1 INTRODUCTION

Digitization and automation will have a great effect

on shipping in the future. Hereby, not only navigation

or single ship operation is affected, but in most cases,

large-scale effects on sea traffic and marine processes

are expected. Depending on the concrete digitization

and automation project, those are potentially also

addressing safety related issues.

As soon as digitization and automation affects

safety, the legal framework and thus the requirement

for a Formal Safety Assessment (FSA) is often touched

as well. A FSA is “a tool to help in the evaluation of new

regulations for maritime safety and protection of the marine

environment or in making a comparison between existing

and possibly improved regulations, with a view to

achieving a balance between the various technical and

operational issues, including the human element, and

between maritime safety or protection of the marine

environment and costs” (IMO, 2007).

A critical step in safety assessments and in the FSA

in general is the transition from the hazard

identification to the quantitative risk analyses. While

international studies agree, that the human factor is

the key component behind most marine accidents

(Sanquist, 1992), (Rothblum, n.d.) recommended risk

assessment methods are mostly classical fault tree

analyses or expert reviews (IMO, 2007), which are

potentially too restricted to fully cover safety effects

of those future technology, as the human element is

hard to assess. Especially regarding the human

element, ship-handling simulation is the only known

method to fully incorporate this into a scientific set-up

and thus “the results, conclusions and recommendations

can be based on a thorough review of technical aspects, as

well as the important human factors, such as response

times and communication” (PIANC, 2014).

So far, SHS was limited to a small number of

simulated vessels in a joint exercise, thus

organizational affects or large scale changes to

waterborne processes, procedures and technology

Assessing Safety Effects of Digitization with the

European Maritime Simulator Network EMSN: The Sea

Traffic Management

Case

-C. Burmeister, T. Scheidweiler, M. Reimann & C. Jahn

Fraunhofer Center for Maritime Logistics and Services CML, Hamburg, Ger

many

ABSTRACT: This paper will give an intro into the technical backbone of the EMSN and derive necessary

specifications to allow for objective testing in large scale. Further, it will demonstrate the potential of EMSN as

a maritime safety test-bed on the STM case. Therefore, the simulations executed within the EMSN are evaluated

regarding their safety level in order to demonstrate the effects of various measures to improve safety. Based on

a fuzzy logic approach, numerical Data from the EMSN Data-Tracker is used as an input to assess a present

traffic situation from the perspective of a specific ship and outputs a comprehensive safety index developed by

expert opinions. The safety index is used to further analyze navigators’ behavior and decisions in different

maritime traffic scenarios that are executed within the EMSN.

http://www.transnav.eu

the

International Journal

on Marine Navigation

and Safety of Sea Transportation

Volume 14

Number 1

March 2020

DOI:

10.12716/1001.14.01.10

could barely be assessed. Thus, the European

Maritime Simulator Network (EMSN) has been

developed to provide a large-scale test environment

for maritime safety (Rizvanolli, et al., 2015).

Chapter 2 will give an overview about the

backbone and technical features of the EMSN as

implemented today, before Chapter 3 outlines the Sea

Traffic Management (STM) case and how this concept

was assessed with the help of EMSN. Chapter 4

draws conclusion regarding the usage of EMSN as

well as about its further potentials for maritime

training and research.

2 THE EUROPEAN MARITIME SIMULATOR

NETWORK

The EMSN is a network that connects numerous ship

handling simulators (SHS) from different simulation

sites across Europe. Enabling joint exercises between

this remote locations with different simulation

manufacturers requires a joint understanding of the

logical and technical connectivity between the SHS

(John, et al., 2014), which is ensured within the EMSN

by the IEEE Standard 1278.1-1995 as well as some

EMSN-specific joint agreements (IEEE, 1995),

(Poschmann & Burmeister, 2017).

This set-up enables users to operate individual

ship bridges in configurable scenarios and to interact

in real time with each other in a simulated

environment.

At the end of 2018, 13 simulation centers in 7

European countries with a total of more than 30

bridges were connected to the EMSN.

In principle, EMSN ensures alignment between

SHS within the following areas:

− Exchange of simulation data (ground truth)

− Exchange of communication data

− Centralized data tracking

− Synchronization of simulation management

2.1 Topology

The mentioned exchange services are deployed in IP

networks and implemented as Virtual Private

Networks (VPN). For this purpose, VPN tunnels are

set up between EMSN simulation sites to provide

confidential and authenticated connections with

integrity over the public Internet. For realization, a

hub-and-spoke topology was selected. Its general

structure is shown in Figure 1 (John, et al., 2014).

Thereby, the DIS interface guarantees the so-called

common ground truth exchange, whose functionality

is described in the section 2.2. As seen in Figure 2, this

interface connects each SHS to the EMSN

infrastructure. NMEA-interfaces at each SHS do

further allow integration of innovative marine

equipment based on standard interfaces, as also used

within the STM project.

Figure 1. Overview of the VPN technology of EMSN.

At each site, the SHS has different interfaces to

allow for an easy integration of EMSN as well as

potential digitization and automation prototypes to be

tested.

Figure 2. Overview of SHS interfaces within EMSN.

(according to (Poschmann & Burmeister, 2017))

2.2 Ground truth and communication

The ground truth exchange basically ensures that all

manually controlled own-ships of one center are

represented by a remote-controlled traffic ship at all

other centers (Rizvanolli, et al., 2015). This is based on

a so called Entity State PDU of the IEEE standard

(IEEE, 1995). Hereby, not only the position and type

of vessels are exchanged, but also the used light and

shape signals, which are currently the only

mandatory way to indicate the own ship navigational

status. Among others, EMSN thus enables an

interactive exchange navigational status, as e.g.

(Poschmann & Burmeister, 2017):

− Not under command

− Restricted ability to maneuver

− Constrained by her draught

− Fishing and

− Anchoring.

Out of the standard marine communication

channels, voice communication by VHF as well as the

Automated Identification System AIS is fully

integrated into the EMSN to allow for realistic

scenarios. VHF is implemented based using a

separate TeamSpeak-Server within the EMSN, while

AIS is fully integrated into the DIS concept as Signal

PDU. Thereby not just the most frequently used

position report (AIS message type 1,2,3) and the Static

and Voyage related data report (AIS message type 5)

is implemented, but also addressed and broadcasted

binary message exchanges (AIS message type 6, 8, 12,

and 14) (Poschmann & Burmeister, 2017), (ITU, 2014).

Besides those standard channels, application specific

channels and communication systems for the

digitization and automation system to be tested can

be integrated into the network.

2.3 Central data tracking and simulation management

By using the DIS-standards Data PDU a wide variety

of data can be centrally recorded in customized

simulation environments and scenarios. This can

serve as the basis for the evaluation of scientific

problems, as in the risk assessment of STM

Validation.

An overview of the data types recorded during the

simulation runs in STM Validation is shown below.

− Ship Data

General information describing specifications of

the vessels used in the simulation scenarios, for

example length overall, breadth overall, IMO

number, MMSI, etc.

− Motion Data

Variety of information describing the movement of

a vessel under consideration, e.g. heading, speed

over ground, engine order telegraph, etc.

− Environmental data

Environmental information that influences the

vessel’s movement or voyage, such as wind

direction and speed, current direction and speed

as well as visibility.

− General data

General simulation information and identifiers for

the assignment of centers and ships, for example

simulation id, site id, timestamp, etc.

Besides, central simulation management tools like

the Start/Resume PDU as well as the Stop/Freeze PDU

enable synchronization of local simulations (IEEE,

1995).

3 EVALUATION

The STM concept contains several functions whose

effects on the navigation behavior of seafarers

involved have been investigated within the EMSN.

These include the following services.

− Chat function

Real-time communication via a stand-alone chat

program on the ECDIS client. It allows the

exchange of simple text messages between vessels

and shore center.

− Receiving navigational warning

Notifying seafarers about the occurrence of new

navigational warnings. Direct presentation of the

area to be avoided on the ECDIS.

− Receiving route suggestion from Shore Center

Submission of a recommended route from a Shore

Center to a ship concerned, to be checked by the

bridge team. At the discretion of the captain, the

route recommendation can either be accepted or

rejected.

− Rendezvous function

Display of the intersection between the routes of

considered vessels using the AIS-based Closest

Point of Approach (CPA) and Time to Closest

Point of Approach (TCPA) on the ECDIS.

− Ship-to-Ship-route-exchange

Display of the next 7 waypoints of a monitored

vessel on the ECDIS.

3.1 EMSN Methodology

The main objective of the EMSN tests is to provide a

further validated input to the FSA risk assessment,

especially concerning the evolvement of situational

awareness and traffic patterns by applying STM.

Hereby, the test methodology itself consists of four

individual stages:

− Selection of appropriate scenarios

− Simulation of scenarios with and without STM

equipment

− Safety assessment of encounters of each scenario

− Comparison of safety assessments.

This study is not attempting to make a

comparative analysis of the possible effects of each

individual STM service on traffic safety separately.

Rather, this study is an attempt to capture the possible

effects of several STM services being available at the

time of the simulation runs based on numerical data

collected during the simulation trials in the EMSN.

Other factors which may have an influence when

analyzing possible effects of introducing STM services

such as usability of ECDIS in general, the

familiarization and training in the use of the services,

the experience of the test participants, etc. have not

been considered in this study. In addition, it should

be emphasized that within the numerical data

analysis only two runs have been analyzed (one with

STM services and one without). This fact is conformal,

since the depicted runs are considered as

representatives of all executed simulation runs.

3.2 Scenario Selection

The English Channel and the Southern Baltic were

selected as they are good examples of heavily

trafficked areas. The Baltic scenario was created

within the Fehmarn Belt representing one of the

worlds’ busiest traffic corridors with numerous

recommended routes, junction areas and crossing

ferry routes, but no Traffic Separation Scheme in

place. For the STM runs, a simulated Shore Centre

“Baltic Shore Centre” was established. The English

Channel scenario was created for the south coast of

England with the port of Southampton playing the

major port of interest and a fictitious “Shore Centre

Southampton” on the Isle of Wight was established

for the STM runs.

Eight scenarios were specified based on the

combination of three variables: location, time of day,

and visibility. Each scenario was executed several

times with and without the availability of STM

services. The simulations were conducted in the

EMSN consisting of up to 30 manned bridges during

four sessions.

For the later evaluation, the simulation data was

used, which was recorded using the data tracker

described in section 2.3. For each run, around 210,000

movement data were available at a distance of one

second from the 30 ships.

3.3 Maritime Safety Index

In order to assess the safety of different encounter

situations and thus the validity of the available STM

services, a Safety Index (SI) was developed. The SI is

used to further analyze navigators’ behavior and

decisions in different maritime traffic scenarios that

are conducted within the EMSN. For the assessment

of navigators’ behavior in encounter situations

between ships, it is required to develop an approach

that accounts for the full complexity of the task. While

most assessment methods conventionally used

depend to a large degree on expert opinions, this

study aims for a more objective and quantitative

approach.

The most widely used approach for the assessment

of ship handling simulation is rating by expert

opinion. An obvious disadvantage of this

methodology is the high influence of subjective

judgement. This means that the same simulation

results can receive totally divergent ratings when

being assessed by different experts. To compensate

this drawback, an alternative approach will be used

for analyzing and evaluating the impact of the

available services at the time of the simulations in the

STM concept on ship traffic.

Within the STM Validation project, the level of

safety of different traffic situations will be measured

based on a fuzzy logic approach, cf. (Bai & Wang,

2006), (Kozlowska, 2012), (Mamdani & Assilion,

1975), (Perera, et al., 2011) and (Zadeh, 1965). The SI

may be used within the Formal Safety Assessment

(FSA) to assess the potential risk reduction by the

implementation of the STM and its various

operational services.

Overall, the safety index consists of a collision

index, a grounding index and an environmental

index, which parameters are presented in Figure 3.

Figure 1. Structure of the safety index for the

evaluation of the EMSN runs.

Following the definition of the input variables, the

membership functions for the fuzzy models

estimating a collision index are created based on the

results of pre-conducted instructor surveys and/or a

comprehensive literature research. In the following a

maritime traffic situation is given by one own ship

(OS) and one or multiple target ships (TS)

encountering in different situations: head-on, crossing

or overtaking.

The maneuverability of a ship is determined by the

block coefficient of length and breadth and the type of

ship transmitted via AIS. If the block coefficient is

small, then there is no good ability to maneuver, the

larger it becomes, the better the ability to maneuver

the ship (American Bureau of Shipping, 2006). A poor

maneuverability causes a deterioration of the Safety

Index. The maneuverability of a vessel strongly

depends on its maneuvering devices. This includes

the rudder, fixed lateral areas, transverse thrusters,

propeller (with fixed pitch or controllable pitch, Voith

Schneider propeller or azimuth thruster). In addition,

the engine has an influence on the maneuverability of

the vessel (two or four stroke engine or electrically

driven). Since AIS does not include this type of

information the maneuverability of a ship has to be

estimated. Thus, the maneuverability is according to

the AIS data a function of the following variables.

( )

[ ]

, , , , f Vesseltype L B good poor none=

(1)

The main idea is to give every ship type a

classification of the maneuverability. For passenger

ships, cargo ships, tanker and tugs an additional

factor will be considered. The ratio length to breadth

will change maneuverability index of the actual ship.

The larger the ratio, the slender the ship. This is good

for speed and course keeping, but rather bad for

maneuvering. For this ship types good L/B ratios will

be estimated. If a ship has a greater L/B ratio then the

estimated one, the estimated maneuverability will be

improved one level and vice versa.

The grounding index is determined by specifying

the squat of each vessel, which represents the

decrease of a ship’s under keel clearance due to

vessel’s movement in shallow waters. A small squat

has a small grounding probability. Given the block

coefficient c

B and the actual speed through water v of

a ship, the squat is given by (Serban & Panaitescu,

2016)

100

squat =

(2)

To determine the environmental conditions, two

fuzzy systems will be used: within a first one, the drift

of a ship given the current, wind and sea state will be

developed. The drift will be used to estimate the

maneuverability later on. The output of the linguistic

variable visibility will be directly used to define the

EI. The drift of a vessel is determined by the

environmental parameters current, wind and sea

state.

For more details on the mathematical backgrounds

of the indices, it is referred to (Olindersson, et al.,

2017).

3.4 EMSN Safety Assessment

To evaluate the effects of STM services, the safety

indices of runs without STM equipment ("base runs")

and runs with equipment ("STM runs") are

determined. For this purpose, each combination of

owned and foreign ships was considered and the

corresponding safety index was determined for each

point in time.

The histograms in Figure 4 and Figure 5 depict the

safety indices for baseline and STM runs. The results

indicate that no significant effect of these combined

STM services on maritime traffic safety could be

observed. If all SI of the runs are compared, it is

verified that both the runs without STM and those

with STM equipment are in a similar range.

Figure 4. Histogram of the safety indices of baseline runs.

(Scheidweiler & Weber, 2018)

Figure 5. Histogram of the safety indices of STM runs.

(Scheidweiler & Weber, 2018)

However, no separate analysis of the effect of

individual services on maritime traffic safety has been

made nor any analysis on usability and/or human

factors assessment. Therefore, the effect of the STM

services reported here should be compared with the

results of other evaluation methods to confirm it.

Due to the ongoing development of the EMSN,

scenario simulations had some sort of irregularity.

Most of the regularities have been ships freezing,

disappearing, and re-appearing, which reduced

throughout the EMSN maturity increased. As those

irregularity were also seldom and mostly of a short

duration of less than ten seconds, the authors

considered that these irregularities were unlikely to

have a large effect on the results of the numerical

analysis regarding the SI.

The results show that these combined STM

services do not significantly improve maritime safety

for a selection of simulation runs. However, neither a

separate analysis of the impact of individual services

on maritime safety nor an analysis of usability and/or

human factor assessment was carried out. Therefore,

the effect of the STM services listed here should be

compared with the results of other assessment

methods (Scheidweiler & Weber, 2018).

Additionally, it must be stressed that even if the

data itself does not significantly indicate an

improvement, the perceived safety benefits by the test

participants was rather positive. With an exemption

of the chat function, all tested services received more

positive than negative comments by the investigated

227 marines (Aylward, et al., 2018). Thus, a resilient

human factor assessment, which is now based on first

experiences and not solely on expert opinions, can be

derived from the EMSN setup providing input to an

objective FSA.

4 CONCLUSION AND OUTLOOK

This paper briefly showed the capabilities of the

EMSN and how it can be used to assess objectively

and human factor oriented the effects of marine

digitization and automation on safety based on the

STM example. In general, the use of the simulator

network EMSN to validate nautical research

hypotheses offers many advantages over large-scale

field tests, which are briefly outlined below.

− Saving money & time

− Focus on relevant research field

− Reduction of noise

− Risk-free investigation

− No impact on environment.

For the future, assessing additional STM services,

which have not yet been incorporated into the EMSN

is recommended and aspired before rolling them out

to shipping. Beyond STM, there are further marine

digitization and automation projects on the horizon,

where proper, simulation-based safety assessments

are required, that fully incorporate the human factor,

like e.g. the development or Maritime Autonomous

Surface Ships. According to the EU Directorate-

General for Research and Innovation, large-scale

virtual test facilities like the European Maritime

Simulator Network are required here to bridge “major

gaps with regards to development of safe waterborne

connected and automated transport” (European

Commission, 2017).

Besides applying the EMSN in research projects, it

is intended to broaden the use of the EMSN for joint

training of nautical cadets and officers during their

studies by incorporation real international training

between cross-border institutes. Therefore, the EMSN

now became project independent by the EMSN

Connect initiative, which ensures further usage and

maintenance of the EMSN beyond STM (Jahn, 2018).

REFERENCES

American Bureau of Shipping, 2006. Guide for Vessel

Maneuverability, Houston: s.n.

Aylward, K., Lundh, M. & MacKinnon, S., 2018. EMSN Test

Report - Evaluation of STM services: Human Factors.

[Online] Available at: https://s3-eu-west-

1.amazonaws.com/stm-stmvalidation/uploads/

20190403113930/STM_ID3.3.6-EMSN-Test-

Report_Evaluation-of-STM-services_Human-Factors.pdf

[Accessed 10 04 2019].

Bai, W. & Wang, D., 2006. Fundamentals of Fuzzy Logic

Control - Fuzzy Sets, Fuzzy Rules and Defuzzifications.

Advanced Fuzzy Logic Technologies in Industrial

Applications. In: s.l.: Springer - Verlag, pp. 17-36.

European Commission, 2017. Connected and Automated

Transport - Studies and Reports. [Online] Available at:

http://ec.europa.eu/newsroom/horizon2020/document.cf

m?doc_id=46276 [Accessed 10 04 2019].

IEEE, 1995. IEEE 1278.1-1995 Standard for Distributed

Interactive Simulation (DIS), s.l.: s.n.

IMO, 2007. Formal Safety Assessment. Consolidated text of the

Guidelines for Formal Safety Assessment (FSA) for use in the

IMO rule-making process

(MSC/Circ.1023−MEPC/Circ.392), s.l.: International

Maritime Organisation.

ITU, 2014. Technical characteristics for an automatic

identification system using time-division multiple access in

the VHF maritime mobile band ITU-R M.1371-5. [Online]

Available at: http://www.itu.int/dms_pubrec/itu-

r/rec/m/R-REC-M.1371-5-201402-I!!PDF-E.pdf [Accessed

10 04 2019].

Jahn, C., 2018. Ship handling in the simulator network - Great

potential for nautical safety. [Online] Available at:

https://www.cml.fraunhofer.de/content/dam/cml/en/doc

uments/Newsletter/Newsletter_4-18_EN.pdf

[Accessed 10 04 2019].

John, O. et al., 2014. Assessing the MONALISA 2.0 Concept:

Establishment of the European Maritime Simulation

Network. ISIS.

Kozlowska, E., 2012. Basic principles of fuzzy logic. [Online]

Available at: http://access.feld.cvut.cz/view.php?

cisloclanku=2012080002 [Accessed 10 04 2019].

Mamdani, E. H. & Assilion, S., 1975. An Experiment in

Linguistic Synthesis With a Fuzzy Logic Controller.

International Journal of Man-Machine Studies, 7, pp. 1-13.

Olindersson, F., Bruhn, W., Scheidweiler, T. & Andersson,

A., 2017. Developing a Maritime Safety Index using

Fuzzy Logics. TransNav (Vol. 11, No. 3), pp. 469-475.

Perera, L. P., Carvalho, J. P. & Soares, C. G., 2011. Fuzzy

logic based decision making system for collision

avoidance of ocean navigation under critical collision

conditions. Journal of Marine Science and Technology, pp.

84-99.

PIANC, 2014. Harbour Approach Channels and Design

Guidelines - Report No 121-2014, s.l.: The World

Association for Waterborne Transport Infrastructure.

Poschmann, P. & Burmeister, H.-C., 2017. European Maritime

Simulator Network Technical Description for STM

Validation. [Online] Available at:

https://s3-eu-west-1.amazonaws.com/stm-

stmvalidation/uploads/20190403113947/STM_ID3.2.1-

EMSN-Technical-Description.pdf

[Accessed 10 04 2009].

Rizvanolli, A., Burmeister, H.-C. & John, O., 2015. The role

of the European Maritime Simulator Network in

assessing dynamic sea traffic management principles.

TransNav (Vol. 9, No. 4), pp. 559-564.

Rothblum, A. M., n.d. Human Error and Marine Safety.

[Online] Available at: http://www.bowles-

langley.com/wp-

content/files_mf/humanerrorandmarinesafety26.pdf

[Accessed 10 04 2019].

Sanquist, T. F., 1992. Human Factors in Maritime

Applications: A New Opportunity for Multi-Modal

Transportation Research. Proceedings of the Human

Factors 36th Annual Meeting, pp. 1123-1127.

Scheidweiler, T. & Weber, R., 2018. EMSN Test Report:

Numerical Data Analysis Southern Baltic Scenario. [Online]

Available at: https://s3-eu-west-1.amazonaws. com/stm-

stmvalidation/uploads/20190403113932/ STM_ID3.3.7-

EMSN-Numerical-Data-Analysis_Southern-Baltic-

Scenario.pdf [Accessed 10 04 2019].

Serban, P. S. & Panaitescu, V. N., 2016. Comparison

Between Formulas of Maximum Ship Squat. "Mircea cel

Batran" Naval Academy Scientific Bulletin, Volume XIX -

Issue 1, pp. 105-111.

Zadeh, L. A., 1965. Fuzzy Sets. In: Information and Control.

s.l.:s.n., pp. 338-353.