Freight Transport Using Short Sea Shipping

Journal of Shipping and Ocean Engineering 5 (2015) 143-150 10.17265/2159-5879/2015.03.006 D DAVID PUBLISHING

Xavier Ametller SENER Ingenieríay Sistemas Ports, Water & Environment Section, C/Creu Casas i Sicart 86-88 - Parc de l’Alba 08290 Cerdanyola del Vallès Barcelona

Abstract: The increasing imbalance between road and maritime transport has led to saturation of the terrestrial infrastructures. For this reason, the development and analysis of an intermodal freight transport which is safe, reliable and sustainable is the first step to a more balanced growth of other transport options. In this sense, SSS (Short Sea Shipping) rises like a real and efficient alternative to the land road-based transport as well as integrated to a multimodal system door to door for the freight distribution. This paper aims to evaluate and identify all those parameters required to determine the characteristics of a SSS service under a theoretical approach. The increasing growth of the transport market demands to the maritime transport an adaptation effort to the redistribution space processes of production centers and consumption and, as a result of it, of the logistic flows and transport. For this reason, a specific methodology has been developed to model all those phases integrated in the logistic chain of SSS. The final purpose is to determine costs and time of each one of them, for its later implementation in a heuristic algorithm of routing analysis. The main driving factors that determine under which conditions the SSS represents an effective and a feasible solution to road freight are also identified. Information regarding European Union programs to promote the SSS is also presented for a better understanding of the ways of funding sustainable freight transport projects. This investigation states an initial basis for evaluating the competitiveness of SSS concepts, and shows where market and environmental circumstances could be handled in order to enhance the competitiveness of SSS.

Key words: Short Sea Shipping, Barcelona, container, port.

1. A Theoretical Approach to the Short Sea chain from many origins to many destinations or a Shipping SSS typical chain that includes the sea link. For the sake of clarity, the model only allows one sea link 1.1 Building the SSS Model between any origin and any destination. Please refer to A many-to-many approach has been developed to Fig. 1 for a concept scheme of those possibilities. build a cost-time SSS model. The door-to-door The model is fed with Cobb-Douglas equations transport model includes three main parts: which have been obtained through the analyses of Road freight transport. Includes all road based and thirty-one (31) RoRo/RoPax vessels and marketland land operations which consume time and distance. and port freight transport costs of the south Port terminal operations (either origin or Mediterranean region. Stock costs have not been destination). This interface includes all port terminal included in the table. operations that allow changing the land transport Ct (dij )  0.966 dij (1) mode to maritime. These activities only consume Tt (dij )  0.0265 dij (2) 0.441 time. Cop(S)  235.064S (3) 0.483 Sea transport. Includes the movement of cargo by Top(S)  0.037S (4) sea which consumes time and distance. 5 Cm (S)   fi (S) For any origin-destination pair, the SSS model i1 (5) 0.937 allows that the freight transport can be a pure road f1(S)  0.248 S (6) 0.937 f 2 (S )  0.009  S (7) 0.216 Corresponding author: Xavier Ametller, MSc, research f3(S)  160.061 S (8) 0.673 fields: ports & logistics, port infrastructures, transport planning. f4 (S)  0.022  S (9) Email: xavier.ametller@gmail. 144 Freight Transport Using Short Sea Shipping

Table 1 SSS model equations. Source: Own elaboration. year). Local differences have a great impact on the SSS Item equations presented in Table 1 and equations should Land costs (€) Road transport be developed accordingly. For further information Land time (h) Port operation costs (€) regarding modeling details please refer to Port operations Terminal time (h) AMETLLER, XAVIER i SAURÍ, SERGI (2007). Sea costs Capital vessel costs (€/day) 1.2 Defining a Transport Problem—a Simple Operation & Maintenance costs(€/day) Approach Sea transport Crew costs (€/day) Fuel costs (€/day) For a certain group of land origins and destinations Port taxes (€/stop) related to a fixed sea link (same port of origin and port Sea time (h) of destination) the logistic problem to solve is to find

0.441 0.544 f5(S)  0.021 S 85.61 S 1.495 S  the routes that minimize the overall cost of freight 79.931 S0.699 (10) transport subject to certain restrictions. Hence, let the dm Tm (S,dm )  origins be i N  1,2,3...,n , 2.877S 0.201 (11) destinations j M  1,2,3...,m , and the platform where, demand pij (or whatever unit of transport) that dij  land distance (km); should be transported from i to j by road or using d  sea distance (nautical miles, nmi); m SSS. The following expressions will determine the S  gross tonnage of the vessel (GT). function to be minimized: It is important to highlight that the gross tonnage of if the platform (or travel demand) p is the vessel (S) has been selected as the variable to write 1 ij X ij   from i to j by road (13) the equations. The gross tonnage can be written as a 0 on the contrary function of the total number of platforms (or demand) if the platform (or travel demand) 1 that the vessel can transport: p uses a port Yij   ij (14) 0 S(n)  3.532 n1.699 (12)  on the contrary where, Wih  number of trucks (1 platform per truck) from n maximum capacity of the vessel in terms of the land origin i to the port of origin; the number of platforms. Whj  number of trucks from the land destination It should be noted that costs are in euro (2005 base j to the port of destination;

Landn-origins Landm-destinations 1 1 2 2 ......

i j . Port operations Sea transport Port operations . . . . . n m

Fig. 1 Many-to-many SSS model. Source: Own elaboration.

Freight Transport Using Short Sea Shipping 145

Wij  number of trucks that make a pure road to the port of origin (h); transport; thj  land time duration between the port of

Wh  vessel capacity (in terms of platforms). destination and the land destination j (h);

Additionally, Top  port operation time (h);

For each origin-destination pair there will be only Tm  sea time duration (h). one transport choice. The cost logistic equation to be minimized is the i 1,2,3,..., n following: X ij  Yij  1 (15) j  1,2,3,..., m n m n i 1,2,3,..., n Ct (dij ) Wij X ij  Ct (dih ) Wih  X ij ,Yij  0,1 (16) j  1,2,3,..., m i11j i1 The total number of trucks at origin or destination m n Ct (dhj ) Whj  Cop (Wih )  (or equivalently the total number of platforms at the   j1 i1 port of origin or destination) will be equal or less to m the overall capacity of the vessels at the port of origin. Cop (Whj )  Cm (S,d m , r, )Wh (20) n j1 i 1,2,3,..., n Wih  Wh (S) (17) Please note that the demand distribution (in terms i1 of platform demand pij between and origin and a m destination given a total freight demand P ) can be Whj  Wh (S) j  1,2,3,..., m (18) j1 estimated by the widely used gravitational model To obtain Wh , Eq. 12 can be rewritten as: based on the total population. 0.441 i  1,2,3,...,n n(S )  1.894 S (19) pij  ij  P (21) j  1,2,3,...,m The following terms are related to cost and time With, estimations and are also used. Please refer to pij  platforms with origin i and destination j ; Eqs.(1)-(19): P total freight demand between the origins i and dij  road distance between the origin i and the destinations j ; destination j (km);  ij  Gravitational weight defined by: pi p j d ih  road distance between the land origin i to    ij n m i  1,2,3,...,n the port of origin (km); (22)  pi  p j j  1,2,3,...,m d hj  road distance between the port of destination i1 j1 p  population of the origin i ; to the land destination j (km); i p j  population of the destination j . dm  sea distance from the port of origin and the port of destination (nautical miles, nmi); 1.3 Solving a Transport Problem—a Simple Approach Ct  land unit cost (€/h);

C op  port operations unit cost (per platform, €/p); Eq. 20 contains one or more unknowns, which are free variables for which values are sought that cause Cm  sea unit costs (per platform, €/p); r  occupation vessel factor (ratio between the the minimum cost condition to be fulfilled. To be existing cargo and the maximum cargo of the vessel); precise, what is sought are often not necessarily actual   profit of the SSS services (%); values, but, more in general, mathematical expressions.

tij  land time duration (by truck) using pure land Additional expressions are given, in a heuristic form, transport (h) from the origin i to destination j ; to assess the logistic problems which are defined as

tih  land time duration between the land origin i steps:

146 Freight Transport Using Short Sea Shipping

matrix is updated. Step 1: The algorithm starts by defining an initial X ij occupation factor r0 (ratio between the existing Step 5: Once the occupation factor r is stable, the cargo and the maximum cargo of the vessel), which algorithm evaluates and compares the time duration of allows to define the cost per platform when using the each mode (road and SSS) for each origin-destination SSS multimodal chain. pair. A new parameter in the algorithm t is Step 2: The algorithm compares the cost of a included, which is the maximum time percentage that transporting one platform from i to j by road or the SSS can be slower compared to the road to retain SSS as follows: the cargo. For example, if t  25% and following i  1,2,3,...,n economic criteria the freight should be moved by SSS if Cij, SSS  Cij, ROAD  X ij  0 j  1,2,3,...,m (23) T  49h ( X ij  0 ), and ij, SSS and Tij , ROAD  40 h , the r  r 0 algorithm evaluates the condition Step 3: The algorithm checks the Eqs. (17)-(18) to Tij,SSS  1 tTij, ROAD , fulfilled in this case as make sure that the capacity of the vessels is not 49h  40h1 25%  50h . The purpose of this overpassed. The vessel is loaded starting with the constraint is to provide a limit by which the time biggest order p simulating the fact that companies ij flexibility is considered regardless of strictly cost with high export volumes have preferred contracts criteria. with shipping companies. After this loading criteria, if X ij  0  Tij, SSS  i  1,2,3,...,n there will be a first approximation of the number of (27) 1  tTij, ROAD  X ij  0 j  1,2,3,...,m trucks, from an origin i that are allowed to go the port This final step makes necessary to update the of origin. Eq. 22 describes the vessel loading criteria: solution X ij and repeat the sequence until the  W ih  maxWih occupation factor is stable. This model was n m  i  1,2,3,...,n successfully tested with real data of South if W ih Yij  Wh   j 1,2,3,...,m (24) Mediterranean SSS routes. i11j    Kindly refer to AMETLLER, XAVIER i SAURÍ, Wih  W1h ,...,W nh  W ih   SERGI (2007) for a complete description of the model Where, and a Microsoft Visual Basic © code implemented in 1 if the platform (or travel demand) pij uses an Excel © spreadsheet. Yij   a port (25) 0 on the contrary 1.4 Key Factors in the SSS Wih  Number of trucks from the land origin i to the port of origin. The model is useful to predict certain key factors Step 4: Following the steps above, an initial that may be considered for transport planners and solution is stored in the matrix X ij which meets decision makers in order to address new regulations or optimized cost criteria. At this point, the algorithm developments regarding this multimodal freight updates the value of the occupation factor r which transport and contribute positively in the future was defined as a default number in Step 1 with its success of SSS. Please note that there are a good actual value until an error tolerance (set at a value of number of publications that describe a long list of neck bottles of the SSS, such as heavy administration r  r0  1% ) is reached: n procedures, port terminal infrastructure deficiencies, WihYij i  1,2,3,...,n  (26) flexibility and frequency of the SSS service. The r  i1 j  1,2,3,...,m Wh purpose of this section is to define the key driving Following the step defined above the solution factors of the SSS which can be inferred directly from

Freight Transport Using Short Sea Shipping 147 the heuristic model. the market share of the SSS and promote the modal The following characteristics are considered to shift. define a reasonable framework of the Short Sea Key Factor 3: A carbon emission reduction of up Shipping in the European South Mediterranean region to 40% (compared to the current situation) can be and set a basis of design: achieved if the predicted modal shift of the model is Sea distances ranging from 450 up to 800 nautical obtained. The model (2005 prediction) estimates that a miles (nmi); modal shift of up to 20% of the pure road to the SSS is Fleet size with a gross tonnage ranging from 8,500 possible (with regard to the cost & time criteria GT up to 30,000 GT; defined above) which may lead to a more sustainable SSS multimodal routes that include from 30% up to transport option. Note that externality costs have not 75% of land freight transport of the total trip distance. been included in the analyses. Making several simulations the key factors 1.5 Conclusion identified are summarized as follows: Key Factor 1: The total road distance in the SSS The increasing growth of the transport market multimodal route should not exceed the 60% of the demands to the maritime transport an adaptation effort total distance to be competitive. to the redistribution space processes of production Key Factor 2: According to the area included in centers and consumption and, as a result of it, of the the scope of the analyses (South Mediterranean logistic flows and transport. The model presented is region) port terminal costs and port operation time just a first step to policy makers or transport planners (vessel loading and unloading) are driving the cost & to demonstrate that Short Sea Shipping is a real time competitiveness of the SSS against the road alternative to the pure land transport. The tool transport. Investments on port infrastructure and presented is able to predict the cost and time increase the efficiency in the operations may increase reductions of the overall freight transport of a region,

SSS key driving factors: Maximum total road distance to be competitive 100% SH: High Gross tonnage = 30,000 GT 450 nmi (nautical miles) - SH 95% SM: Medium Gross tonnage = 15,000 GT 600 nmi - SH SL: Low Gross tonnage = 8,500 GT 800 nmi - SH 90% 450 nmi - SM 600 nmi - SM 85% 800 nmi - SM 450 nmi - SL

80% 600 nmi - SL 800 nmi - SL

75%

70%

65%

60%

55% % Cases Cases %where is SSS costcompetitive against pure road transport 50% 25% 35% 45% 55% 65% 75% 85% % Total road distance in the SSS logistic chain. Minimum is set at 30%

Fig. 2 Freight transport analyses considering different gross tonnage capacities and different sea links in the multimodal SSS option. Source: Own elaboration

148 Freight Transport Using Short Sea Shipping

1.400

1.200

1.000

800 42% 100% 600

Cost comparison (€) 400 36%

200 22% 0 SSS - Cost Pure Land - Cost Land transport Port terminal Sea link

25 61% 20 100%

Time comparison (h) 10 26%

5 13% 0 SSS - Time Pure Land - Time Land transport Port terminal Sea link

Fig. 3 Cost (up) and time (down) structure of the typical situation where SSS is more competitive than the pure road freight transport. Costs (2005 EUR) are per platform. Source: Own elaboration. which may be used to plan programs to develop and procedures and a change in the existing incent this kind of sustainable freight transport regulatory framework supported by the multimodal option. development of electronic data exchange systems and surveillance capabilities by coastal and authorities in 2. European Framework of the SSS ports. 2.1 EU Initiatives Two different initiatives have been introduced in the EU to promote the SSS. The first one is the The EU vision is to plan a maritime transport space TEN-T, a high level program which includes the without barriers which extends the Common Market improvement of existing maritime links to establish to sea transport. The effective implementation new viable, regular and frequent maritime links for the pre-supposes a simplification of administrative transport of goods between Member States in order to

Freight Transport Using Short Sea Shipping 149 reduce road congestion. The second program is the should be used for a preliminary estimation of the Marco Polo which aims to shift freight from the roads funds to be assigned to the promotion program. to more environmentally friendly modes and to 2.2 A SSS Reference: Port of Barcelona support commercially-oriented services in the freight transport market and finance actions involving The Port of Barcelona is the leading Port in Spain candidate countries. The characteristics of both in Short Sea Shipping (SSS) traffic. The demand for programs are briefly summarized below: this type of transport has led the Port to set up an In general, the approach to corridors (TEN-T) or increasing number of short-distance maritime lines, through the subvention of services (Marco Polo I and which link it to other Mediterranean and North II) have not been fully convincing as they seem to be African ports. The competitiveness of Short Sea based more on lobbying than on objective findings. Shipping and the investment made by port facilities Promoting the greater involvement of PPP (Public and shipping lines to make this a real alternative to Private Partnerships) is the new approach that may road transport can be clearly seen by the spectacular overcome the difficulties of the past. Please refer to growth figures. In 2013, the SSS moved a total of the European Commission web portal for more details 100,716 intermodal transport units (ITU-equivalent to regarding these initiatives. In both cases, a freight trucks, platforms and trailers) which represents a demand model like the one presented in this paper year-on-year increase of 10 %, driven mainly by renewed

Table 2 EU programs to promote the SSS. Source: Own elaboration from Anne Bårseth: Executive Agency for Competitiveness and Innovation (EACI) Marco Polo Unit. Program TEN-T MARCO POLO Driving force Public sector driven Private sector driven Actors Top-down (Member States) Bottom-up (undertakings) Schedule Long-term Short-term Mission Creation of transport network Modal shift objective Target Infrastructure Transport services Infrastructure Strategic infrastructure Ancillary infrastructure

SSS services/week from the Port of Barcelona 30 Non EU destination EU destination 25

10 Services/week

0 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

Fig. 4 SSS services per week in the Port of Barcelona. Source: Own elaboration from Port of Barcelona - 2014 data.

150 Freight Transport Using Short Sea Shipping

Fig. 5 Grimaldi SSS Terminal in the Port of Barcelona. Source: Port de Barcelona, 2014. growth in traffic with Italy and a high export capacity References of the hinterland. [1] Ametller, Xavierand Saurí, Sergi 2007. “Optimización Thanks to its strategic location and its ever-larger del transporte de mercancías mediante el transporte and important hinterland, the Port of Barcelona is marítimo de corta distancia.” Universitat Politécnica de Catalunya. Barcelona. Accessed at Aug 10th, 2014. developing strategies to increase the number of SSS http://upcommons.upc.edu/pfc/bitstream/2099.1/4942/1/ services and offer a more flexible and competitive Mem%C3%B2ria.pdf maritime transport service to its companies and clients. [2] Autoritat Portuària de Barcelona 2014. Accessed at Aug The Port currently has berthing, loading and 10th, 2014. http://www.portdebarcelona.cat/ [3] Camarero, Alberto, and Polo, Gerardo. 2005. “Ro-Ro Ships unloading areas dedicated to this particular mode of for Short Sea Shipping.” PIANC Magazine (188): 35-41. transport and has become a touchstone for this type of [4] Cullinane, K., and Khanna, M. 1998. “Economies of short distance traffic. Scale in Large Container Ships.” Journal of Transport Economics and Policy (33): 185-208. The Port has three specialized terminals for trucks, [5] Jansson, J. O., and Sheneerson, D. 1988. “A Model of vehicles and passengers and frequent SSS services Scheduled Liner Freight Services: Balancing Inventory between Barcelona and the main ports of the Cost against Shipowners’ Costs.” The Logistics and Transportation Review (21): 3. Mediterranean and North Africa (refer to Fig. 4). The [6] Martínez de Osés, F. Xavier and Castells Sanabra, Marcel·la last SSS terminal in the Port of Barcelona (refer to Fig. 2005. “Análisis de los buques dedicados al transporte 5) received a TEN-T contribution of 1.4 million euros marítimo de corta distancia internacional en España.” Grupo de investigación TRANSMAR. Departamento de Ciencias e (10% of total project works cost) under the key Ingeniería Náuticas-UPC. Spain. objective of the improvement of intermodal transport, [7] Ministerio de Fomento de España 2002. “Informe sobre as it will allow upgrading the efficiency along the el cabotaje comunitario o transporte marítimo de corta whole logistics chain between the Iberian Peninsula, distancia.” Documento para el consejo informal de ministros de transportes de la Unión Europea. Gijón. North Africa and Italy through the removal of existing [8] Robusté, F. 2005. Logística del transporte. Barcelona bottlenecks. UPC.