555
1 INTRODUCTION
The shipping industry is responsible for a large
portion of the worlds greenhouse gas [GHG]
emissions. According to IMO the emissions increased
from 977 million tonnes in 2012, to 1076 million
tonnes in 2018 [1]. As of 2018, shipping is responsible
for 2,89% of the global anthropogenic emissions. To
handle rising GHG emissions from shipping, IMO has
implemented several measures. One important
measure is the Ship Energy Efficiency Management
Plan [SEEMP] which is mandatory for all ships above
400 GT. It was adopted as an amendment to MARPOL
Annex VI at MPEC 62 in 2011 [2]. The SEEMP is an
operational measure that fosters fuel efficiency on
board ships. It consists of goals for energy saving,
measures that should be followed and how to monitor
energy usage in daily operations. Typical measures
for energy efficiency are speed optimization, weather
routeing and efficient use of ship equipment [3]. The
SEEMP is usually developed by the crew of each
vessel and thus specified for its operation.
Furthermore, the document should be revised on
certain intervals, for example when experiencing
extensive changes to daily operations.
Most passenger ferries in Norway are required to
create a SEEMP and implement energy saving
measures. For years the ferry operators in Norway
have strived towards lowering emissions and
reducing the cost of operations. Economic rewards in
combination with strict government regulations have
fostered company policies demanding efficient ferry
operations. Tough competition between shipping
A Study of Efficiency Regarding Port Operations on
a
Passenger Ferry
E.M. Kløvning
Norwegian University of Science and Technology, Tro
ndheim, Norway
ABSTRACT: While revising their “Ship Energy Efficiency Management Plan” [SEEMP] the crew of a roll-
on/roll-off [Ro-Ro] passenger ferry in Norway discussed new measures for energy saving. One suggestion was
to lower transit speed by reducing idle dwell time in port.
The concept was to arrive just in time to handle all
cargo before departing. On short distances a couple of minutes would make a noticeable impact on transit
speed. A speed reduction leads to lower fuel consumption which has both economic and environmental
benefits. Although port operational efficiency has been studied by maritime researchers for some amount of
time, there exists a limited supply of literature dealing with passenger ferries and their cargo handling in port. It
was therefore necessary to gather more information before implementing this measure in the ferry’s SEEMP. An
observational field study was carried out by one of the navigators, where different variables related to cargo
handling in port was measured. Throughout the field study it was discovered that the ferry had idle time in
port on several occasions. Among the factors that were discussed it was recommended that transit speed should
be reduced by 0,5 knots to save fuel. The results were then summarized in this article and distributed with the
intent of sharing our experiences.
http://www.transnav.eu
the
International Journal
on Marine
Navigation
and Safety of Sea Transportation
Volume 18
Number 3
September 2024
DOI: 10.12716/1001.18.03.
09
556
companies in combination with increasing fuel prices
puts pressure on operating personnel to reduce
unnecessary expenses. On the other hand, this has
also led to an exciting development in the industry of
ro-ro passenger ferries. New ferries are built with
other energy carriers besides marine diesel oil [MDO]
and liquid natural gas [LNG], such as electric [4] and
hydrogen power [5]. Autocross and autonomous
sailing are also in the making [6], both providing
more energy efficient operations.
2 BACKGROUND
The crew of a passenger ferry in Norway were
discussing possible changes to the ships SEEMP. The
ferry operator encouraged new ideas and challenged
crew members to implement these energy saving
measures. Among several suggestions, port efficiency
was highlighted as a possible energy saving area.
Studies on container shipping has shown that more
efficient port operations have led to less fuel
consumption because of speed optimization [7]. On
this particular connection it was widely known that
the ferries had idle time during port operations. The
next section will present the ferry and the connection
in detail.
2.1 MF Glutra
The crew worked on MF Glutra as seen on figure 1,
which is a ro-ro passenger ferry that runs on LNG.
The ferry was delivered in 2000 from Langsten Yard
and later modified in 2010 at Remontowa in Poland.
MF Glutra has IMO number 9208461 and it is built as
a monohull, aft-bow symmetrical vessel with one
Schottel STP 1010 azimuth thruster in each end [8].
The ferry operated on the Molde-Vestnes connection
as part of E39 in Norway during this field study.
Figure 1. MF Glutra viewed from starboard side, arriving in
port during normal operation [8].
Figure 2.General arrangement and main dimensions of MF
Glutra [8].
MF Glutra is designed to transport maximally 120
car equivalent units [CEU] and 350 passengers as
presented in figure 2. CEU is a standard reference unit
for 4,3m long vehicles [9, p. 30] used by The
Norwegian Public Roads Administration.
Furthermore, figure 3 presents a loading condition
with 120 cars and a loading condition with 10 trailers
and 62 cars. This is meant as an illustration to how the
lanes are organized. The main deck also includes two
designated spaces for dangerous cargo in each end, as
well as evacuation zones and walkways. All large
vehicles are placed on this deck while the side house
deck is used for cars. Notice that trucks are meant to
be stowed in the two lanes close to the centreline, to
maintain stability.
Figure 3. Different loading conditions as seen from above,
where squares represent a vehicle. Side house deck is not
displayed on the bottom half [8].
2.2 Molde-Vestnes connection
This connection is the 4th largest in Norway
measured in cargo transported annually [10]. Molde-
Vestnes is part of E39 and at the time of this study is
operated by 4 ro-ro passenger ferries from Fjord1,
each with a capacity of 120 CEU. The distance from
each port is approximately 6,5 nautical miles and with
a transit speed of 12 knots the voyage should take 37
minutes. Dwell time is usually 8 minutes, but the
ferries can adjust arrival time without economic
repercussions from the government. Similar to other
connections, there are major traffic variations during
the day.
Most navigators on this connection chose to sail
with a transit speed of 12 knots to avoid delays in
port. One major reason for this is that the schedule is
quite strict making it harder to reduce possible delays
by increasing vessel speed. Furthermore, a delay on
one ferry could cause delays on the other ferries
considering that each port has infrastructure to handle
just one ferry simultaneously.
557
2.3 Dwell time and fuel efficiency
Dwell time has an impact on the other phases of
operation for a passenger ferry. However, the extent
varies depending on the ferry and the connection. Ro-
ro passenger ferries are an essential part of the
Norwegian road network with 133 active connections
[10] across a series of fjords, channels and straits. In
total, these connections transport over 34.000.000 CEU
yearly [11]. In recent years, the ferry operators have
met increasing and stricter demands concerning
emissions, punctuality, and operational reliability.
Depending on the region, authorities will demand
economic redress from shipping companies that do
not uphold these regulations.
Ferry connections in Norway are quite diverse.
Some connections are maintained by multiple ferries
while other connections require just one. Timetables,
crew size, ferry design and equipment on board
differs across the ferry fleet. Tourism or major traffic
variations also induce a seasonal increase in ferries on
specific connections.
Although there are some differences, the operation
is similar. Ferry operations are split in different
phases as shown below [12, p. 3]:
1. Ferry arrives at dock and keeps itself in place using
its propulsion equipment/mooring.
2. Hatches and doors open which let the vehicles and
passengers off the ferry.
3. Ferry personnel guides waiting vehicles and
passengers on-board the ship.
4. Hatches and doors close.
5. Ferry undocks from the current harbour and starts
transiting to the next one.
6. During transit and docking/undocking ferry
personnel takes care to follow the International
Regulations for Preventing Collisions at Sea
(COLREGS) to avoid any collision.
When sailing between ports, ferries have three
distinct phases. These are acceleration, transiting and
retardation. Depending on the connection, some
ferries also use the propulsion equipment to position
the ship along the pier during cargo handling. These
phases have a great influence on each other as
exemplified in figure 4.
Figure 4. Voyage options for MF Glutra [8].
The information in figure 4 is gathered from MF
Glutra when sailing on the Molde-Vestnes connection.
Although the vessel operated on LNG, it is natural to
use the term kWh when discussing energy efficiency.
This is because the ferry operator standardized kWh
as the unit for energy consumption in their vessel
fleet. The integrated automation system will therefore
display kWh on all ferries, for all types of energy
carrier used for propulsion. Kilowatt-hour (kWh) is a
well-established measure of energy consumption and
consists of the SI-unit Watt, multiplied with 1000
(kilo) and 1 hour.
Four different phases are shown with the
corresponding energy consumption during normal
operations. The total amount of time for one voyage is
45 minutes and this cannot be exceeded. If the ferry
accelerates to 12 knots, it will need approximately 5
minutes and 60 kWh during calm weather. When
accelerating the consumption per minute is identical
in both cases but reaching 11 knots is achieved faster.
Transiting in 12 knots would require more energy per
minute, but the destination is reached faster. Thus, the
consumption is almost identical. The retardation
phase is a little bit longer when sailing in 12 knots
because the vessel needs longer time to reduce speed.
Here the power demand is estimated to be at 500 kW
to maintain proper steering. Naturally the last phase
in quay is shorter when sailing in 11 knots, but the
total energy consumption for the whole voyage is
lower. Notice that the ferry uses its propulsion to
position itself along the pier, and that power demand
is estimated at 300 kW. Reducing dwell time while
still handling all cargo within the timetable would
therefore be a positive measure for energy efficiency.
Based on the example in figure 4, optimizing transit
speed benefits the ferry operator. However, it is
difficult to estimate the proper transit speed because
terminal time is uncertain. In essence, estimating
terminal time in port is crucial to save fuel during
other phases of operation. The reason why is further
explained in figure 5 below.
Figure 5. Combined power demand on the propulsion
equipment at different velocities [8].
Figure 5 is copied from the ferry’s SEEMP. It
shows how the power demand is affected by vessel
speed. Although vessel speed is affected by several
factors, such as wind, current, growth or loading
condition [13], the propulsion system is the main
cause for velocity. The propulsion equipment requires
power from the main engines, which in turn
consumes fuel. Depending on the engine, fuel
consumption is described as having an exponential
correlation with engine load on most ships [14].
On the other hand, reducing speed to a bare
minimum is not profitable either. Conventional
marine engines usually work most efficiently between
70-90% load [15]. The key theme here is therefore to
reduce speed if it is fuel efficient.
558
To summarize, terminal time has a great effect on
the other phases of ferry operations. In theory there
could be potential economic and environmental
benefits to be gained through careful planning of
every voyage and terminal operation. Adapting a
vessel speed that provides sufficient time for
unloading existing cargo as well as loading new cargo
would be more effective than what is practiced today
on some connections.
3 FIELD STUDY
Hoping to be able to reduce transit speed and save
fuel, it was necessary to observe cargo handling in
port. What needed to be uncovered was how much
idle time the ferry experienced and how much the
traffic varied. Loading and unloading efficiency was
also observed. The specific research questions were
formulated like this:
− How long is the dwell time?
− How long time is spent on equipment handling?
− How efficient is the unloading?
− How efficient is the loading?
− How long is the idle time?
To answer these questions a field study was
organized. An observational field study falls within
the category of descriptive quantitative research [16,
p. 154] where a phenomenon is described as it is while
interfering as little as possible. All crew members
were told to act as normal during these observations.
3.1 Data gathering
The first research question was to observe total dwell
time. This is the period between vessel movement
dedicated for cargo handling in port. In other words,
terminal time starts the moment the ferry has
positioned itself alongside the pier and reached a
speed of 0 knots. Terminal time ends when all hatches
and doors are closed, and the ferry starts embarking.
This definition is chosen because the on-board Ship
Performance Monitor (SPM) also applies this
definition. Dwell time was registered from the ship
SPM on the bridge. All navigators were told to sail
with 12 knots speed over ground during transit.
Upon arrival the navigator stood on the bridge
with a stopwatch. This was used to measure
unloading and loading of vehicles. Equipment
handling was also measured by stopwatch. This
includes handling of gates, hatches and doors when
arriving and embarking. This is a fixed value that
does not change.
To measure efficiency the crew also requested
traffic logs for the observations. These were provided
by the ferry operator. Finally, idle time had to be
observed. This was done by taking the dwell time for
each observation and subtracting the time spent
handling cargo or equipment.
When planning the field study, it was necessary to
establish how many observations were needed and a
timeframe. The SEEMP had to be finished in
November. It was therefore decided that the
observations were conducted in October and that we
aimed for as many as possible. The observations
would be done at different times of day when the
author was available.
3.2 Molde ferry terminal.
Every observation was made in Molde, as shown on
figure 6. When vehicles are unloading, they use a road
consisting of two lanes that travel unobstructed for
approximately 300 m. Here it reaches an intersection.
During heavy traffic this could lead to slower
unloading and a possible traffic jam. The two lanes
that are assigned for unloading combines into one
lane near the intersection. Unloading vehicles must
therefore merge as efficiently as possible to avoid
congestion. Vehicles waiting to embark are stowed on
the three lanes in the upper half of the figure. In the
end it should also be mentioned that pedestrians are
physically separated from vehicles using designated
walkways. These two groups are handled
simultaneously but still unaffected of each other.
Figure 6. Molde ferry terminal [17].
3.3 Unloading phase
Figure 7 presents a detailed description of unloading
in port. Shortly before arrival the visor in the stern is
lifted in upright position. When the ferry is carefully
positioned alongside the pier, the able seaman (AB)
will adjust the linkspan’s height depending on tidal
waters using a remote control. When the front of the
linkspan rests on the ferry’s stern loading shelf, the
AB will open the loading ramp. Finally the automatic
gate will open, indicating that unloading can safely
commence. AB will decide which lane is unloading
first. On MF Glutra
the ramp is quite wide and
therefore it is possible to unload two lanes
simultaneously. Pedestrians walk on a passenger
walkway, sheltered from the vehicles.
Figure 7. Overview of cargo handling operations on MF
Glutra.
Unloading operations on passenger ferries are
regulated in Norwegian law [18] and company
procedures [8]. It states that drivers must follow crew
instructions and that vehicles only move when a
signal is given. Figure 8 illustrates an unloading
operation where the AB has just finished choosing
which vehicles unload first.
559
Figure 8. Unloading vehicles in Molde ferry terminal.
There are several situations that can arise when
unloading. Some ferries have had a heeling angle so
large that it could not unload [19], others have
experienced tidal waters that restrict heavy vehicles
from unloading [20]. Some cars fail to start, some
drivers are asleep during unloading or some are not
even present in their cars. All these situations lead to
longer terminal time, although they are out of control
for the crew.
3.4 Loading phase
When the ferry is completely empty, loading starts.
This phase is also regulated in law. It states that
vehicles might only drive onboard on signal and that
they follow the queue of which they arrived on the
quay. An exemption here is if stability or other factors
makes this impractical. There are also some vehicles
with priority, such as buses and emergency services
on the job. The AB must place vehicles according to
the stowage plan. Emergency zones must be
unobstructed, and the passenger tally must be
reported to the bridge before departure.
Figure 9. Loading and stowing vehicles in Molde ferry
terminal. Designated area for dangerous goods is
thoroughly marked on the deck, using red paint.
In this phase it is only possible to load one lane at a
time as illustrated on figure 9. There are also some
scenarios that can lead to longer loading time, for
instance cars that park in the evacuation zone and
must be moved. Furthermore, there is a risk of
collision between vehicles and passengers or crew
members.
3.5 Limitations
Although this method is believed to be reliable, there
are some limitations that affect the results. One
problem is related to sea-level fluctuations caused by
tides. During high tide or low tide, some large
vehicles tend to drive slower to reduce the risk of
material damages underneath the car. To some extent,
this limiting factor could also be applicable when
there are lots of heavy trucks stowed near the
perpendiculars, providing an unwanted forward trim
during unloading. Figure 10 provides an example of
this limitation where two heavy trucks are placed in
the bow. Unfortunately, these factors have not been
accounted for in this study. A different limitation is
related to capacity problems on the port facility
during unloading. When a fully loaded ferry unloads
all vehicles at once, there is a potential risk of traffic
congestion present.
Figure 10. Example of one limitation where heavy cargo is
placed in the bow, potentially slowing down unloading
operations.
Another limitation is associated with the
complexity of port operations. This method does not
necessarily cater for the diversity of driver
characteristics, where some might accelerate faster
than others or maintain a higher speed during loading
or unloading.
These factors are hard to cater for and as a result
the observations cannot be viewed as undisputed
facts but descriptions of a trend. Although there are
some limitations to this study, they have been deemed
not severe enough to invalidate the results. The study
was therefore completed, knowing about these
limitations.
4 RESULTS
The result of the study is presented in the following
pages. It was conducted 36 observations in Molde
ferry terminal, between the 2. of October and 30th of
October 2020. There are certain gaps between the
dates, but this is due to the working schedule of the
560
author. Furthermore, the time of day is not the same
for each date. This is because the observations were
conducted by the author while performing other work
related tasks. Sometimes it was not feasible to conduct
an observation. If any inconsistencies were present,
such as a car failing to start, the observation would be
deleted from this study.
Equipment handling was measured at 45 seconds.
This includes opening or closing hatches, handling the
gate and linkspan. These seconds are already
accounted for in the idle time column in table 1 below.
Information about registered traffic during loading
and unloading on these dates was provided by the
ferry operator. The cargo for both loading and
unloading is presented using CEU, which is a
standard reference unit for vehicular cargo, as
mentioned in chapter 2. The time column show the
next scheduled departure from Molde.
Table 1 show that there are several observations
that have long periods of idle time. One exemption is
observation 36 which exceeded the timetable by 36
seconds. The most important figures from table 1 are
presented in table 2 below.
Table 1. Registered traffic during the observational study.
________________________________________________
DT – Dwell time; CU – CEU Unloaded; TU – Time unloaded
CL – CEU Loaded; TL – Time Loaded; IT – Idle time
________________________________________________
Date Time(LT+1) DT CU TU CL TL IT
________________________________________________
1 02.10.20 14:00 480 65,15 133 63,89 249 53
2 02.10.20 17:00 470 81 171 35 149 105
3 02.10.20 18:30 465 85,33 185 17,23 64 171
4 02.10.20 20:00 483 17,55 38 16,93 61 339
5 19.10.20 09:30 477 10 31 19,6 72 329
6 19.10.20 11:00 445 30,11 67 21,52 60 273
7 19.10.20 12:30 491 27,58 65 13,74 43 338
8 19.10.20 14:00 435 19,74 50 12,30 51 289
9 19.10.20 15:30 444 37,82 56 19,24 76 267
10 19.10.20 17:00 487 34,15 66 25,65 77 299
11 19.10.20 18:30 490 10,91 24 13,82 46 375
12 20.10.20 09:30 456 24,68 41 30,02 98 272
13 20.10.20 11:00 420 11,53 25 11,23 39 311
14 20.10.20 12:30 434 27,89 56 17,59 56 277
15 20.10.20 14:00 456 35,70 85 55,34 193 133
16 20.10.20 15:30 477 39,18 92 21,23 80 260
17 20.10.20 18:30 456 18,23 53 4 16 342
18 22.10.20 09:30 454 24,5 39 27,23 95 275
19 22.10.20 12:30 489 41,16 75 22,55 72 297
20 22.10.20 15:30 411 41,79 93 50,61 179 94
21 22.10.20 18:30 434 41,91 72 26,97 70 247
22 23.10.20 09:30 454 22,05 44 31,16 114 251
23 23.10.20 15:30 467 45,17 93 68,25 257 72
24 23.10.20 18:30 487 70,06 140 16,09 62 240
25 26.10.20 09:30 454 17,9 19 27,55 84 306
26 27.10.20 09:30 421 48,55 75 16,08 41 260
27 27.10.20 12:30 489 14,8 40 9,825 28 376
28 28.10.20 09:30 456 28,51 60 14,84 55 296
29 28.10.20 12:30 485 43,97 89 18,55 47 304
30 28.10.20 15:30 498 43,48 83 35,68 105 265
31 28.10.20 18:30 434 18,30 47 11,45 32 310
32 29.10.20 09:30 411 30,44 65 21,82 88 213
33 29.10.20 12:30 453 27,05 59 8,627 30 319
34 29.10.20 15:30 421 22,44 47 31,83 129 200
35 30.10.20 12:30 456 49,11 94 39,2 122 195
36 30.10.20 14:00 482 56,68 134 80,25 339 -36
________________________________________________
Table 2. Key figures.
________________________________________________
Dwell time Idle time
________________________________________________
Shortest observation 411 seconds 53 seconds
Longest observation 498 seconds 376 seconds
Average observation 458,94 seconds 247,69 seconds
Least efficient observation 23,10%
Most efficient observation 88,95%
Average efficiency 45,98%
________________________________________________
As shown in table 2, there are large differences
between the observations. Efficiency describes the idle
time as a portion of the total dwell time for each
observation. This is further illustrated in figure 11.
Figure 11. Port efficiency
Figure 11 compares idle time and dwell time for
each observation in the field study. Notice the
fluctuating results for idle time in port. Dwell time
also deviates from the goal of 8 minutes, but not
substantially.
Table 3. Key figures from the study
________________________________________________
Unloading Loading
________________________________________________
Longest observation 185 seconds 339 seconds
Shortest observation 19 seconds 16 seconds
Average observation 72,38 seconds 93,86 seconds
Most efficient observation 1,06 sec/CEU 2,53 sec/CEU
Least efficient observation 3,10 sec/CEU 4,26 sec/CEU
Average efficiency 2,10 sec/CEU 3,44 sec/CEU
________________________________________________
Table 3 highlight essential data from the loading
and unloading phases of the operation. There is a gap
between longest and shortest observation. This
indicates major traffic variations in port. The average
loading observation seems to take longer time than
unloading. This is bolstered by the efficiency
calculations, where seconds spent handling one CEU
is presented.
Figure 12. Results from observations with a corresponding
trend line for each operation.
Figure 12 exhibit cargo handling data from every
observation. Although there are some deviations,
there seems to be a somewhat linear trend line for
561
both operations. These lines visualize the general
pattern and direction of the data. As seen above, both
lines rise with an increase in CEU, albeit the loading
line has a marginally steeper growth. The trend lines
have values of R² close to 1, which indicates high level
of reliability. Lastly there are no extreme values that
differ greatly from the other data in figure 12. These
observations suggest that loading is slower and less
efficient than unloading. In addition, the observations
vary for similar amount of cargo.
Figure 13. Necessary transit speed for each
observation.
The figure above presents the necessary transit
speed that the ferry should have sailed to avoid idle
time in port. Calculations are based on information
from the ferry’s SEEMP [8]. The traffic fluctuations
lead to great differences between each voyage.
To summarize, the data gathered from the field
study seems to be sufficient to answer the research
questions. There are several interesting findings that
will be discussed in the upcoming chapter.
5 DISCUSSION
In this section the results of the observations will be
discussed. As mentioned earlier the scope of this
study was to learn more about port operations on a
ferry. The goal was to uncover measures to optimize
cargo handling in port, thus saving energy and
implementing this in the ferry’s SEEMP.
There are a number of interesting factors that
should be addressed in this study. Table 1 in the
previous chapter presented the total dwell time for
each observation. As mentioned earlier in this article,
the navigators tried to keep total dwell time at 8
minutes (480 seconds). It was assumed that this would
be difficult, considering that the ship handling is not
automated, except using autopilot during transit.
External factors such as wind, current, visibility or
ship traffic would make it difficult to arrive precisely.
It is also possible that navigators manoeuvre
differently and needs an uneven amount of time to
berth. These factors would most likely explain why
dwell time varies, although not significantly. In any
case, this information in itself is interesting because
the study was performed by different navigators.
Considering that all were told to sail in 12 knots, their
dwell times differ. A recommendation would be to
learn from each other and collectively become adept
at ship handling. Navigators should also prioritize
consistency to ensure positive results over time.
The next factor is related to cargo handling.
Apparently the traffic varies tremendously, both for
unloading and loading. As presented in table 3, there
is a gap between longest and shortest observation for
both situations. Most observations were made for 55
CEU or less, and no observations include an empty
deck or a fully loaded ferry. Unfortunately, this makes
it very hard to predict necessary dwell time in port. A
speed reduction would only be applicable if total
dwell time is sufficient to complete cargo handling.
Delays would lead to less disposable time for the
upcoming voyage. In most cases it would be
necessary to increase transit speed, thus diminishing
any fuel savings from the previous voyage.
Furthermore, it is important to discuss the
unloading and loading efficiency. Table 3 presents
how many seconds is spent handling one CEU. Again,
there is a significant difference amidst the most
efficient and least efficient observation. As mentioned
above, this study focused on observing cargo
handling as it was usually performed. For this
purpose, the only registered variables were time and
amount of cargo. Placement of vehicles on the deck,
how many lanes were used and how close vehicles
were stowed was all decided by the deck department.
Identifying the cause might therefor be difficult, but it
is not unlikely that the deck department could
improve their consistency and learn from each other.
Another cause might be that drivers have different
driving styles. Drivers that accelerate quickly and
maintain a relatively high speed would benefit overall
port efficiency. Cargo handling could also be affected
by meteorological and oceanographic factors. Sea
level-fluctuations create a heeling angle on the
linkspan when loading and unloading cargo. Drivers
tend to slow down when the linkspan has a steep
angle. Efficiency could be further decreased in poor
visibility, such as heavy rain, darkness and fog.
Finally, strong winds in combination with waves
create a rolling motion along the ferry’s longitudinal
axis and a pitching motion along the transverse axis.
Cargo handling in these conditions is often less
efficient, but it was not included in this study and
therefor hard to define precisely.
A major part of the observations shows that
unloading is faster than loading. This is logical,
considering that the ferry can unload two lanes
simultaneously while loading just one lane. During
loading, vehicles must also be sorted out and placed
on designated spots, while unloading do not have to
cater for this. On the other hand, traffic congestion on
the road network in Molde often reduce unloading
efficiency.
The previously mentioned topics are interesting,
but the most important factor in this study is related
to idle time. The last column in table 1 shows the idle
time for each observation. It is very interesting,
because all but one observation shows relatively large
amounts of idle time. The average efficiency is 45,98%,
but the range is from 23,10% to 88,95%, as presented
in table 2. At first glance this indicates possibilities for
fuel saving. However, cargo handling is not the only
operations conducted in port. In addition the crew use
this time to dispose of garbage, maintain cargo
handling equipment, change crew or receive
packages. Navigators should therefore not focus
solemnly on cargo handling when planning to reduce
idle time in port.
562
What this study fails to highlight is the passenger
perspective. Reducing transit speed to save fuel
would mainly benefit the ferry operator. Even though
the ferry does not correspond with other public
transportation, it is not unlikely that passengers
would have a negative reaction towards increased
travel time. Tickets also cost the same, no matter how
fast the ferry sails. Navigators should therefore
determine what is most important, predictability for
passengers or fuel savings for the ferry operator.
There is a noticeable difference between 43 minutes
and 36 minutes voyage time if you travel with a ferry.
Throughout this discussion, several topics have
been commented upon. It has been established that
the ferry has idle time in port when sailing in 12 knots
during transit. Considering that the traffic fluctuates it
is not realistic to have zero idle time in port. Focus
should therefore be given to uncover appropriate idle
time, without experiencing delays.
Figure 13 showed the necessary transit speed that
should be sailed to avoid idle time for each voyage.
For the majority of the observations, it would not be
necessary to sail with 12 knots. On another hand, it is
hard to identify one fixed speed for all voyages. A
recommendation would be to reduce the usual speed
but adjust it if needed. If transit speed was reduced to
11,5 knots, this would reduce dwell time to 6,5
minutes and the total consumption would be 446,83
kWh. These numbers are based on data from the
SEEMP and not sea trails. 33 of 36 observations would
still have idle time, but the fuel savings would be
noticeable, estimated at 3,8% reduction from 12 knots.
Most likely the passengers would not react on such a
small speed adjustment. Furthermore, the deck
department should investigate why efficiency for both
loading and unloading varies. If they find ways of
optimizing their task, maybe port operations could be
further lowered. This goes for the navigators as well.
One way to further optimize operations would be to
reduce the range in dwell time. There might be ways
to share experiences and find the best possible route
between ports.
It should also be discussed how applicable this
information is to other ferries. Most likely, some of
this is relevant to others. However, connections with
fixed arrival and departures would not benefit from
better port efficiency. This also applies to connections
with bus correspondence at certain times or electric
ferries that need to recharge batteries while in port.
Most ferries that do not fall into aforementioned
categories would probably benefit from this study.
6 CONCLUSION
The main objective of this study has been to learn
more about cargo handling in port on a passenger
ferry. As mentioned earlier in this article, efficient
port operations lead to a lower necessary transit
speed. A speed reduction has both an economic profit
as well as an environmental gain. To achieve more
efficient port operations, a series of observations were
conducted on a passenger ferry in Molde.
The field study resulted in some interesting
findings. It was discovered that the dwell time varies,
even though navigators were told to sail with 12 knots
speed over ground. This indicates a possibility for
improving overall efficiency. To some extent, this is
also applicable to the deck department. Cargo
handling efficiency varies, but it is uncertain if this is
related to the actions of the deck department or
external factors. This should be studied further but a
recommendation would be that the crew share
experiences and focus on consistency in their work-
related tasks. Better manoeuvring and faster cargo
handling would bolster overall efficiency and
generate room for reduced transit speed.
Furthermore, it was uncovered that there are large
traffic fluctuations on this connection. This makes it
hard for crew to predict necessary dwell time for each
voyage. As a result of this, navigators have usually
sailed with 12 knots speed over ground to avoid
delays in port. The observations indicate that this
speed is too excessive. A recommendation would be
to sail in 11,5 knots during transit, but increase speed
if necessary, during periods of high traffic. This would
lead to increased fuel efficiency and should be
included in the ferry’s SEEMP.
ACKNOWLEDGMENTS
This work was supported by the Norwegian ferry operator
Fjord1. They have provided several internal documents and
company procedures, not to mention authorizing and
approving the field study.
BIBLIOGRAPHY
[1] International Maritime Organization, “Fourth IMO GHG
Study 2020,” IMO, London, United Kingdom. Published
2021 [Online]. Available: https://www.imo.org/en/
OurWork/Environment/Pages/IMO-Publications.aspx
[2] International Maritime Organization, Energy Efficiency
Measures. Available: https://www.imo.org/en/ourwork/
environment/pages/technical-and-operational-
measures.aspx Accessed on: 14.02.23.
[3] DNV GL, Ship Energy Efficiency Management Plan
(SEEMP). Available: https://www.dnv.com/maritime/
hub/decarbonize-shipping/key-drivers/regulations/imo-
regulations/seemp.html Accessed on: 10.02.23.
[4] Skipsrevyen, World’s first battery driven car ferry.
Available: https://www.skipsrevyen.no/aktuelt-battery-
driven-ferry-fjellstrand/worlds-first-battery-driven-car-
ferry/670542 Accessed on: 20.01.23.
[5] Riviera, World’s first hydrogen ferry set for sea trials.
Available: https://www.rivieramm.com/news-content-
hub/news-content-hub/worlds-first-hydrogen-ferry-set-
for-sea-trials-73501 Accessed on: 25.01.23.
[6] Kongsberg, The pioneer trail. Available: https://www.
kongsberg.com/maritime/about-us/news-and-media/
our-stories/the-pioneer-trail/ Accessed on: 16.02.23.
[7] Golias, M.M., Saharidis, G.K., Boile, M., Theofanis, S. and
Ierapetritou, M.G. (2009) The berth allocation problem:
Optimizing vessel arrival time, Maritime Economics &
Logistics, 11(4), pp. 358-377. 10.1057/mel.2009.12
[8] Fjord1, Interne dokumenter [PDF]. Personal
communication, 2 Oct. 2020.
[9] Statens Vegvesen, Kapittel 1-8 Brutto Bognes-Skarberget
og Drag-Kjøpsvik (versjon 3). Available: https://eu.eu-
supply.com/app/rfq/publicpurchase_docs.asp?PID=2784
74&LID=327084&AllowPrint=1 Accessed on: 15.11.20.
[10] Statens Vegvesen, Ferjesamband. Available: https://
ferjedatabanken.no/Samband Accessed on: 17.01.23.
[11] Statens Vegvesen, Ferjestatistikk. Available: https://
ferjedatabanken.no/Statistikk Accessed on: 12.03.23.
563
[12] Bitar, G.I. (2017) Towards the development of
autonomous ferries. Master thesis, Norwegian
University of Science and Technology. Available:
https://ntnuopen. ntnu.no/ntnu-
xmlui/handle/11250/2465617 Accessed on: 20.02.23.
[13] Bialystocki, N. and Konovessis, D. (2016) On the
estimation of ship’s fuel consumption and speed curve:
A statistical approach, Journal of Ocean Engineering and
Science, 1(2), pp.157-166. 10.1016/j.joes.2016.02.001
[14] Psaraftis, H.N. (2019) Speed optimization vs speed
reduction: The choice between speed limits and a bunker
levy, Maritime Economics & Logistics, 21, pp.524-542.
10.1057/s41278-019-00132-8
[15] Solem, S., Fagerholt, K., Erikstad, S.O. and Patricksson,
Ø. (2015) Optimization of diesel electric machinery
system configuration in conceptual ship design, Journal
of Marine Science and Technology, 20, pp.406-416.
10.1007/s00773-015-0307-4
[16] Leedy, P.D. and Ormrod, J.E. (2016) Practical research:
Planning and design. 11th edn. Essex: Pearson Education
Limited.
[17] Google Maps, Molde ferry terminal. Available:
https://www.google.no/maps/@62.7367427,7.1722391,315
m/data=!3m1!1e3 Accessed on: 18.10.20.
[18] Forskrift om transport med ferje. Forskrift 26. march
2003 nr. 403 om transport med ferje. Available:
https://lovdata.no/dokument/SF/forskrift/2003-03-26-403
Accessed on: 20.03.23.
[19] Killingberg, A. (2018) Fikk ikke bilene i land da ferja var
skjevt lastet, Fosna-Folket.no. Available:
https://www.fosna-folket.no/nyheter/2018/04/04/Fikk-
ikke-bilene-i-land-da-ferja-var-skjevt-lastet-16413019.ece
Accessed on: 10.02.23.
[20] Nyhaug, E. (2016) Bastø Fosen fikk trøbbel med å få
bilene på land, Gjengangeren.no. Available:
https://www.gjengangeren.no/ferjer/basto-
fosen/horten/basto-fosen-fikk-trobbel-med-a-fa-bilene-
pa-land/s/5-60-83902 Accessed on: 11.02.23.
