.ac.jp ABSTRACT The wreck of the tanker TORREY CANYON occurred in This resulted in the biggest oil pollution accident ever recorded up to those times. At the official report of the Liberian Board of investigation, only the Master of the TORREY CANYON was owned the responsibility of this marine casualty. But they did not indicate the specific quantified data why this marine casualty had happened. This is because there is no synthesized method to major human factors. In this study this marine casualty was analyzed by the method of the m-shell model to extract the factors. Then the important factors are chosen and quantified. Finally, these quantified factors are applied to the System Dynamics so as to know the transition process of the wreck and safety voyage for future work. The System Dynamics is a simulation tool to understand the complex system over time which allows us to simulate over time without model time into the system. And the models are created graphically, so the relationships of the factors are easily understood. Furthermore, we can model feedback structures, that is, structures where cause and effect feed back on each other. For those reasons, this simulation tool is applied to various fields such as economics and social studies and could even be applied to almost unlimited fields. Through this study we intended to clarify how human factors affected the wreck of tanker TORREY CANYON just immediately before it and we also intended to identify whether the application of System Dynamics to the marine casualty is totally effective. Keywords : System Dynamics, TORREY CANYON, Human Factors 1. INTRODUCTION 117

2 It is said that industrial accidents will never disappear from the world. This opinion could be applied to the field of merchant marine transportation. In these decades navigational equipment and the structure of ship herself have advanced much. On the other hand, the number of marine casualty has not decreased. These are due to two main reasons. Firstly, the number and the size of ship have emerged. The quantity of cargo is also increased along with it. The other reason is that the problem of human factors is not solved. In the statistics of Japan, 80 percent of marine casualties are caused by the human factors and almost all of marine disaster relates to human actions (1). It could be said that to lose or to weaken the damage of marine casualties, we have to understand human factors more Objective of This Study Though the most of investigation about the marine casualties refers to human factors as reasons of accident, almost all of them do not show the quantitative data of them. This is because there is no synthesized method to major human factors. But the quantitative data is needed to know the details of casualties and to prevent future accidents. This study had tried to show the effect of human factors quantitatively by using the System Dynamics and applied this to the wreck of TORREY CANYON. The merits of using the System Dynamics are not only that we can gain the quantitative data, but also that we can visually understand the transition process. Furthermore we can output many kinds of data as a graph or chart. From these reasons, this study also intended to identify whether the application of the System Dynamics to the marine casualty is totally effective. 2. LITERATURE SURVEY 2.1. The wreck of the tanker TORREY CANYON On 18th March, 1967 the tanker TORREY CANYON struck Pollard s Rock in the Seven Stones reef, just west of Land s End, England. She left Mina al-ahmadi on 19 th February with full cargo of crude oil then planed to go Milford Haven, England. She was the first of the big supertanker, carrying a cargo of 120,000 tons of oil. In this accident the 80,000 tons of oil leaked from the ship and spread along the sea between England and France, killing most of the marine life it touched along the whole of the south coast of Britain and the Normandy shores of France (2). The Liberian Board of the Investigation had investigated the wreck of the tanker TORREY CANYON to appear why this marine disaster had happened. They said that there was no failure of machines and ship herself. They judged that this marine casualty had happened because the Master couldn t practice Good Seamanship. Especially, he had judged to change the course to the east of the Schilly Isles by himself. Furthermore, they stated that he had not taken proper actions such as reducing the speed and steering the ship in hand mode. The track of the TORREY CANYON to the wreck is shown in Fig

3 0850 Wreck Planed Course Hard a Port to to 013 Figure 1: The track of the TORREY CANYON 2.2. m-shell model The m-shell model is a theoretical framework designed to aid in the analysis of accident s interactions between people and all other aspects composed of Software, Hardware, Environment, Management and other people. In this model people is referred to as Liveware. Edwards had firstly proposed SHEL model in SHEL is an abbreviation of Software, Hardware, Environment and Liveware. Hawkins who was a pilot of KLM Royal Dutch Airlines had developed this SHEL model to SHELL model in He added Liveware to SHEL model and change the shape as is shown in Fig.1. The m-shell model is a concept which the management of NTBS s 4M is added to SHELL model (3). At first SHELLl model was developed in the Airlines field as a model of human factors. As is shown in the Fig.2, Liveware is put on the center and the others surround it. This concept means that human errors are ought to happen when there are problems at the borders between Liveware and the others (4). In m-shell model the problem between Liveware and the others are picked up and classified. The factors of Liveware and management are also included. 119

4 Figure 2: Figure of SHEL, SHELL and m-shell model (5) Application of the m-shell model In table 1, Examples of m-shell model are shown. The factors could be picked up and classified to L-L, L-S, L-H, L-E, m and Liveware itself. Application of the m-shell model to the wreck of TORREY CANYON is shown in Table 2. Factors Table 1: Examples of m-shell model Examples Management M organaization atomosphere of workplace safety culture Software S manual check list text for training or education Hardware H interface of man and machine automation arrangement of machine design of machine 120

5 Environment working environment E working characteristics Liveware L phsycological state Somatic condition abilities Liveware(others) L communication leadership teamwork L-L Master, First Officer Table 2: Application of the m-shell model First Officer noticed that the ship is off the course enormously. But the Master decided to pass through the west of Schilly isles without discussion with First Officer. Master, Third Officer Master, Quarter Master At the wreck, Third Officer was on duty. Ship's position was not sure. Before the wreck, the Master ordered Quarter Master 'Hard a Port'. Quarter Master did not change the steering mode from 'Auto' to 'Manual' Master, Fishing boats Master couldn't communicate with Fishing boats. L-S Software There was no principle of changing steering mode. L-H Hardware1 Maneuvering characteristics of the ship. Hardware2 No Decca was on the ship. Hardware3 No communication tool between the ship and light-ship. L-E Environment1 Fishing boats in the area. Environment2 Strong sea current. 121

6 Environment3 To enter the port by 23:00 because of tide. Environment4 To shift the cargo oil. M Management1 The problem of working time. Management2 Limited voyage information to the crew. Management3 Laws about how to voyage. L Master Master didn't make Stand-By Engine. Master had no experience to pass through this area Master didn t set the emergency positioning of the crew 2.3. The System Dynamics In 1956, Jay W Forrester from MIT Sloan School of Management initiated the System Dynamics. At the beginning it was applied to study industrial management field and then was rapidly implemented for other field of study such as economics, social studies, transportation engineering and could even be applied to an almost unlimited field. The field is now matured through more advanced research and applications. System Dynamics (SD) is based on the theory of non-linear dynamics and feedback control. It provides a framework to model complex systems and embedded feedback structures between various elements. SD concerns a system s dynamic behavior over time under various conditions. SD is designed to investigate cause-effect relationships among equipment as well as the characteristic and functions of a complex system. Uncertainty is another reason why SD is used. Using SD, we can simulate the uncertainty of system easily to investigate the system response and interaction. Through this better understanding, a conscious learning on interaction among components in a system can help decision makers in doing a better operation and/or maintenance of the component and system in order to reduce possible risk. Using SD, the cause-effect interactions and relationship linking components in a system could be obtained and a better understanding of system behavior expected (6) Application of the System Dynamics As a software of the System Dynamics, Powersim Studio 2007 was used. In the model, geographical data like depth of the water, ship maneuverability, the planned course and speed of the TORREY CANYON are structured. Fig 2 shows basic modeling for the movement of her. In this study, three cases of simulation had implemented. Each case is implemented 500 times and the rates of the wreck are gained. Case 1. The instability of the positioning by Third Officer and the delay of action by Quarter Master are structured with no fishing boats in the area. Case 2. The instability of the positioning by Third Officer, the delay of action by Quarter Master and existence of fishing boats in the area are structured. Case 3. Existence of the fishing boats in the area is structured without the instability of the positioning by Third Officer and the delay of action by Quarter Master. 122

7 Pass through N- S Pass through E- W Position of ship N- S Depth of the position Position of ship Movement of ship Position of ship E- W Depth of the position2 longth of cube Third Officer Drought wreck sin speed cos speed Quarter Master New course Course Speed maneuverability maneuverability2 NEWS2 North NEWS1 West South East NS EW Figure 2: System Dynamics model 3. RESULTS AND CONSIDERATION 3.1. Results The rates of wreck are gained from three types of simulation. These are shown in the Table 3. Table 3: Rates of wreck The wreck rate Structured factors Case 1 0% Third Officer, Quarter Master Case 2 60% Third Officer, Quarter Master, Fishing boats Case 3 32% Fishing boats 123

8 3.2. Consideration These results show that the rates of wreck are 0%, 60%, 32% in each simulation. The result of case 1 shows that if there were no fishing boat in the area, the wreck could not be caused. In this situation the human factors of Third Officer and Quarter Master had not affected the wreck rate. But the actual situation was that fishing boats were there and they had disturbed the safety movement of the TORREY CANYON. In case 2, the rate of wreck is 62% and this is the highest rate. This result appears that the fishing boats affect a lot to the wreck. The comparison for the case 2 and case 3 shows that human factors of Third Officer and Quarter Master also affect the risk of the wreck. The comparison for the case 1 and case 2 shows that there is influential level of human factors by Third Officer and Quarter Master. It depends on the area space. If the area is like narrow channel, these human factors affect a lot. On the other hand if there is enormous space, these factors do not affect. 4. ANALYSES AND DISCUSSION The factors of these simulations were picked up by the m-shell model. So, all of the factors could be considered in the system to which the Master who has all of the responsibility in the ship relates. From these results it could be quantitatively said that the judgment of the Master to pass through this area was the biggest fault. With no effect of factors by Third Officer and Quarter Master there was a profound danger in this area. The Master who has all the responsibility must judge whether there is a space to safety voyage. Also, he had to consider the relations between the space and influential level of the human factors. In this model the speed of her has not changed as the actual simulation. If the Master had decided to reduce the speed, the space to pass through the area could make. These things could be also embedded in the simulation model. From all of the factors like ship maneuverability, geographical restrictions, other ship s restrictions and human factors, the tool to know the risk of voyage beforehand could be made in the System Dynamics model. This simulation tool could help people to understand marine casualties from various quantitative angles. At the same time, this simulation tool could be used as a risk prediction model for the determiner in the ship and as an educational tool. 5. CONCLUSION In this study the m-shell model and the System Dynamics are applied to the wreck of TORREY CANYON and the main factors of the wreck are compared quantitatively. And it appeared that this model could use in many situation. It could be said that the model structured by the System Dynamics have the possibility to be used as analyses, prediction and educational tool. 6. REFERENCES (1) Yusuke, Y. (2002), Outline of Ship Safety in Japanese (2) Crispin, G., Booker, F., and Soper, T. (1967). The Wreck of the TORREY CANYON (3)(4) Yusuke, Y. (2002), Outline of Ship Safety in Japanese (5) Access Date : (6) Powersim Software(2007), Powersim Studio 2007 user s guide 124