Smart Ships and implications in logistics chains a case study in Zeeuws Vlaanderen

Autonomous shipping is expected to be gradually adopted in the coming years. While many scientific studies in the field have focused on technological development, recent research has started to explore the effects of this innovation on the cargo transportation industry. This study investigates the economic dynamics that can drive logistics entrepreneurs to adopt teleoperated barges, a specific type of smart shipping. By conducting a case study of cargo transportation between two companies using roundtrips with trucks and barges, the study evaluates a modal shift to intermodal transport and looks into the conditions that are affected if barge teleoperation is implemented. A major conclusion of the study is that the transport distance, the equipment size, and the mix of captain-only tasks versus all the sets of crew tasks affect the expected economic gains that are obtained as a result of implementing smart shipping. In this case, a conventional modal shift to waterborne transport is already economically attractive. When opting to operate a smaller barge, teleoperation becomes preferable when a shore-control captain can only focus on exclusive sailing tasks and when more than one ship is monitored simultaneously.


Introduction
As we step into a future of digital transformation, there seems to be a consensus among scholars on the benefits that will come as a result of the introduction of autonomous ships.According to [1,2,3] it is expected that smart ships will help in decreasing crew costs, generate fuel savings due to the improved aerodynamics (e.g. by removing the bridge), reduce investment costs (as life support systems are no longer needed), increase crew and cargo safety due to reduced human error, increase cargo holding space, and enable flexibility and optimization opportunities for logistics chains if new business models are introduced.
Because the research focus has been primarily on the technology development domain, more studies focusing on the application aspects and value-generation capabilities of smart shipping are needed [4].To make smart shipping attractive to stakeholders, evidence of its value-creation capabilities resulting from autonomous ships' introduction must be produced.Some studies [5,6,7] fully elaborate on the economic aspects of smart shipping and demonstrate positive results.It is however demanded that more evidence is created to support this argument.
In general, two perspectives in the economic analysis of smart shipping are presented in the literature: (1) the pure 'as-is' replacement of a traditional vessel for an autonomous or teleoperated vessel.The focus is on cost reduction due to smaller crew size, via fuel efficiency, or via vessel redesign to operate more efficiently or carry more cargo [5,6,7,8].Another (2) perspective is the introduction of a new logistics concept in which cargo flows are shifted from road to waterborne transport [9,10].Relevant examples of this shift include the Yara Birkeland and the ASKO cargo ferries.Also in this second perspective, some studies investigate the opportunities of smart shipping in enhancing supply chain performance such as increased reliability and stock optimization [11].
Smart shipping is claimed to contribute to the business case of the modal shift [12].However, to our knowledge, very few studies substantiate this claim by presenting economic arguments showing that autonomy is the breakthrough element that removes the obstacles for executing the modal shift.Therefore, it is not clear to what extent smart shipping is a precondition to realize the proposed logistics concept or a contributing factor in realizing lower logistics costs.For the logistics industry, it is useful to have the insight if or in what conditions adopting smart shipping is the tipping point for a positive business case for modal shift or an alternative logistics concept.The modal shift of short distances is always a challenge because it requires high volumes and involves relatively high transshipment costs [13].In various markets, there is a minimum distance before intermodal becomes attractive.Smart shipping could make the modal shift more attractive if it would reduce the cost of personnel.However, if the costs and benefits of smart shipping provide a tipping point in the modal shift tradeoffs, meaning that intermodal transport becomes attractive in even shorter distances, is not explicitly addressed in studies.
Furthermore, smart shipping may be a solution for commercial flows where small volumes need to be transported and may enable the use of smaller vessels in which personnel costs represent a large share of the overall ownership costs.By investigating whether smart shipping can provide a game-changer advantage for shorter distance transport flows in smaller vessels, we aim at filling the gap that is found in the literature.
This paper explores the economic potential of a modal shift in a dry bulk use case in inland shipping in the region of Zeeland (The Netherlands).In this case, inland shipping is already used to transport fertilizer (the end product) between a factory and a logistics terminal over a distance of 6 kilometers.The analysis focuses on the question of whether a modal shift of inbound raw material destined for the factory from the same terminal would be possible with or without the adoption of smart shipping.Although the use case may not be a typical inland shipping logistics operation, the use case has several characteristics that make smart shipping and saving on labor costs an attractive option: (1) the transport service is not continuous, i.e.only operates on specific days, (2) the transport distance is very short and hence the sailing time is much smaller compared to the loading and unloading time.We expect that the operational and cost model can be generalized to other dry bulk inland shipping use cases and be compared to other markets, for instance, in container shipping [7,10].
The paper is structured as follows: Section 2 provides an overview of current smart shipping literature which focuses on the application and value creation aspects of smart shipping.Section 3 introduces the case study, providing relevant background of the working scenario.Section 4 provides at length the economic evaluation results of the case study.Finally, in Section 5 we interpret the results in the light of the dry bulk inland shipping segment and reflect on its implications.Section 6 concludes with what has been achieved in the study.

Literature Review
When it comes to autonomous ships or smart ships there is not a universally accepted definition.Citing different authors, Van Dijk et al. [14] define smart ships as "a further evolution of already existing subsystems of a ship, which together constitute an (autonomous) vessel".Kretschmann et al. [5] cite the definition of an autonomous ship from the Waterborne Implementation Plan in 2011 as "a ship with next-generation modular control systems and communications technology [that] will enable wireless monitoring and control functions both on and off-board.These will include advanced decision support systems to provide a capability to operate ships remotely under semi or fully-autonomous control".
Autonomy is not a black-and-white categorization of a ship but rather a spectrum of different levels of autonomy.Accordingly, four levels of autonomy are distinguished by the International Maritime Organization (IMO).The first level refers to a ship with automated processes and decision support with crew onboard to operate the ship.A second level refers to a remotely controlled ship with a crew onboard to take over navigation if necessary.A third level refers to a remotely operated ship without a crew onboard.A final fourth level refers to full autonomy, where the ship is capable of making decisions and taking actions by itself.The Central Commission for the Navigation of the Rhine (CCNR) considers in its framework the specific characteristics of inland shipping such as passing through locks, and bridges and considering the water level.They define six levels ranging from non-automated (level 0) to fully automated (level 5).

Expected Benefits from Smart Shipping
Most published studies on autonomous or smart shipping highlight with some level of consensus the benefits that such innovation can bring once it is implemented in practice.Rodseth [2] provides a comprehensive list of what can be a positive gain: for instance, smart shipping can improve working conditions, as sailing can be regarded as a more dangerous activity compared to jobs on land.By unmanning ships, human exposure to risks will be reduced.Smart shipping can lower damage-related costs as automation will help avoid incidents due to human fatigue.It will lead to crew and fuel cost reduction due to low steaming (if business conditions allow).Smart shipping will also help in lowering structural costs as no human facilities will be needed in the vessels and this can also lead to energy efficiency.Higher energy efficiency can also lead to a reduction of emissions and greenhouse gases [8].Finally, Rodseth [2] also argues that smart shipping can enable new ship designs which can trigger the birth of new business models and operating strategies.
Tsvetkova and Hellmstron [3] take a wider view of the benefits and raise the question of how smart shipping can improve the maritime industry as an ecosystem.For instance, Maritime Autonomous Surface Ships (MASS) can play a role in transforming supply and logistics chains.The value creation potential of these ships depends on the degree to which they disrupt logistics as these ships will have a considerable impact on supply and logistics chains beyond that of cost reduction and individual business cases [3].They also argue that with MASS new routes can be created and savings on ship insurance can be achieved due to a detailed log of ships operations.
Pauwelyn and Turf [15] refer to the current struggle for several years in inland shipping with the shortage of skippers.As a consequence smaller inland waterway vessels may disappear and smaller waterways get no longer used, making inland shipping less competitive versus road transport.According to [15] there is an international consensus that the automation of inland vessels can help alleviate the shortage of skippers.

Business gains of Smart Shipping: findings in the literature
Practical tests have shown that ships can autonomously navigate around obstacles and for them to achieve full commercial acceptance in a highly competitive industry such as shipping, the business gains need to outweigh the costs of implementations [6,16].
Kretschmann et al. [5] analyzed the economics of a newly designed deep-sea autonomous ship (i.e. an unmanned dry bulk carrier).The main drivers for economic gains are crew reduction and improved fuel efficiency and they conclude that autonomous ships can have a positive impact on the profitability of shipping companies.The cost dynamics of different subsystems and cost components in the vessel according to [5] are shown in Table 1.[6] investigate the economic viability of an autonomous cargo vessel using a short-sea container vessel.They conclude that savings can be achieved by low manning the ship, however, fully unmanning was found to be not economically viable.Also, the size of the monetary benefits is not as large to represent a big economic incentive to implement the studied concepts.Akbar et al. [17] focused on a short-sea shipping network moving containers between Europe and ports along the Norwegian coastline.By using daughter routes to be executed by autonomous vessels, an average operational cost reduction of 11% can be achieved.Alias et al. studied a decentralized chain of smaller autonomous container vessels in the Rhine [7].They conclude that remote operation can yield a 17% cost reduction, whereas a fully unmanned scenario yields a 31% cost reduction.A 2021 systematic literature review on the economics of smart shipping [18] arrives at similar conclusions and provides a list of open questions that remain to be answered: costs of insurance, costs of cybersecurity, and costs of contingency operations.
As shown above, fewer studies have focused on the inland shipping sector compared to the maritime world.Waterway authorities like Rijkswaterstaat in The Netherlands and De Vlaamse Waterweg in Belgium have commissioned formal studies on the economics of inland smart shipping in their respective areas [19,20].Both studies result in positive business cases and make distinctive conclusions for different industry segments, i.e. dry bulk, liquid bulk, and containers.Both studies argue that autonomous shipping can play an important role in the modal shift due to the increase in competitiveness with road transport.
In summary, most authors agree that there will be an economic incentive to adopt smart shipping.In most studies, these conclusions emerge after comparing baseline reference scenarios (i.e.current scenarios) with the replacement of conventional ships by smart ships.This transition forms the base for conducting cost comparisons and drawing conclusions.Some studies also incorporate economic gains beyond an 'as-is' ship replacement, for example, by transporting more cargo for a longer time with smaller smart ships.Consequently, there seems to be strong deductive support for the notion that smart shipping adoption leads to economic gains.When it comes to arguing why smart shipping is a prerequisite to shifting road transport to waterborne transport, there are hardly any studies providing evidence of this argument.In the next section, we aim at exploring this gap in theory by analyzing the potential of modal shift in a dry bulk use case in inland shipping in the region of Zeeland (The Netherlands).

The case study
Based on the motivations expressed above, we formulate the following research questions: (i) To what extent is teleoperation a precondition to enable a modal shift from road to waterborne transport?(ii) In what conditions does teleoperation enable cheaper logistics concepts for inland shipping?

Current logistics flow
The Gent-Terneuzen Canal and its strategic location facilitate the smooth and cost-effective movement of bulk goods.The logistics flow under study is located along this canal.At one end of the flow, there is an advanced bulk logistics service provider (i.e.3PL) and at the other end, there is the fertilizers producer.These two stakeholders have access to the waterway and they are separated by 6 kilometers along the canal.
The 3PL in question is a renowned logistics service provider that specializes in value-added bulk logistics solutions.The fertilizer producer is a prominent player in the agricultural sector, with demand both domestically and internationally.The producer utilizes dry bulk raw materials (some of them stored at the 3PL) and produces dry bulk finished fertilizer to be stored, bagged, and distributed by the 3PL.The material flow between 3PL and the producer is displayed in Figure 1, including volume sizes.Note that tipper truck movements happen daily whereas barge movements happen three times per week.

Figure 1. The current setup of operations
While this reciprocal movement of products operates without major issues, logistics engineers from the 3PL believe that the reliability of the supply chain can be improved.The direction of products from 3PL to producer (i.e.dry bulk raw materials) is done by tipper trucks through tertiary roads.Because conventional trucks can move up to 25 tons, the number of truck trips per year can raise to 4000 trips per year.This is a considerable burden for the tertiary roads that are used by these trucks.Finished fertilizer is transported by using barges of about 1000-ton capacity.Because this barge is operated on a spot basis, there is not always certainty that a barge will be found when needed.In addition, the freight rate is volatile and the operating costs of the 3PL are highly dependent on the market conditions at that time.Because of the spot market dynamic, the 3PL and the fertilizer producer are not always certain of the reliability of the supply chain and costs.
The barge sailing time is of about one hour from producer to 3PL.Loading and unloading can be done typically at 300 tons/hour, meaning that a normal day trip consists of 3 hours of loading, 1 hour of sailing, and another 3 hours of offloading.The allowed working hours of such a type of barge is 14 hours a day, meaning that to execute this flow the barge operator would need 8 hours to do a complete cycle, plus a couple of hours to move to its next destination.In short, this reflects a big imbalance between sailing and docked time and the time that the captain is performing exclusive captain duties.Finally, the type of barge used in this flow today is classified as CEMT Klasse III, RWS Klasse M4 with crew size 2 (captain and second crew member).

Concept of Operations
The concept to be studied consists of the deployment of a smaller retrofit teleoperated barge (Klasse M2, Kempenaars with 500-ton cargo capacity).The team envisioned a sailing schedule with daily departures with smaller loads.Reducing batch sizes and increasing frequency is a known effort to streamline material flows [21].Expected gains on batch reduction efforts resulting from the use of autonomous ships have also been suggested in the literature [11].Moreover, the elimination of truck movements was proposed to convert the barge operation into a round trip without empty legs.The barge captain will be located at an already established remote control center (RCC) and the onboard crew size is one, i.e. the sailor.High-level requirements for this operation are shown in further detail in Table 2.

Economic Evaluation
The above concept will be evaluated by quantifying various scenarios.The Current situation will be compared to two proposed scenarios, namely, a full Modal Shift scenario with conventional barges and a full modal shift scenario with Teleoperation.Within each scenario, the original size of the barge will be compared to the 500 tons barge.Furthermore, the involvement of the remote captain will be varied.In total, five scenarios will be assessed.For all the scenarios, data is collected via published sources and an extra validation step with an expert was conducted to ensure the direction and magnitude of the figures.Below we explain the base calculation model for this evaluation.

Methodology
The underlying calculation model is the RWS Kostentool Binnenvaart1 (RWS KB).The goal of this tool is to enable transparency of the fundamental cost structure of inland shipping for different vessel types and operating modes.Its outputs are best used as input for policy-making and societal cost-benefit studies [22].This model is based on actual data from the Dutch fleet in the year 2017, and in our study, the figures are corrected to the year 2022 based on yearly indices from [23].
In the RWS KB, the user can modify multiple parameters and select one option out of several alternatives 2 .For ship class, all RWS ship classes are available; for operating mode, solo sailing, day sailing, semi-continuous and continuous are possible; for the sailing area, the options are inland & Belgium, North-South to/from France, the Rhine, and average; for load type, alternatives include sand-gravel, containers, liquid bulk and other.
Accordingly, the model returns cost indicators in Euro and mostly at per annum level.For instance, annual costs for insurance, maintenance and repairs, interest, depreciation, port fees, and salaries.For fuel, it returns fuel usage in liter/km, sailing speed in km/h, fuel price per liter, and fuel costs per year; last, it returns multiple cost indicators such as cost per hour, cost per ton-km, and cost per teu-km.
Because of the different stakeholders present in the inland shipping market (i.e.ship owner, ship operator, charterer, and shippers), it is decided for this study to compute costs from the perspective of the barge owner.Similar to many studies in the literature [5,9,6], the cost of operating and owning a ship are defined according to [24], which is also in line with the cost split obtained from the RWS KB.

Scenarios and parameters
The underlying model is used only as a standard upon which different assumptions are made for each scenario, and additional cost elements are incorporated as a result of the application of smart shipping.Within the scenarios, variations are made in the ship class and the extent of efforts needed by the control center captain.These variations are not performed in the current situation, which only acts as a baseline.The rationale for the selection of the parameters for each scenario is explained in Table 3 and further description is provided in the following sections.

Current situation
This scenario is made of a flow using trucks for raw materials transportation and M4 barges for moving bulk fertilizers.For road transport, the cost per ton-km, loading/offloading, and waiting time are obtained from [25].Based on interviews and data collected on the current operation, it is known how many hours last a round trip and how many tons are moved in each transport leg.Accordingly, a unit cost per ton for the truck portion of 3,1 Euro/ton is obtained.For the barge transport, figures of the RWS KB model are used, and based on the annual total ownership costs of a barge and number of effective hours in a year (i.e.3360 hours based on a 14-hour per day schedule), a cost per hour and cost per ton is obtained.As a consequence, a round trip in this scenario results in a rate of 3,7 Euro/ton (see Current Situation in Table 4).

Modal shift
This scenario calculates the impact of a full modal shift alone without teleoperation.Thus, a round trip using only M4 and only M2 barges is computed.There is no crew reduction, however, a massive cost cut is achieved by eliminating truck and empty movements.The empty movement optimization refers to the removal of an empty transport leg from an assumed offloading location situated 40 km away (e.g.another unrelated shipper).Instead, the barge performs a loaded round trip from the 3PL and/or from the fertilizer producer.The results of this scenario are further expanded in Section 4.3 and Table 4.

Teleoperation
This last scenario deals with similar round trips as in Modal Shift but also incorporates new cost elements resulting from implementing teleoperation.Changes in costs come in the form of variations of the onboard crew salaries, fuel optimizations, implementation of autonomy technology, and use of remote control centers.Each one of these elements is further explained.
The cost of retrofitting a barge is represented in the model taking literature as a basis.Several sources are considered for the selected value.[19] quantified in 100.000Euro with +/-20% possible variation in the autonomy systems for a barge that can perform straight navigation in waterways and for remote sailing through bridges and locks.[5] quantified specific autonomous ship technology for a deep-sea bulk carrier in USD 1.7 million.[6] used in their cost assessment 80.000 Euro for a smart system that can do navigation in open waters and 160.000Euro for a system that can allow automation in navigation near the shore.[20] uses 125.000Euro for the systems needed to perform inland barge teleoperation.In this study, an intermediate one-off value of 150.000Euro is assumed which will be depreciated in ten years.
Different barge class and captain schedule configurations are evaluated in this scenario: • M2-TO is a teleoperated M2 barge whose captain is located at a remote control center.The captain is responsible for the duration of the transport process, i.e. sailing to the loading location, mooring, loading, sailing to the destination, and unloading.This captain operates only one vessel simultaneously.• M2-TO CR replicates M2-TO but its captain can operate three ships at a time.In addition, a reduced schedule for the captain is restricted to only sailing tasks plus the first and last 30 minutes of loading and unloading.• M4-TO CR uses similar assumptions to M2-TO CR but with all costs recalculated to a M4 barge configuration.
Equations ( 1) and ( 2) capture this notion of a reduced schedule for the captain.In (1), T capt is the ratio of captain-exclusive tasks over the total duration of a full trip.This ratio is in function of the duration of the different transportation segments.S load and S dest concern the sailing time in hours to a loading location and to an offloading location.The numerator in (1) includes the 0.5 hour in offloading plus 0.5 hour in loading where the captain ensures the barge is properly engaged for the long duration of loading and offloading operations.L l and L o are the time for loading and offloading in hours.No waiting times or delays are included in any of the scenarios which makes it possible for a fair comparison.
In ( 2), C is the total cost of the remote control center.These costs are in function of different variables.N is the year network cost, which concerns a highly available communication network infrastructure that provides ever-present availability, ample bandwidth, and satisfactory quality on data loss and delays [26].[20] assumes a value of 15k, and in our study, this value is assumed at 18k Euro.R is the hourly rate of a remote control center captain.Captain rates vary based on the CEMT barge class.Values are obtained from [20] and we assume the largest rate of 50 Euro/hour as the captain should be able to remotely operate all kinds of CEMT class barges.H year is the effective hours in a year.M2 and M4 barges can only operate a maximum of 14 hours per day and 240 working days are expected, hence this value is 3360 hours.Finally, S is the number of vessels that the remote control captain is expected to operate.In M2-TO this is set at 1 and in M2-TO CR and M4-TO CR we assume 3. In [5] they assume a value of 5.The year costs obtained for the remote control center are in line with what [8] found in their study.
Similar to [19], reparations and maintenance are assumed an extra 10% of the value of the newly installed technology.For fuel optimization, we use a 6% reduction according to [5].
Finally, when it comes to insurance costs, in [11] it is argued that these can increase in the short term but as smart shipping adoption takes off, premiums can be expected to drop in the medium and longer term.[3] argues as well for a reduction of insurance premiums and [18] leaves this as an open question yet to be fully answered in literature.Because the direction of these costs is not yet clarified, insurance costs have been maintained similar to what is suggested by the RWS KB model.4 shows the effects of eliminating truck movements and using only barges in this flow.This decision alone represents a 3,2 Euro per ton reduction (from 3,7 to 0,47 Euro/ton) which gives more than enough incentive to perform the modal shift.This is obtained not only for the cheaper cost/ton rates observed in barges but also for the fact that an empty leg is eliminated, which means that when performing this service the barge will always have cargo onboard while sailing.Moreover, when performing this same flow by using smaller M2 conventional barges this will result in even lower rates, i.e. 0,39 Euro/ton.This is due to lower ownership costs in smaller barges, i.e. insurance, reparations and maintenance, depreciation, and fuel costs.

Results and sensitivity analysis Modal Shift in Table
Teleoperation for M2-TO does not result in better rates compared to what is already achieved by a conventional modal shift (0,72 > 0,39 Euro/ton).Results show that retrofitting a smaller barge will create new costs, i.e. the costs of new technology and the expected maintenance of it.The onboard captain is removed from the crew which reduces the fixed costs.On the contrary, the variable costs see the effect of reduced fuel usage but also see the addition of the remote control center costs with an on-shore captain and the network costs.When productivity arrangements are made so that the captain can control more than one ship, i.e. 3, and only captain-exclusive tasks are charged, only then, a smaller unit cost in the M2 barge is achieved (0,39 Euro/ton in M2-TO CR ).As explained in the next lines, this effect vanishes as distances increase and at some point, the M4-TO CR barge becomes more attractive.The sailing distance in our case study is 6 kilometers, however, when extending the calculations for larger sailing distances, a linear relationship is found between the number of kilometers sailed and the unit cost per ton.Because an M2 barge with half the capacity of an M4 barge would need to execute twice the number of trips to carry the overall total volume, our initial expectation was to find a higher unit cost.However, for a short-distance flow such as this one, the differentiating factor is the imbalance between sailing hours and docked hours (while loading/offloading).Because the sailing time is so small compared to the loading/unloading hours, the operation leverages the advantage of much lower ownership costs (see Figure 2 for 6 km).
Further analysis showed that for flows of more than 22 km, the sailing portion starts dominating the mix of sailing-operations/docked-operations, which leads to obtaining higher unit costs in an M2 barge compared to a larger M4 barge (see Figure 2 at 22 km).The rate at which this cost deterioration occurs, as the distance is larger and more time of the captain is needed, takes place at a higher rate with the M2-teleoperation scenario, followed by M2 TO Captain Reduced and M4 Captain Reduced.This means that not only the size of the barge but also the ratio of captain-exclusive vs. non-exclusive tasks have a direct link to the ultimate transportation unit cost.

Discussion
The comparison of an envisioned redesigned operation using teleoperated smaller barges versus the present baseline situation has revealed a cost advantage.This statement has nonetheless a   This indicates that the modal shift benefits are evident.Autonomy is not, however, required to realize these gains.Logistics managers can opt to partially optimize this flow by implementing a daily barge service between the two nodes and obtain substantial cost reductions.On the other hand, our results show that upgrading the barge and deploying the captain in a remote control center does not result in a cost-reduced operation beyond what has already been achieved in the conventional modal shift scenario.When looking at the results without considering the conventional scenarios, it can lead to the wrong conclusion that smart shipping in the form of teleoperation was requisite to enable the modal shift.Making this comparison explicit gives support to the calculation approach used in this study to make this effect noticeable.
These results also shed light on the conditions that are needed to achieve the cost-reduction benefits from smart shipping.Most studies in literature based their results on bigger ships with larger crews deployed with semi-continuous or continuous operations, i.e. 20 to 24 hours a day.In studies in which the evaluated scenarios such as short-sea shipping dominates, i.e. for example in Norway or Finland [9], autonomous shipping does provide a game changer advantage.Crew reduction results in larger savings, which outweigh the costs of any service needed from the shore control centers, and ships can also benefit from sailing in a continuous operation.In inland shipping, which plays a much larger role in the Benelux and Germany and is the focus of this study, our results suggest that cost reduction effects of teleoperation will not directly materialize in smaller ships unless some conditions are met.
First, new business practice needs to be introduced in which a captain can oversee more than one vessel at a time.We investigated it using an assumed 1:3 captain:vessel ratio.Moreover, the captain ideally needs to fully monitor only during sailing and during the initial phases of starting loading and offloading.From a liability perspective, this creates questions, but these need to be addressed in the right settings and with the right audience.
Secondly, for flows in which the loading/offloading time dominates over the sailing time, smaller barges show an advantage in very short distances.As the transport distance is increased, the involvement of captain increases.Furthermore, the usual economies of scale of large barges come into play and they become cheaper to operate as they require less fuel per kilometer per ton transported.Such factor needs to be carefully evaluated in potential implementations of this technology.
Our results suggest that productivity gains, without compromising safety and regulations, are needed in the remote control center if small inland ships in the evaluated context expect to observe economic advantages from teleoperation.Future studies can focus on the safest and most economical split of tasks between the on-shore captain and the onboard crew member.Future research can also investigate the effects of fully unmanning the barges in this particular setting.

Conclusions
Recent smart shipping literature is now creating interesting studies into the economic expected gains from autonomous shipping.Some of this literature investigates the effects of replacing conventional vessels with autonomous vessels.In addition, some studies go into investigating the effects of modal shifts to waterborne transport and claim that smart shipping can support this change.While most of this research returns positive business cases, it does not sufficiently address to what extent smart shipping is a necessary precondition for implementing a modal shift.Moreover, the actual impact of variables such as sailing distance and barge sizes on the business case is not fully substantiated.In this article, we aim to bridge this knowledge gap by providing valuable insights through a detailed case study.
The results of our analysis reveal that the benefits of a modal shift, in this case, are evident.It was also found that autonomy is not required to realize these advantages.Because of the larger cost of moving cargo via road versus the more competitive barge rates, the calculated cost reduction is significant.Even if new costs result from enlarging the inventory capacity and making investments in the loading and unloading areas (due to using smaller batch sizes), it is likely that the modal shift is still favorable.
Our results also suggest that teleoperation does not provide additional benefits for flow distances larger than 22 km in the case of small barges, i.e.M2.The longer the distance traveled, the more hours of captain are needed.In shorter distance flows, the differences between an M2 and M4 barge are relatively small because economies of scale and utilization of personnel do not come into play.In short distances, loading and offloading time dominate over the sailing portion, and in such cases, a smaller barge pays off and benefits from its lower capital costs.This means that teleoperation for small ships and short distances can be beneficial for shuttle-like services and for operations where loading/unloading has a much larger proportion than sailing.
Finally, the current practice in which the onshore captain remains 100% assigned to a single vessel, including during non-sailing periods, (which is done from a liability perspective) does not provide an economic benefit and increases costs.New business practice needs to be introduced so that without compromising safety, a better mix of tasks between the onshore captain and crew onboard can be found and lead to cost-efficient operations.
Volume: 200 K tons/Year.(dry bulk fertilizer and dry raw materials) Distance: 6 km, Barge trip duration: +/-1 hour at 12 km/h Target trip volume: 500 MT Frequency: 5 times/week, seasonal One leg duration (load=1.7h,sailing=1h., unload=1.7h)= 4.5h Barge: Retrofit, CEMT II, RWS Klasse M2 with crew size = 1 Allows remote operation from Remote Control Center (RCC) Needs to have 360 view and nighttime vision Ensure strong 4G/5G signal (potential disruption due to proximity to BE border) Operating Rules: Certification of the crew = 1 sailor, i.e. matroos ++ The sailor will be on the bridge during sailing time Mooring and unmooring are done remotely (Matroos assisting) Loading / Offloading takes place only during dry weather RCC on a single-based operation (1 captain → 1 ship) per complexity in the area The ship spends the night at berth at 3PL.Operation begins at 8 am The ship has dedicated berths at both ends of the flow (3PL and Producer) Default Safety Rules: If lost connection, move to side of the fairway and stop until connection is re-established The vessel cannot pass the SVG bridge simultaneously with a deep draft vessel

Figure 2 .
Figure 2. Effect of leg distance on unit cost

Table 3 .
Main parameters used in model