International Journal

on Marine Navigation

and Safety of Sea Transportation

Volume 1

Number 2

June 2007

145

Mental Workload of the VTS Operators by

Utilising Heart Rate

S. Kum

Kobe University, Graduate School of Science and Technology, Kobe, Japan

M. Furusho

Kobe University, Graduate School of Maritime Sciences, Kobe, Japan

O. Duru & T. Satir

Istanbul Technical University, Istanbul, Turkey

ABSTRACT: This study clarifies the mental workload of VTS Operator; by understanding their

characteristics during carrying out their task, with a physiological index. The objective is to determine VTS

Operators’ mind stress that might trigger any human error based on their mental workload during their

watches. For this purpose, Heart Rate Monitor (HRM) is utilized as physiological index. The VTS Operators

fitted the HRM and all of them have experience as a Master Mariner. During the all experiments, their heart

rates and behaviours were recorded on the Event Record Form based on the time scale. After getting the heart

rate variability, it is matched with the events, and then Operators’ behaviour is understood as the mental

workload due to such kinds of events. Furthermore, these workloads include the Operators’ mind stress and

their decisions under these circumstances. This study provides the fundamental information for understanding

the VTS Operators’ characteristics.

1 INTRODUCTION

This study is one of authors’ systematic studies for

understanding the behaviours of seafarers. And, the

investigation of heart rates is found a suitable

physiological index for evaluating mental tasks.

1.1 Heart Rates and Heart Rate Variability (HRV)

A healthy person’s heart will pump between 60 and

140 times per one minute, depending on whether the

individual is under exertion or at rest. Beating of the

heart is caused by rhythmic contraction of the atria

and ventricles under the control of Autonomic

Nervous System, ANS, (Malik 1996). ANS has two

components; Sympathetic Nervous System (SNS)

and Parasympathetic Nervous System (PNS).

Heart rate is a term used to describe the frequency

of the cardiac cycle. It is calculated as the number of

contractions (heart beats) of the heart in one minute

and expressed as "beats per minute (bpm)”.

Heart Rate Variability (HRV) is the variation in

times between successive heartbeats (i.e. R-R

intervals that is a term from the peak of R wave to

the next peak in microseconds). R-R interval always

fluctuates to reflect the physical and the mental

conditions. The frequency components of R-R

interval reflect Low Frequency, LF, (0.04 - 0.15 Hz)

and High Frequency, HF, (0.15 - 0.40 Hz). This

index is able to evaluate both SNS and PNS activity

at the same time. The LF provides a quantitative

index of the sympathetic and parasympathetic

activities controlling the heart rate, while HF

provides an index of the parasympathetic tone

(Ishibashi & Yasukouchi 1999).

The relationship between R-R interval and heart

rates (bpm) is when the interval time of R peaks

increases conversely heart beats decrease in minute.

It means that the fast heart pulsates, the smallest R-

R intervals. HRV analysis is based on measuring

variability in heart rate; specifically, variability in

intervals between R-R. Power spectral analysis

146

(obtaining HF and LF) on HRV is useful to assess

the mental workload (Sloan et al. 1995).

HF relates to parasympathetic activity (such as

controlled respiration, cold stimulation of the face,

and rotational stimuli), and LF relates to both

parasympathetic and sympathetic activities such as

tilt, standing, mental stress, and moderate exercise

(Malik 1996). Some studies suggest that LF is a

quantitative marker of sympathetic modulations;

other studies view LF as reflecting both sympathetic

activity and parasympathetic activity. Consequently,

the ratio of LF/HF is considered to reflect the

sympathetic modulations. That’s why this ratio can

be considered to the SNS index for evaluating

mental workload as defined by Kobayashi & Senda

(1998).

Before carrying out experimental studies, authors

carried out pre-study for recording heart rates of

VTS Operators (VTS-Os) while sitting operation

desk (console table) when they are doing nothing

and when they arrange display of the operating

system before taking the watch. These two

conditions compared and the results can be

summarized as; two samples (resting and arranging

display at the console table) compared with ANOVA

analysis in heart beats and also R-R intervals. The

analysis results show that two samples data are

different (F statistics is significant at the confidential

level, p < 0.05). The physical actions of operator are

at the minimum level due to always sitting. It can be

assumed that this difference mainly occurs for the

effort of performing task as mental behaviours. This

information is useful for considering the results that

would be mainly affected by the fluctuation of heart

rates caused by sympathetic activities as mental

workload (until the respirations doesn’t influence for

the contra verse situation such as instantaneous

emotion).

The relation between HR and HRV is found from

the pre-study as; recording R-R intervals and

predicted R-R intervals (that estimated by heart

beats, using by average values of R-R for every beat

times) are the same pattern, (the correlation is 0.99

significantly (p < 0.01)). And also, two samples

belong to same population (F (0.0005), p: 0.98 >

0.05).

There is very few study related to HR in the

maritime field, and they mainly consider to the

navigators’ heart rate variability during on board and

in the simulator experiments. These studies (Kum et

all. 2004, Kobayashi & Senda 1998, Murai et all.

2004a and Murai et all. 2004b) obviously show that

there were remarkable effects for the navigators’

heart rates during the task performing. On the other

hand, there is not found such kind of study for the

VTS-Os. That’s why heart rate variability has

enthusiastically been employed in physiological

assessment of respective working environmental

factors during the executing task process of VTS-Os

in this paper.

1.2 Mental Workload – MWL

Workload is defined as the physical and/or mental

requirements associated with a task or combination

of tasks (Gudipati & Pennathur, unpubl.). Workload

refers to the operators have limited capacity that

actually required to perform a particular task; it is

the interaction between the operator and assigned

task (Gopher & Donchin 1986).

Workload is classified as Physical Workload

(PWL) and Mental Workload (MWL). PWL is the

measurable portion of physical resources expended

when performing a given task (Gudipati &

Pennathur, unpubl.). MWL is defined as the level of

processing capacity while performing the task or the

difference between the capacity to affect the usable

real performance and human-information processing

system (Eggemeier & Wilson 1991). The common

idea for the MWL is directly proportional with the

difference between existing sources and required

sources by the task.

Under these explanations, we can consider three

kinds of loading for performing a task. Firstly, the

operator capacity is over to the required

performance. This situation tends to the operator

bored and then also tends to make mistakes.

Secondly, the capacity is equal to the required

performance that is the best fitting of the

employment not only for human resources but also

for the efficient performance of the task. Finally,

when the operator capacity is not enough to perform

a task due to overlapping task items; this situation

tends to the operator has stress. And, if the last

situation ordinarily continues, occupational stress

will occur automatically and it makes the operator

has not satisfied his/her job. That’s why; the

acceptable workload can be determined as the level

of workload not to impede the operator to manage

the system safely and effectively (Jung & Jung

2001).

The concept of mental workload has become an

important issue for all kinds of industry after 1960’s.

The main reason of that is the computer and it

becomes indispensable component of the life.

Therefore, there is a rapid increase for the publishing

of papers related to the MWL. On the other hand, the

quantity of research related to MWL in maritime

field is so restricted. Authors hope that this study

could take an interest to consider not only shore side

staff in maritime sector but also for personnel on

board and in shipping companies. It is assumed that

147

the factors to cause MWL and its affects can be

analyzed by the common measures (such as NASA

TLX, SWAT and so on) for more understanding of

maritime human factors. Because when we

investigate the studies in the maritime field,

regretfully saying that there is not any study found

related to measurement to MWL in these common

measures such as utilized in the air and land

navigation field, just a few studies consider to MWL

of navigators by only using the heart rate variability

as mentioned by section 1.1. Authors also submitted

a questionnaire to the VTS-Os for determining the

MWL factors and applied NASA-TLX for measure

their MWL as the forward stage of this study.

In this study, the mental task of VTS-O is defined

as shown in Figure 1. VTS-O obtains cognitive

information from the VTS System and performs the

task under his skill based on the regulations. That’s

why the “mental workload” is the difference

between human information process and the

operator’s capacity which affects to the actual

performances that can be expended by operator (for

covering the required performance). Under this

excessive mental workload, operators may exhibit

delayed information processing, or even not respond

at all to incoming information, because the amount

of information surpasses their capacity to process it

(Ryu & Myung 2005).

It is mentioned in all studies (the common sense)

of the air and land navigation (except of maritime

navigation due to not found any related study) that

the factors to cause MWL is not clarified. On the

other hand, some studies show that nature of work,

training and age (Duru et al. 2005), physical

conditions of working environment (Leplat 1993),

the structure of task such as task aim, performing in

the most proper way and operator’s task perception

(Hart & Staveland 1988) have affect on MWL.

According to ISO 10075:1991 (Ergonomic

principles related to mental workload), the

fundamental factors of MWL are; task, work

equipment, physical work environment and social

work environment.

Fig. 1. Mental task of VTS-O

2 EXPERIMENTAL STUDY

Turkish Straits VTS (TSVTS) consists of Istanbul

VTS and Çanakkale VTS by covering area of the

Istanbul Strait, the Marmara Sea and Çanakkale

Strait (total length of the area is 164 nm). This

experimental study is carried out at the Istanbul VTS

Center. Istanbul VTS has 4 sectors; (North to South)

Sector Türkeli is the north entrance of the Istanbul

Strait (between the northern limit of the Istanbul

VTS and the line joining Fil-Çali Point), Sector

Kavak covers the area between the line of Fil-Çali

Point and the Fatih Sultan Mehmet Bridge, Sector

Kandilli’s southern limit is the line of Haydarpasa

Breakwater and Sector Kadiköy is the southern limit

of the Istanbul VTS.

TSVTS is operated in 24 hours; there are two

watching system as day shift and night shift. The

shifts have 2 hours watches. There are; two VTS-Os

(while one of them operates the system the other is

standby), one Assistant Supervisor and one VTS

Supervisor, VTS-S, (who is an appropriately

qualified VTS-O carrying out supervisory duties on

behalf of the VTS Authority) in every watch. Additi-

onally, some of the VTS-S (who engaged for daily

administrative works other than interaction with the

vessels), Data Input Operator (who engaged to

input the vessels sailing plans to the system) and

other staffs are involved day shifts.

The number of VTS-Os is 48 (including VTS-S)

at the Istanbul VTS and 30 operators at the

Çanakkale VTS. All of the Turkish VTS-Os has sea

experience (their average sea experience is 13.6

years) as a Ship Master (5.5 years is the average

level of their experience as actual Ship Master).

Their average age is 33 years and all of them have

Bachelor’s degree from a maritime faculty.

2.1 The Profile of VTS-Os

There are 48 persons employed for the Vessel

Traffic Services at the Istanbul VTS Center, 32 of

them are actually engaged for operating the system

to communicate with the ships [two operators for

every sectors at the each (A, B, C and D) shift].

Eight of 32 VTS-Os have fitted HRM under

different conditions for experimental studies. There

are not any specific criteria to choose these VTS-Os,

they were chosen randomly. But authors mainly want

to investigate different environmental situation in the

VTS. That’s why; four of experimental study was

particularly carried out simultaneously for the four

sectors. It was quite difficult to carry such kind of

experiment during the actual task execution not only

for the authors but also for the VTS Authority for

giving any permission. In this concept, during the

148

permission time to carry out these experiments,

authors made as much as experiments for

understanding of the harmonized environments.

And, finally 8 VTS-Os were analysed from total of

32 VTS-Os by the ratio of 25%.

Table 1 shows the profile of these operators.

VTS-Os have 6.8 years experience as Ship Master

and 2.5 years as VTS-O (in average). Majority of

operators doesn’t smoke and they are healthy. Their

average weight is 91.5 kg, average height is 178.3

cm and average age is 41 years.

Table 1. Profile of VTS-Os

2.2 Conditions of Experimental Study

The experimental studies are carried out in different

conditions such as; the flow of traffic direction,

watch keeping times, weather conditions and sectors

in VTS area. Authors carried out totally 8

experimental studies at the Istanbul VTS Center as

shown in Table 2. Four of them are carried out

simultaneously for determining the interactions

among the Operators by covering all sectors in the

VTS area. The others are carried out randomly.

Normally, the traffic situation in the Istanbul

Strait is two ways traffic based on the Traffic

Separation Schemes. But at that time, the flow of

traffic was just one way due to the operation that

carried at the southern entrance of the Strait for

underground railway system. That’s why during our

experimental study, the traffic in the Strait was just

one way, it is important to indicate this information

due to considering workload of VTS-Os’. Their

workload level was under the normal case because

the capacity of the system was not routine. As the

mentioned by VTS-Os, they don’t feel that it is

really as difficulty as comparing the normal traffic

situation in the Strait, so it is observed that they are

in quiet behaviours (especially VTS-O in the middle

sectors for just observing to passage of one way

traffic). One of the experiment was carried out when

the traffic direction was changed due to estimating

the VTS-O’s MWL in more congested traffic cases,

and also authors mainly focused the south/north

entrance of the Strait depends on the traffic

direction. The environmental conditions in the Strait

are fine.

Table 2. Outline of the experimental studies

* Traffic direction was change to South to North.

2.3 Recording Data

The Heart Rate Monitor (HRM) which is equipment

that consists of a wrist watch, a chest belt (strap) and

software produced by POLAR Co. Ltd. is utilised in

this study. The type of this model is S 810i which

can measure not only heart beats (bpm based on time

sequence as 5 sec., 15 sec. and so on) but also R-R

intervals.

Before executing the experimental studies,

authors needed to prepare equipments (such as

HRM, voice data recorders and camera) and

documents (such as Information Data Sheet for

recording ships’ data (e.g. ship’s name, type, length,

cargo, pilot on board or not and so on) that

interrelated to VTS-Os’ communication and Event

Record Form for recording their all behaviours and

actions as shown in Figure 2) for reliable keeping

records.

Fig. 2. Event Record Form for VTS-Os

2.4 Analysing Data and MWL Assessment

In the analysis, the software (Polar Precision

Performance SW Version 4.03.044) is used for

getting the characteristics of heart rate and frequency

components. And for the further analysis, the

149

statistical analysis techniques (to determine the

significances of differences and similarities), power

spectrum analysis and Maximum Entropy Method

(obtaining LF, HF and the ratio of LF/HF) are also

investigated.

Heart beats (recorded as bpm for every 5 sec.) of

VTS-Os are the raw data, then R-R intervals are

predicted from heart beats by using the average

value of R-R intervals for every bpm (Figure 3

shows the R-R intervals of VTS-O_7). Later, the

components of R-R intervals (LF and HF) are

calculated, and then LF/HF ratio is obtained.

VTS-O_7

700

750

800

850

900

950

23:56 00:01 00:06 00:11 00:16 00:21 00:26 00:31 00:36 00:41 00:46 00:51 00:56

Real Time (min:sec)

R-R Interval (msec)

Fig. 3. R-R intervals of VTS-O_7

When authors checked the direct relation (by the

regression analysis) among heart beats, R-R interval

and LF/HF ratio, it is found that the coefficient of

correlation is 0.68 (p < 0.01).

The mental workload assessment is performed as

follows (Fig. 4);

− Individual assessment for every VTS-O; firstly to

consider on fluctuation of heart beats and then the

ratio of LF/HF.

− To consider the rise and declines in the heart

rates and to check the events for this fluctuation

simultaneously and also to check the ships’

specifications in these events.

− To consider the remarkable events and their

effects on the heart rate.

− To consider the remarkable vessels and their

effects on the heart rate.

− General assessment (including individual and

mutual assessment) for getting common and

sharing behaviours; to consider the sharing

behaviour for VTS-Os in the different sectors

based on the events and/or the ships’

specifications.

− Assessment based on the sectors; to consider the

any relation between MWL of VTS-O and the

sector that he operates, e.g. there is not strong

effect for MWL of VTS-O who operates the

middle sectors due to traffic direction was one-

way and mainly they focused to observe the

passage and just communicated for position

reports and so on.

− Finally, the assessment made for any difference

because of the different watch times,

environment, weather and sea condition that may

affects to MWL of VTS-O.

Fig. 4. Procedure for mental workload assessment

3 RESULTS AND CONSIDERATION

As indicated by the summary of VTS-Os’ heart rates

in Table 3, the average heart rate is 79 bpm (standard

deviation is 5 bpm) is slightly high, because of they

are ready to have any task, and every time they keep

their attention.

Table 3. Summary of VTS-Os’ heart rates

When comparing the means of heart beats for

four sectors during the simultaneous experimental

study, it is determined that VTS-Os have different

behaviour significantly (p < 0.01) as shown in Fig-

ure 5.

General assessment made for getting the

interaction affects and sharing behaviours of VTS-

Os. The remarkable points are as follows:

Fig. 5. Changes in HR at the 4 sectors

150

− At the beginning of watches their MWL was

highest (2-3 minutes of the watches), and they

mainly arranged the display of monitors,

observing the existing traffic during this time.

And also, it is obtained that when they made

observing the traffic, their MWL was higher and

continuous during any time of the watch.

− It is determined that 90 bpm of heart beat is the

baseline for remarkable MWL. When heart beats

are over 90 bpm, the increase in the MWL is

obtained.

− There is not found any statistical relation between

physical condition of VTS-Os and fluctuations of

heart rates. As an example; the result of Chi-

Square test between smokers and non-smokers of

VTS-Os shows there is not any difference

between these two groups based on their heart

rate and heart rate variability (p > 0.05). Indirectly

saying that the smoking situation of VTS-Os

doesn’t affect to their MWL.

− The ships over 150 mt. affected to increase

MWL. Also, tankers, gas carriers and the ships

carry dangerous material slightly affected to

increase MWL. In generally, ship’s length and

type had close relations, but some cases when the

ship’s length was less than 100 mt., mental

workloads were affected by ship’s type.

− The ships which took pilot increase their MWL.

Ship’s size and taking a pilot had also close

relation, but similar case was found as above.

− When the other operator asked questions to not

any affect on MWL, but when the operator asked

questions to the others his MWL was increased.

− When VTS-Os spoke the individual items with

the other operators/persons, their MWL increased

temporary (comparing the situations related to

operation) and calmed down again.

− When they gave “advice” and “instruction” to the

ships, MWL increased but that was not the

highest workload (the sample of such kinds of

case were not enough, it occurred 3 times during

the time of all experimental studies).

− There is not strong effect for MWL of VTS-Os

who operate the middle sectors due to traffic

direction was one-way and mainly they focused to

observe the passage and mainly communicated

for position reports and so on.

− Cut down of electric power or some system

technical problems did not affect their MWL.

− There are not found any remarkable effect due to

the difference of environment, during the

experimental studies weather and sea conditions,

current were fine. And visibility is clear in the

Strait.

− The difference of MWL for VTS-Os depends on

the sectors which they operate, traffic density

(especially changing the traffic direction mainly

affects to increase the traffic volume and as a

result of this, the communication is increased

until the stable navigation is performed for

covering the ships would enter the Strait).

The figures of VTS-Os’ heart rates clearly show

that there is a rapid decline of heart fluctuations

when the beginning of mental workload, and it

instantaneous increases during the task execution

then slightly keeps the condition (depends on the

performing time), and finally repeats the rapid

decline.

4 CONCLUSIONS

The specifications of ships such as length, speed and

draft (when they are in high values that mean to

reach the limitations of the passing area (e.g. some

limitations are applied for the ships with a length

over all of 150 m. or upwards in the Istanbul Strait)

and the traffic volume have an effect on the mental

workload of VTS-Os. On the other hand, one of the

remarkable points is the vessel that took pilot affect

to VTS-Os’ mental workload.

The factors to cause MWL of VTS-Os can be

summarised as follows; ships’ specifications, the

area where ship is sailing (sailing area), traffic

density, either ship takes a pilot or not,

communication skills of Ship Master, and work

fatigue, tiredness.

In conclusion, this study provides the

fundamental information for understanding the VTS

Operators’ characteristics. Our future work is to

analyse more situations in the different VTS areas

(e.g. the Japanese VTS-Os (especially who don’t

have any maritime background)); not only utilised by

HR as physiological assessment but also utilised by

subjective assessment measures such as NASA-

TLX, SWAT, etc. for clarifying the relation of these

two assessment techniques.

ACKNOWLEDGEMENTS

This study was supported in part by a Grant-in-Aid

for Scientific Research from the Japanese Ministry

of Education, Science, Sports and Culture

(No.18402006). The authors gratefully acknowledge

to the VTS Operators at the Istanbul VTS Center and

the VTS Authority, General Manager of Coastal

Safety and Salvage Administrations.

151

REFERENCES

Duru A., Ermis A., Akay D. & Kurt M., 2005. Measurement of

Mental Workload in Information Technology Sector Using

Subjective Technique (NASA-TLX) (in Turkish),

Teknoloji, vol. 8 (2), pp. 173-180.

Eggemeier F.T. & Wilson G.F., 1991. Subjective and

performance-based assessment of workload in multitask

environments, Damos D.L. (ed.), Multiple Task Perfor-

mance, pp. 217-278, London.

Ekman P., Levenson R.W. & Friesen W.V., 1983. Autonomic

nervous system activity distinguishes among emotions,

Science, vol. 221, pp. 1208-1210.

Gopher D. & Donchin E., 1986. Workload-An examination of

the concept. Handbook of perception and Human

Performance, Cognitive Processes and Performance, vol. 2,

chp. 41, Boff K.R., Kaufman L. & Thomas J. (ed.), John

Wiley, New York.

Gudipati S. & Pennathur A. (unpubl.) Workload Assessment

Techniques for Job Design Department of Industrial

Engineering University of Texas at El Paso.

Hart S.G. & Wickens C.D., 1990. Workload Assessment and

Prediction, Harold R. Booher (ed.), An integrated approach

to systems integration, pp. 257-296, New-York.

Ishibashi K. & Yasukouchi A., 1999. Analysis of Heart Rate

Variability during Mental Task with Reference to Ambient

Temperature, Applied Human Science, vol. 18 (6), pp. 219-

223.

Jung H.S. & Jung H.S., 2001. Establishment of overall

workload assessment technique for various tasks and

workplaces, International Journal of Industrial Ergonomics,

vol. 28, pp. 341-353.

Kobayashi H. & Senda S., 1998. A Study on the Measurement

of Human Mental Workload in Ship Handling Using SNS

Value, Journal of JIN, vol.98, pp.247-255.

Kum S., Satır T., Furusho M. & Hayashi Y., 2004. Evaluation

of BTM Effects with Heart Rate, International Maritime

Lecturers’ Association’s 13th Conference, ‘Maritime

Education and Training’, St. Petersburg, 13-17 September.

Leplat J., 1993. Person-Centred Ergonomics: A Brantonian

View of Human Factors, Oborne, D.J., Branton, R., Leal,

F., Shipley, P. & Stewart, T. (ed.), International Journal of

Man-Machine Studies, vol. 39 (3), pp. 521-528.

Malik M., 1996. Heart Rate Variability, American Heart Asso-

ciation, Circulation, vol. 93 (5), pp.1043-1065.

Murai K., Hayashi Y. & Inokuchi S., 2004. A Basic Study on

Teammates' Mental Workload among Ship's Bridge Team,

IEICE Transactions on Information and Systems, Vol. E87-

D (6), pp. 1477-1483.

Murai K., Hayashi Y., Nagata N. & Inokuchi S., 2004. The

Mental Workload of a Ship's Navigator using Heart Rate

Variability, Interactive Technology & Smart Education, vol.

1 (2), pp. 127-133.

Ryu K. & Myung R., 2005. Evaluation of mental workload with

a combined measure based on physiological indices during

a dual task of tracking and mental arithmetic, International

Journal of Industrial Ergonomics, vol. 35, pp. 991–1009.

Sayers B.M., 1973. Analysis of Heart Rate Variability,

Ergonomics, vol.16 (1), pp.17-32.

Sinha R., Lovallo W.R. & Parsons O.A., 1992. Cardiovascular

differentiation of emotions. Psychosomatic Medicine, vol.

54, pp. 422-435.

Sloan R.P. et al., 1995. Consistency of heart rate and

sympathovagal reactivity across different autonomic

contexts, Psychphysiology, pp. 452-459.

Smith K. & Dickhaut J., 2005. Economics and Emotion:

Institutions Matter, Games and Economic Behavior, Special

Issue on Neuroeconomics, vol. 52 (2), pp. 316-335.