Locational Characteristics of Dry Ports in Developing Economies: Some Lessons from Northern India

This article was downloaded by: [Hong Kong Polytechnic University]On: 29 May 2012, At: 23:40Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: MortimerHouse, 37-41 Mortimer Street, London W1T 3JH, UK

Regional StudiesPublication details, including instructions for authors and subscription information:http://rsa.tandfonline.com/loi/cres20

Locational Characteristics of Dry Ports in DevelopingEconomies: Some Lessons from Northern IndiaAdolf K. Y. Ng a & Ismail B. Cetin ba Department of Logistics and Maritime Studies, Faculty of Business, The Hong KongPolytechnic University, Hung Hom, Kowloon, Hong Kong, Chinab Maritime Faculty, Dokuz Eylul University, Cumhuriyet Bulvarı No. 144, 35210, Alsancak,Izmir, Turkey

Available online: 21 Sep 2011

To cite this article: Adolf K. Y. Ng & Ismail B. Cetin (2012): Locational Characteristics of Dry Ports in DevelopingEconomies: Some Lessons from Northern India, Regional Studies, 46:6, 757-773

To link to this article: http://dx.doi.org/10.1080/00343404.2010.532117

PLEASE SCROLL DOWN FOR ARTICLE

The Regional Studies Association (http://www.regional-studies-assoc.ac.uk) has licensed the Taylor & FrancisGroup to publish this article and other materials. To join the Regional Studies Association please visit http://www.regionalstudies.org/join/benefits.asp.

View the Regional Studies Association Disclaimer (http://www.regional-studies-assoc.ac.uk/disclaimer.asp)

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form toanyone is expressly forbidden.

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions. For articlespublished as an Open Select article please note Part II. Intellectual property and access and license types, §11. (c) Open Access Content.

For articles published as Open Select articles, please note that the use of these articles for commercialpurposes is strictly prohibited.

The publisher does not give any warranty express or implied or make any representation that the contentswill be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug dosesshould be independently verified with primary sources. The publisher shall not be liable for any loss, actions,claims, proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly orindirectly in connection with or arising out of the use of this material.

http://rsa.tandfonline.com/loi/cres20

http://dx.doi.org/10.1080/00343404.2010.532117

http://www.regional-studies-assoc.ac.uk

http://www.regionalstudies.org/join/benefits.asp

http://www.regionalstudies.org/join/benefits.asp

http://www.regional-studies-assoc.ac.uk/disclaimer.asp

http://www.tandfonline.com/page/terms-and-conditions

Locational Characteristics of Dry Ports inDeveloping Economies: Some Lessons from

Northern India

ADOLF K. Y. NG* and ISMAIL B. CETIN†*Department of Logistics and Maritime Studies, Faculty of Business, The Hong Kong Polytechnic University, Hung Hom,

Kowloon, Hong Kong, China. Email: [email protected]†Maritime Faculty, Dokuz Eylul University, Cumhuriyet Bulvarı No. 144, 35210 Alsancak, Izmir, Turkey.

Email: [email protected]

(Received March 2009: in revised form October 2010)

NG A. K. Y. and CETIN I. B. Locational characteristics of dry ports in developing economies: some lessons from Northern India,Regional Studies. Contemporary economic development has triggered the importance of establishing distribution centres at appro-priate locations in developing economies. However, so far, academic studies addressing this issue are rather scarce. Hence, by inves-tigating dry ports within two industrial clusters in Northern India, this paper investigates the locational characteristics of distributioncentres in developing economies. Analytical results indicate that the spatial dynamics of dry ports in developing economies aredifferent from Western, advanced economies due to geographical diversifications, the different paces of regional developmentand local practices. This paper also suggests that dry ports in developing economies are more cluster (rather than supply chain)oriented.

Dry port Location Developing economies India

NG A. K. Y. and CETIN I. B.发展中经济体无水港的区位特征：来自北印度的相关证据，区域研究。当代经济发展强调了在发展中国家特定区位建立配送中心的重要性。然而迄今为止学术界对该问题的关注甚少。通过考察位于北印度两大产业集群的无水港，本文考察了发展中经济体配送中心的区位特征。分析结果表明，发展中经济体无水港的空间活力与西方发达经济之间由于地理差异、不同的区域发展节奏以及地方实践而有所不同。研究同时表明，发展中国家无水港多是集群（而非供应链）导向的。

无水港区位发展中经济体印度

NG A. K. Y. et CETIN I. B. Les caractéristiques des emplacements des ports secs situés dans les économies en voie de développe-ment: des leçons à tirer du nord de l’Inde,Regional Studies. Le développement économique contemporain a déclenché l’importanced’établir des centres de distribution dans des emplacements appropriés dans les économies en voie de développement. Cependant,jusqu’ici, plutôt rares sont les études théoriques qui abordent cette question. Donc, en examinant les ports secs situés au sein de deuxclusters industriels dans le nord de l’Inde, cet article cherche à examiner les caractéristiques des emplacements des centres de dis-tribution dans les économies en voie de développement. Les résultats analytiques laissent voir que la dynamique géographique desports secs situés dans des pays en voie de développement se distinguent de celle des économies avancées de l’Ouest à cause desdiversifications géographiques, de la variation du rythme de l’aménagement du territoire, et des pratiques locales. Cet articlelaisse supposer aussi que les ports secs situés dans les économies en voie de développement sont orientés plutôt vers les clusters(que vers les chaînes d’approvisionnement).

Ports secs Emplacement Economies en voie de développement Inde

NG A. K. Y. und CETIN I. B. Standortmerkmale von Trockenhäfen in Schwellenländern: Lehren aus Nordindien,Regional Studies.Aufgrund der modernen wirtschaftlichen Entwicklung ist es in Schwellenländern wichtig geworden, Vertriebszentren an geeigne-ten Standorten zu errichten. Dieses Thema wurde jedoch bisher in wissenschaftlichen Studien nur selten behandelt. Aus diesemGrund werden in diesem Beitrag die Standortmerkmale von Vertriebszentren in Schwellenländern anhand des Beispiels vonTrockenhäfen in zwei industriellen Ballungsräumen Nordindiens untersucht. Aus den analytischen Ergebnissen geht hervor,dass die räumliche Dynamik von Trockenhäfen in Schwellenländern aufgrund von geografischen Diversifizierungen, unterschie-dlichen Geschwindigkeiten der regionalen Entwicklung und lokalen Praktiken unterschiedlich ausfällt als die Dynamik in westli-chen Industriestaaten. Ebenso wird in diesem Beitrag der Schluss nahegelegt, dass Trockenhäfen in Schwellenländern stärkerclusterorientiert (statt lieferkettenorientiert) sind.

Regional Studies, Vol. 46.6, pp. 757–773, June 2012

0034-3404 print/1360-0591 online/12/060757-17 © 2012 Regional Studies Association http://dx.doi.org/10.1080/00343404.2010.532117http://www.regionalstudies.org

Dow

nloa

ded

by [

Hon

g K

ong

Poly

tech

nic

Uni

vers

ity]

at 2

3:40

29

May

201

2

Trockenhafen Standort Schwellenländer Indien

NG A. K. Y. y CETIN I. B. Características de ubicación de los puertos secos en las economías en desarrollo: lecciones del norte de laIndia, Regional Studies. El desarrollo económico contemporáneo ha impulsado la necesidad de establecer centros de distribución enubicaciones apropiadas de las economías en desarrollo. Sin embargo, hasta ahora se han llevado a cabo poco estudios académicossobre este tema. Por este motivo, en este artículo analizamos los puertos secos en dos aglomeraciones industriales del norte de laIndia para investigar las características de ubicación de los centros de distribución en economías desarrolladas. Los resultados analí-ticos indican que las dinámicas espaciales de los puertos secos en economías en desarrollo son diferentes de las economías avanzadasde Occidente debido a las diversificaciones geográficas, los diferentes ritmos del desarrollo regional y las prácticas locales. En esteartículo también sugerimos que los puertos secos en las economías en desarrollo se orientan más hacia las aglomeraciones (en vez dehacia la cadena de suministros).

Puerto seco Ubicación Economías en desarrollo India

JEL classifications: O53, P2, R12, R40

INTRODUCTION

Technological improvement, economic developmentand the establishment of a global supply chain pose sig-nificant implications to ports, not only in the nature ofdemands for their services, but also in geographicalmonopolies, where intensified inter-port competitionhas contributed to the blurring of boundaries demarcat-ing port-specific fore- and hinterlands (HEAVER, 2002).To tackle such challenges, it is a natural corollary forthem to extend further inland by patronizing, formingstrategic alliances and buying out existing dry ports soas to control the supply chain (NOTTEBOOM andRODRIGUE, 2005; ROSO et al., 2009), and this trendis especially prominent in Western, advanced econom-ies, like the United States and the European Union.Hence, dry ports have gradually affirmed their signifi-cance in complementing market development, havingseamless integration and having closer collaborationwithin the supply chain.

Generally speaking, a dry port can be understood as adistribution centre where freight (de-)consolidationtakes place with functions similar to those of seaports,including cargo handling, providing intermodal trans-port connections, information exchange, as well asother ancillary services like custom clearance, inspec-tions, storage, maintenance and repair of empty contain-ers, and tax payments (BERESFORD and DUBEY, 1990;UNITED NATIONS ECONOMIC COMMISSION FOR

EUROPE (UNECE), 2001). In this respect, the establish-ment of dry ports allows shippers to undertake variouslogistics activities at more remote, often inland, siteswhich are often more proximate towards productionfacilities, reducing transaction costs and accompanyingrisks, thus triggering the competitiveness of their pro-ducts in the global market.

The establishment of dry ports is also acceleratingwithin developing economies, as typified by India.1

Until 2009, more than 200 dry ports have been estab-lished throughout India, mainly by the national

government. Among them about sixty are operated byContainer Corporation of India Ltd. (CONCOR),India’s flagship state corporation in providing transportand logistics services. However, dry ports in India areoften established with little serious locational analysisnotably the existence of genuine demands in usingsuch facilities.2 Indeed, locating a dry port properly,that is, what particular hub(s) should be called/estab-lished (for the so-called location–allocation problem,see SHERALI et al., 2002), is of utmost importance foreffective freight movements between production basesand gateway seaports. As per the New EconomicGeography, making appropriate locational choices canresult in cost savings, notably transport costs, thusenhancing the competitiveness of locally produced mer-chandize within the global market due to increasingreturns to scale (KRUGMAN, 1991, 1996; BEHRANS

et al., 2007) and, ultimately, encouraging trade withother countries/regions and improving that particularcountry/region’s overall welfare. Moreover, while strat-egies and decisions in ‘weightless economy’ industrieslike information and communication technology(VICENTE and SUIRE, 2007) are relatively short-termand can thus be altered rather easily in response tomarket demand and the availability of land and capital,the locations of distribution centres (and facilities),including dry ports, are often geographically fixed andthus difficult to reverse in the short- or medium-term.Like industrial plants, inappropriate site selections canwaste considerable capital costs, resulting in the estab-lishment of economically unsustainable distributioncentres, thus affecting the efficiency of the long-termoperations of supply chains (ZHU et al., 2010).However, while a number of research works investi-gating the locational characteristics and spatial dynamicsof dry ports in Western, advanced economies exist (forinstance, RUTTEN, 1998; MACHARIS and VERBEKE,1999; HESSE, 2004; HESSE and RODRIGUE, 2004;NOTTEBOOM and RODRIGUE, 2005; RAHIMI et al.,

758 Adolf K. Y. Ng and Ismail B. Cetin

Dow

nloa

ded

by [

Hon

g K

ong

Poly

tech

nic

Uni

vers

ity]

at 2

3:40

29

May

201

2

2008; ROSO, 2008; ROSO et al., 2009), investigationsinto developing economies have remained ratherscarce, with the possibly works of NG and GUJAR

(2009a, 2009b) and NG and TONGZON (2010) beingnotable exceptions.3

The scarcity of research focusing on dry ports indeveloping economies has left a huge gap in the researchthat is yet to be filled. In recent years various workswhich mainly focused on the experiences of Western,advanced economies (like the United States andWestern Europe) identified the development of dryports as an ‘outside-in’ strategy to relieve seaport con-gestion and/or control the supply chains, with shippinglines and seaports being the major consumers of dry portservices (cf. SLACK, 1999; ROSO et al., 2009; RODRI-

GUE et al., 2009). Such an understanding of theconcept and functions of dry ports is not necessarilytrue for developing economies, however, especiallygiven their reliance on exporting agricultural andlabour-intensive products, as well as the vast existenceof small- and medium-sized producers/shippers (NG

and GUJAR, 2009a). In some cases, the very existenceof dry ports can even act as a prerequisite for some ofthese rather low-value products to compete in the inter-national market (and thus export), especially given theoften disorganized and highly fragmented nature oftheir logistics sectors. In these cases, the positiveimpacts of dry ports include: (1) the preservation (andeven improvement) of a product’s quality, thus sustain-ing (or increasing) its value; and (2) the lowering oftransport costs (through consolidation) and damage tocargoes, which in turn reduces the cost of production.4

For emerging markets with a large landmass, like Braziland India, dry ports can act as catalysts encouragingintra-national trade and interactions between differentregions, thus promoting regional development(UNITED NATIONS ECONOMIC AND SOCIAL COM-

MISSION FOR ASIA AND THE PACIFIC (UNESCAP),2006). Under such circumstances, contrary to theexperiences of advanced economies, the strategic con-sideration of developing dry ports in developing econ-omies can sometimes be the other way round – being‘inside-out’ with shippers, rather than shipping lines orseaports, being the major consumers of dry port services.Such differences would affect not only the roles andfunctions of dry ports, but also possibly their competi-tive structure. Clearly, the existing literature has yet tooffer a comprehensive theoretical understanding onhow dry ports work under diversified developmentalstages in different geographical regions.

Moreover, such differences have also left severalimportant questions yet to be answered. First, infacing globalization, the establishment of supply chainsand increasing regional institutional convergence glob-ally (HALL, 2003), can generic solutions (dry portswith similar roles, functions and facilities) be appliedto different geographical regions? Also, what should bethe roles, functions and direction of the development

of dry ports in developing economies? Finally, giventhe diversifications between different regions, forexample, geographical, economic, social systems, etc.,are there any significant differences, in terms of spatialpatterns, between dry ports in advanced and developingeconomies, especially given the understanding that itscounterpart, the seaport, has been generally less affectedby globalization (LEE et al., 2008)? By understandingsuch deficiency and using two case study regions inNorthern India, this paper aims to address the aboveresearch questions from a spatial perspective. Bearingin mind that the location of particular facilities, andtheir success, can depend on the locations and oper-ations of other players within the market (CHAPMAN

and WALKER, 1991; VICENTE and SUIRE, 2007), it isproposed that the spatial dynamics of dry portsbetween different types of economies are dissimilardue to fundamental diversifications between regions,the pace of development, policies and user behaviours,and that the conventional experience from advancedeconomies on the roles, functions and orientation ofdry ports is not necessarily applicable to developingeconomies.

The paper is structured as follows. The secondsection gives the theoretical background. The thirdand fourth sections describe the case study regions andmethodology, respectively; while the fifth and sixth sec-tions illustrate the simulated results and discuss the majorfindings, respectively. Finally, in the seventh section,conclusions, including major lessons drawn from thispaper, are given.

THEORETICAL BACKGROUND

The core concern of locational studies relates not onlyto the optimal usage of available space, but also to theprecise sites where particular facilities should be estab-lished, as well as the major criteria in sustaining theirlong-term survival. Indeed, the essence of the geo-graphical approach is that spatial issues are central andfundamentally concerned with how economic activitiesare organized in particular chosen areas and the under-lying processes which lead to the creation of spatial pat-terns. As noted by ULLMAN (1941), traffic generated bya particular place is at least partly due to its geographicalsetting.

The study of location has traditionally been the focusof concern within the subfields of economic geography,including industrial, urban and transport geographies(CHAPMAN and WALKER, 1991). Generally speaking,the study of location can be classified into the followingmajor categories, namely location and land use, thelocation of manufacturing facilities, as well as thelocation of distribution centres within a retail networkor supply chain (THAI and GREWAL, 2005). Thus,one cannot address the location of distribution centreswithout mentioning location theory, its concepts and

Locational Characteristics of Dry Ports in Developing Economies in India 759

Dow

nloa

ded

by [

Hon

g K

ong

Poly

tech

nic

Uni

vers

ity]

at 2

3:40

29

May

201

2

influences, as this theory has long been receiving closeattention from both scholars and practitioners datingback to earlier centuries. Johann von Thünen’s treatiseThe Isolated State (1826) was widely accepted as thepioneer study focusing on the relation between locationand economic development (cited in BARNES, 2003),while other human geographers, especially Germanand American – say, WEBER (1909), ULLMAN (1941),LOSCH (1954), KRUGMAN (1991) and GERTLER

(2001) – also provided substantial inputs. Largely basedon the minimization of transport costs (BALLOU,1999), classical location theory mainly addresses twoparadoxes. First, where do particular firms/facilitieslocate in order maximize profits? In other words, whydo certain locations develop rather than others? Also,by considering the existing locations, where are theappropriate locations for firms/facilities to market theirproducts in the most profitable way? Indeed, theemphasis on transport costs is so influential that researchworks based on transport cost-based models are abun-dant, for instance, the conditional logit model (forexample, COUGHLIN et al., 1991; FRIEDMAN et al.,1992), mixed-integer programming (for example, NG

and GUJAR, 2009a), the dynamic programming model(for example, CANEL and KHUMAWALA, 2001), andthe centre of gravity (CoG) model (for example, THAI

and GREWAL, 2005; CETIN and CERIT, 2008).5

Nevertheless, as noted by WEBER (1909) in hislocation-preference theory, and later also by NOTTE-

BOOM and RODRIGUE (2007), site selection can beaffected by both ‘economic’ and ‘non-economic’factors, which can be quantitative or qualitative innature. In this respect, two important concepts mustbe highlighted. The first attribute emphasizes the needfor locating a facility as close to the production basesand/or market places as possible, so as to gain competi-tive advantages not only by minimizing transport costs,but also other potential benefits (FUJITA andKRUGMAN, 1999; HESSE, 2007). This concept is a cor-ollary to the agglomeration process, where a centralplace, be it local, regional, national or transnational,has positive impacts on its size, functions and traffic-generating potentials. In this respect, through his land-mark dissertation analysing German urban centres,CHRISTALLER (1933) constructed symmetrical hexa-gons and highlighted the ‘central place’ as possessing ahigher order of significance against its surroundingregion (HOTTES, 1983). Such pulling force by pro-duction bases and/or market centres is referred to ascentrality by FLEMING and HAYUTH (1994), beingpart of the non-transport frictions identified byBEHRANS et al. (2007), or the so-called ‘location-specific’ factors identified by BERKOZ and TURK

(2008). The significance of central places is also notedby ULLMAN (1941) who stresses the relation betweenproductive land supporting the market centre andother essential services provided by the market centreto the surrounding areas. This proposition is also

supported by TOBLER (1970), LUNDVALL andJOHNSON (1994) and GERTLER (1995), who pointout that interactions work better when users and produ-cers are geographically proximate. Hence, sooner orlater central places would become gateways to distantplaces outside the region and the foci of social, econ-omic, transport or even political activities (LOSCH,1954; CHAKRAVORTY and LALL, 2005). However,while such force can be induced by the artifice of trans-port infrastructure (DICKINSON, 1961), their attractive-ness also greatly depends upon the perceptions of facilityusers, as pointed out by SLACK (1985) and NG (2006)during their analysis of seaport selection and compe-tition. Thus, centrality regards locating a transport facil-ity close to the production sites and/or market places asnot only gaining competitive advantages through costminimization, but also lifting surrounding facilitieswith positive economic spillover, including non-econ-omic attributes, for example, reputation, informationflows, convenience, etc.

On the other hand, the location of distributioncentres can be also affected by intermediacy. At suchsites, local services connect with national/internationalservices, as well as one transport mode with another(FLEMING and HAYUTH, 1994). Indeed, sometimessuch places have nothing but their geographical settingsso as to enable them to become distribution centres.Hence, intermediacy can be understood as the spatialquality that can be identified in the context of the trans-portation system and can generate additional traffic, iffavoured by transport service providers as connectinghubs, usually as connecting points between origins anddestinations. However, this does not only imply directmeasurements of geographic distance and impediments,where such importance is also imparted to the distri-bution centres by users who may decide to take theimportance away by shifting their business to other dis-tribution hubs due to several reasons, for example, theestablishment of new sites, the rise of alternatives, tech-nological innovation, changing trade pattern, etc. In thiscase, the magnitude of a user’s perspective is equallyimportant and is often directly proportional to thelevel of traffic. Furthermore, while locational choicesare, in many cases, guided by economic considerations(HESSE, 2004), at the same time, as illustrated by NG

and GUJAR (2009b), they can also be affected by gov-ernment policies and exogenous factors, especiallywhen services are not purely profit motivated(HOOVER and GIARRATANI, 1985). This is especiallytrue for developing economies where politics canmatter at least as much as economics within themarkets (BREMMER, 2005), and this is not helped bythe fact that locational decisions nowadays can be dueto compromises of contrasting dynamics betweenglobal and local forces (HALL, 2004).

However, centrality and intermediacy are not necess-arily clear-cut, or even overlapping (FLEMING andHAYUTH, 1994). While the initial establishments of


Dow

nloa

ded

by [

Hon

g K

ong

Poly

tech

nic

Uni

vers

ity]

at 2

3:40

29

May

201

2

new distribution hubs can be due to favourable physicalconditions and intermediate locations, their attractive-ness can increase and develop into central places, asbusiness and complementary industries start to movein and mutually assist each other through agglomera-tion, bringing different types of positive effects to sur-rounding economies (BUTTON, 2004), thus reducinguncertainty and better information flows and, comple-mented by improving accessibilities, encouraging evenmore interactions (HESSE and RODRIGUE, 2004;VICENTE and SUIRE, 2007). This is especially visiblefor ports, given their roles as gateways between fore-and hinterlands (WEIGEND, 1958; BIRD, 1973). Thisprocess can also be observed within distribution hubsin advanced economies, where many North Americanand European dry ports have been transformed intomultifunctional establishments, attracting serviceagglomeration and developing into regional centres(NOTTEBOOM and RODRIGUE, 2005, 2007). Hence,it is clear that locational choice can be highly compli-cated, and where a favourable location does not necess-arily create genuine demands to ensure a facility’ssustained survival (LOSCH, 1954).

THE CASE STUDY REGIONS

This paper investigates the spatial dynamics of public dryports, that is, those which are not dedicated to particularenterprises or cargoes, from two industrial clusterslocated within one of the world’s important developingeconomies – India. For the former, it will focus on theindustrial clusters within the central and northern statesof Maharashtra and Gujarat, respectively. Existing, localdry ports within the regions include Sabarmati InlandContainer Depot (SICD) (Gujarat) and Nagpur InlandContainer Depot (NICD) (Nagpur) (Fig. 1). By under-standing the importance of international trade andexport to India’s economic development, the studywill focus on cargoes produced within these tworegions being exported to the international market,with shippers serving as the main consumers of the ser-vices offered by the stated dry ports.6

There are several reasons explaining the abovechoices. First, these regions are generally accepted asimportant industrial clusters within India, well knownfor the production of merchandize (for example, tex-tiles, pharmaceuticals, chemicals, etc.), agricultural andmineral products (for example, cotton, soya, rayon,iron, etc.) for exports which is critical for the well-being of India’s national economy. For instance, in2008, textile production contributed about 14%, 4%and 17% to India’s industrial production, gross domesticproduct and national export earnings, respectively(MINISTRY OF TEXTILES OF THE GOVERNMENT OF

INDIA, 2009). Hence, perhaps unsurprisingly, thesecase study regions are also proximate to important sea-ports, notably Jawaharlal Nehru Port ( JNP), which

handles about 80% of India’s total container throughputs(INDIAN PORTS ASSOCIATION, 2009). Furthermore,despite such clear significance within multimodalsupply chains, as mentioned above works on dry portswithin developing economies have been foundwanting, thus leaving a large research gap which is yetto be filled. By investigating the case study regions, itenables this study to address a relevant issue that doesnot receive sufficient coverage, particularly in thecontext of developing economies. Thus, although byno means fully comprehensive in generalizing India,not to mention developing economies as a whole, thecase study regions can still provide representativeexamples illustrating the morphology of distributioncentres in developing economies.

Ahmadabad, Gujarat

With a population of 5 million spread over 50 km2,Ahmadabad is located within the state of Gujarat. Theregion is mostly well-known for its textile mills datingback to the last century, spreading over four major pro-duction bases nearby, notably Gondia, Mehsana, Vatwaand Viramgam. Apart from textiles, other industrieshave also flourished around the region, for example,pharmaceuticals, paper, sheet glass, chemicals, agricul-tural products (such as oilcake and edible oil), etc. Itslocal dry port, Sabarmati Inland Container Depot(SICD), located about 4 km from Ahmadabad’s citycentre and spread over 10 hectares, is connected byroad and railroads to JNP, Mundra and Pipavav.According to industrial information,7 67%, 20% and13% of cargoes are shipped out through these ports,respectively.

Nagpur, Maharashtra

With a population of 3 million spread over 40 km2,Nagpur is located within the state of Maharashtra. It isa region rich in forestry and mineral resources, andthus the industrial outputs from this region mainlyconsist of agricultural and mineral products, forexample, cotton, soya, rayon, iron and steel, aluminium,etc., spread over several production bases, notably Buti-bori, Gondia, Raipur and Wardha. Nagpur’s local dryport, Nagpur Inland Container Depot (NICD), islocated about 12 km from Nagpur’s city centre.Despite the fact that both JNP and Vishakhapatnamare approximately equidistant from Nagpur (and alsoconnected by railroads), nearly all cargoes (98%) areshipped out through JNP.

METHODOLOGY

The foremost concern of locational analysis is the impe-diments to the movement occurring due to spatial sep-aration, which often involves a cost, be it economic,


Dow

nloa

ded

by [

Hon

g K

ong

Poly

tech

nic

Uni

vers

ity]

at 2

3:40

29

May

201

2

financial or social. This has certain important impli-cations, as the methodology in searching for this opti-mally located dry port should be multifaceted andreflect the dynamics of human spatial interactions,notably economic activities and the pace of regionaldevelopment. The analytical framework can be foundin Fig. 2.

In the following analysis, several (sensible) assump-tions have been made. First, given the understandingthat dry ports are more than just cargo distributioncentres which also serve additional functions in facilitat-ing the shipment process, as well as the fact that Indianshippers largely consist of medium and small-sized firms,as also agreed by ISLAM et al. (2005), it is practicallyimpossible for them to get around dry ports and shiptheir cargoes to seaports directly, and so not using adry port is not an option. Also, given the existence ofsignificant overcapacities in both case study regions,modelling is based on a single- (rather than a multi-)facility location model. In other words, during the simu-lation, only one dry port should be called on each time

(Fig. 2). Finally, this study’s focus lies on containerizedcargoes, and thus the 20 foot-equivalent unit (TEU) isused as the unit of measurement.8 In terms of method-ology, it can be divided into two steps: economic mod-elling and in-depth interviews.

Centre of gravity (CoG) model

Various analytical techniques have been applied to dealwith the facility location problem, for instance, heuristicalgorithms, simulation methods, mixed-integer pro-gramming, factor-rating systems, linear programming,analytic Delphi model, artificial neural network, fuzzyset theory, etc. This study starts by regarding the casestudy regions as facility location problems, and thusstarts with a normative analytic approach by undertakingquantitative economic modelling by applying the CoGmodel, a static continuous model (BALLOU, 1999), toinvestigate the physical locations of particular facilitieswithin a system, taking theweighted average of the econ-omicmasses surrounding a single facility (in this case a dry

Fig. 1. Case study regions

Fig. 2. Analytical framework


Dow

nloa

ded

by [

Hon

g K

ong

Poly

tech

nic

Uni

vers

ity]

at 2

3:40

29

May

201

2

port) and finding its theoretical best location. The CoGcan be understood as an imaginary point where all theweights of the object are considered to be concentratedthrough that point, and hence there is no momentumarm to unbalance the objective concerned (Fig. 3).

The application of the CoG model in spatial inter-action analysis has been widely applied by economicgeographers (ULLMAN, 1956; MIKKONEN andLUOMA, 1999; WALKER, 2006), of which considerabletheoretical endeavours and empirical results have sup-ported its relevance in explaining human spatial inter-action with reasonable accuracies (for instance,CAREY, 1858; REILLY, 1931; ZIPF, 1946; STEWART,1948; CASEY, 1955; CARROTHERS, 1956; SCHNEI-

DER, 1959). Given the understanding of the influencesof centrality and intermediacy in deciding the locationof distribution centres (see the second section), despitesome limitations, for example, the inability of incorpor-ating total logistics costs, its static nature, etc., the CoGmodel is still a highly relevant tool for this study consid-ering its ability to identify the characteristics of centralplaces and the spatial arrangements which give rise totransportation between them. Such a wide applicationof the CoGmodel is not restricted to human geography,but it can also be used in other disciplines within thesocial sciences, notably urban planning, regionalscience, demographics, marketing (MIKKONEN andLUOMA, 1999; SHEN, 2004), retailing (CLARKSON

et al., 1996), as well as transport planning and trip distri-bution (BLACK, 1997). Such relevance is also enhancedby the existence of significant overcapacities of dry portswithin the case study regions (as mentioned above), aswell as the fact that cargo consolidation serves as oneof the most important functions of dry ports within

these regions. As supported by CHAPMAN andWALKER (1991), with economies of scale and increasinggeographical concentration taking place, it is sensible toapply the least-cost approach of identifying a singleoptimal location for the industrial plant concerned.

The CoG model assumes that the object concernedcarries a collection of weights (W1, W2, …, Wn) andthe relevant distances from these weights to an imagin-ary CoG (d1, d2,…, dn). The forces being exerted on thisobject concerned at various locations can be reflected bythe following formulation:

F1 = W1 ∗ d1,F2 = W2 ∗ d2, . . . , Fn−1

= Wn−1 ∗ dn−1, Fn = Wn ∗ dn (1)

To ensure that the object concerned can be balanced,the CoG should be placed where the followingequation is satisfied:

F1 + F2 + · · · + Fn−1 + Fn = 0 or

∑ni−1

Fi∗di = 0 (2)

Equation (2) implies that all weights within the objectconcerned are mutually related, and that any changeswithin one direction can be reflected in other direc-tion(s) of the other weight(s) so as to maintain thebalanced status. Also, as the force being exerted from aweight depends on the weight and the distance to theCoG, an increase in weight can be reflected by thedecrease in that particular distance so that the object

Fig. 3. Centre of gravity (CoG) modelSource: Redrawn by the authors based on THAI and GREWAL (2005)


Dow

nloa

ded

by [

Hon

g K

ong

Poly

tech

nic

Uni

vers

ity]

at 2

3:40

29

May

201

2

can maintain its balance. Therefore, the CoG will shiftand become more proximate to a new area withheavier weighting. As a diagnosis from equation (2), tomaintain the status quo, the CoG model would rec-ommend the dry port concerned to locate to a sitenear major weights so as to offset other lighter weightsat further distances (THAI and GREWAL, 2005). In thisstudy, where numerous dry ports have been establishedthroughout India, it means that the CoG model can beused identify the dry ports with the highest potential ingenerating maximum flows, and thus business, based onvolumes of cargoes generated within the productionbases and the lowest possible transportation costs(based on distances) (hereinafter called the ‘optimal dryport’). Given the nature of dry ports as mentionedabove, the use of transportation costs based on marketdemands as the basic analytical unit is a sensiblechoice. In other words, this step attempts to identifythe ‘best’ location from a cost view. The CoG can beexpressed as the following formulation:

MinG =∑i

ViRidi (3)

where G is the total transportation cost; Vi is the cargovolume at i; Ri is the transportation rate at i; and di is thedistance at i from the facility to be located. The optimallocation can be simulated by applying two equationsbased on the coordinates of different related facilities,as illustrated in the following formulations:

X =∑iViRixi/di

∑iViRi/di

(4)

Y =∑iViRiyi/di

∑iViRi/di

(5)

where X and Y are the latitude and longitude of theoptimal location, respectively; and xi and yi are the coor-dinate points of source and demand points. On the otherhand, the distance (di) is estimated by applying the fol-lowing formulation:

di = k��(xi − X)2 + (yi − Y )2

√(6)

where k is a scaling factor in converting one unit of acoordinate index to a more common distance unit (inthis paper, kilometre, or km, will be used). When apply-ing the above formulations, several steps will beinvolved. The first step involves the determination ofthe X- and Y-coordinate points for each source anddemand point, along with point volumes and lineartransportation rates, followed by the approximation of

the initial location from the CoG formulae by omittingthe distance term (di), as follows:

X =∑iViRixi

∑iViRi

(7)

Y =∑iViRiyi

∑iViRi

(8)

UsingX and Y, di can be calculated by applying equation(6) (the scaling factor k does not need to be used at thispoint). Then, by substituting di into equations (4) and(5), the revised X- and Y-coordinates can be solved anddi will be recalculated based on the revised X- and Y-coordinates. The above steps should be repeated untilneither of theX- andY-coordinates change for successiveiterations (or change so little that continuing the calcu-lations is not practical) and, finally, the total costs forthe optimal location should be calculated by applyingequation (3). While not ignoring the fact that not allcargoes are directed towards local dry ports, that is,SICD and NICD, as warned by CHAPMAN andWALKER (1991) where potentially profitable sites oftenexist around an optimal location, given the insignificanceof these alternatives in terms of cargo volumes, it is sen-sible to assume that all cargoes are currently passingthrough the local dry ports as indicated above. Thispoint will be further discussed in the next section.

Freight trains, instead of trucks, can be used, as long asthe annual cargo size along this route reaches the annualthreshold of 453600 metric tonnes,9 as well as this routebeing supported by railroads to the gateway seaports.Based on industrial information, the unit shipmentcosts of cargoes carried by trucks and trains (providedthat the threshold can be reached) is US$0.25 and US$0.15 per metric tonne per km, respectively. All othernecessary data required for the simulation were obtainedfrom various industrial sources.

In-depth interviews

So far, the CoG-simulated optimal location is only basedon cost minimization. This assumption does not necess-arily work for developing economies as their develop-ment often concentrates within a few particularregions accompanied by underdeveloped transportinfrastructure (cf. KRUGMAN, 1991; FUJITA andKRUGMAN, 1999; TODARO and SMITH, 2005) while,as mentioned above, one should also not overlook thepotential impacts of government policies on developingeconomies. As noted by MARTINSONS and HEMPEL

(1998), these factors encourage firms towards informal,relationship-based business practices which can divert adry port away from the optimal location. Thus, to


Dow

nloa

ded

by [

Hon

g K

ong

Poly

tech

nic

Uni

vers

ity]

at 2

3:40

29

May

201

2

understand the morphology behind, apart from quanti-tative modelling, in-depth interviews with carefullychosen industrial stakeholders were conducted (herein-after called ‘interviewees’) consisting of dry port oper-ators, government officials and shippers, all of whompossessed extensive professional experience within theindustry and were authorized to make key decisionsfor their respective institutions when the interviewswere conducted.

The core objective of conducting such interviews wasto understand the reasons behind explaining the currentlocational characteristics of dry ports, so as to address anyanomalies arising from modelling, especially since quan-titative methodology alone, in some cases, may not givethe most feasible location (MACCORMACK et al., 1994),as reflected by the existence of various works applyingqualitative methods to address the location of distri-bution centres (for example, LARSEN, 2001; LEE andHOBDAY, 2003; OUM and PARK, 2004; LU andYANG, 2006). Thus, the collection of qualitative data isimportant, especially given the lack of serious attemptsto explain the morphology of dry ports in developingeconomies, as well as the dynamics behind it.10

SIMULATED RESULTS

Ahmadabad

Cargoes originate from three production bases(Mehsana, Vatwa and Viramgam) and are transportedto three gateway seaports ( JNP, Mundra and Pipavav)via a dry port in between (Fig. 1). By applying theCoG model, the optimal dry port’s location can be

simulated, as illustrated in Fig. 4. According to the simu-lated results, Ahmadabad’s local dry port, SICD, is situ-ated close to the CoG-simulated optimal location.Although the optimal dry port is located away fromSICD in a south-westerly direction, in terms of distanceit is located less than 40 km between the two. Hence, inthis case, it seems that the CoG model has effectivelyassessed the optimal location of the dry port.

Nagpur

Cargoes originate from four production bases (Butibori,Gondia, Raipur andWardha) and are transported to twogateway seaports ( JNP and Visakhapatnam) via a dryport in between (Fig. 1). By applying the CoG model,the optimal dry port’s location can be simulated, as illus-trated in Fig. 5. In accordance with the simulated results,Nagpur’s local dry port, NICD, is far from the CoG-simulated optimal location, which lies more than170 km away from NICD towards JNP, which is themajor gateway seaport for this region. Hence, in thiscase, the CoG model seems ineffective in identifyingthe realistic situation.

Results indicate that Ahmadabad’s local dry port,SICD, is located close to its respective CoG-simulatedoptimal dry ports. However, this is not the case forNagpur, where its local dry port, NICD, lies nearly200 km away from the CoG-simulated optimal site,which is relatively distant from the major productionbases. Nevertheless, in both cases, existing dry portsshare important common characteristics, of whichtheir locations are proximate to their respective pro-duction bases, while the optimal dry ports simulated

Fig. 4. Location and coordinates of Ahmadabad’s optimal dry port


Dow

nloa

ded

by [

Hon

g K

ong

Poly

tech

nic

Uni

vers

ity]

at 2

3:40

29

May

201

2

by the CoG model are situated farther away from theirrespective production bases (Table 1). Hence, the con-tradictions between optimal and realistic locationsseem to indicate that the CoG model has underesti-mated the friction of distances, where dry ports needto be located closer towards their respective productionbases, that is, central places, so as to enable them tobecome self-sustainable.

The results also suggest that production bases createstrong forces pulling dry ports away from optimallocations towards them. Of course, this does not implythat there are currently no dry ports established at (orvery near to) the optimal locations. However, the situ-ation is that dry ports currently established at optimallocations are lying outside the profitability margin andare thus (economically) unsustainable. It is clear thatlocal dry ports are much more competitive in terms ofgetting cargoes from the surrounding region than theirmore distant counterparts (but very close to theoptimal location), for example, Ankleshwar InlandContainer Depot (ICD), Gandhidham containerfreight station (CFS) (both versus SICD), BhusawalICD, and Daulatabad ICD (both versus NICD), etc.

According to CONCOR (2009), nearly all thecargoes generated from the production bases (greaterthan or equal to 90%) are exported via their respectivelocal dry ports, that is, SICD and NICD. Indeed,those dry ports which are located close to the optimallocation (as mentioned above) persistently had fewerthan 100 TEUs of annual container throughputs,while a number of them were even forced to shutdown due to their inability to attract any users. Suchresults indicate that when the spatial pattern of econ-omic masses is uneven, the dry ports should be locatedvery close to the dominant economic mass (or a groupof economic masses) that are geographically proximate.These results lead to a suspicion that the normative,least-cost approach is an insufficient specification,which is in stark contrast to Western, advanced econ-omies, where it is noticed that most dry ports are nowa-days located at intermediate places, often along theintersections of transport corridors (RAHIMI et al.,2008). Thus, an anomaly exists where the pullingforce of centrality is more influential than intermedi-acy,11 supporting the notion that the impacts of differenttypes of costs on the locational choice of facility userscan be diversified (BEHRANS et al., 2007).

DISCUSSION

In this respect, based on the interviews conducted by theauthors (see the fourth section), detailed analysis wasundertaken and several important factors were ident-ified. First, they are related to the nature of thecargoes and their values. Within the case studyregions, most of the products entirely comprised volu-minous, non-highly time-sensitive freight (see thethird section) and, thus, in general, their values cannotdiminish the significance of transport costs within the

Table 1. Distances between existing and simulated dry portsand their respective production bases (km)

Productionbase

SICD(Gujarat)

NICD(Nagpur)

Simulated optimaldry port

Viramgam 56.81 n.a. 59.30Mehsana 66.07 n.a. 99.78Vatwa 10.18 n.a. 35.52Gondia n.a. 133.53 169.72Butibori n.a. 23.82 152.53Wardha n.a. 65.13 106.92Raipur n.a. 264.25 421.00

Note: n.a., Not available.

Fig. 5. Location and coordinates of Nagpur’s optimal dry port


Dow

nloa

ded

by [

Hon

g K

ong

Poly

tech

nic

Uni

vers

ity]

at 2

3:40

29

May

201

2

manufacturing process (CHAPMAN and WALKER,1991). Such a pattern is not dissimilar to Losch’smodel where distant facilities would result in highertransport costs, and thus the quantity demand for suchfacilities would decline with an increase in distance(LOSCH, 1954), which is especially relevant when indus-tries producing low-value, widely produced consumergoods are involved (FOUST, 1975). Indeed, simulatedresults even suggest that the explanatory power of theCoG model is directly proportional to cargo values, asillustrated by Nagpur (mainly producing raw materialssuch as cotton, soya and rayon) versus Ahmadabad(mainly producing comparative higher-valued manufac-tured products such as pharmaceuticals, paper, sheetglass and chemicals).

Second, the establishment of such a phenomenonwas enhanced by the need for flexibility and bettercontrol. In India, although the number of shippers inthese regions is fairly large, the average quantity ofcargo exported by each shipper is quite small, whichrequires consolidation to fill the containers. Forexample, in India, decentralized hosiery and the knittingsectors form the largest section of the textile industrieswithin the country (MINISTRY OF TEXTILES OF THE

GOVERNMENT OF INDIA, 2009). Hence, in the casestudy regions, shippers often prefer local dry ports soas to obtain better information and they find it mucheasier to tackle upcoming challenges requiring quickor immediate responses. The importance of flexibilityhas been emphasized by SCHOENBERGER (1997) andHALL (2004), especially given the increasing uncertaintyarising from technological changes and global compe-tition. In this respect, shippers feel more secure byretaining control over their consignment until the lastmoment, which also enables them to negotiate betterfreight rates. Moreover, shippers have establishedmuch closer relations with personnel within local dryports, which often saves significant resources andallows them to gain important knowledge by reducingrisks and uncertainties (VICENTE and SUIRE, 2007)due to shared culture, norms, attitudes, values andexpectations (GERTLER, 2001) resulting in the gener-ation of trust, notably documentary clearance, custominspection and preferential treatment. QuotingDUCRUET et al. (2009), such non-economic consider-ations often make residents of the region ‘stickier’ thanever. Indeed, this scenario strongly suggests that localdry ports, that is, SICD and NICD, are actually auxiliaryindustries forming parts of the agglomeration process ofrespective production bases, where the latter are under-going spatial clustering and hence gradually evolvinginto central places, especially given the relative smallsizes of shippers, and thus often unable to internalize ser-vices necessary for the exportation process (WEBBER,1972). Non-local dry ports, on the contrary, areunable to benefit from this agglomeration process dueto their lack of ‘economic connections’ (VICENTE andSUIRE, 2007) with the production centres. As suggested

by GERTLER (2001), extensive interactions betweenstakeholders can be limited if the distance betweenthem is too great where shippers tend to take the firstalternative which can satisfy (rather than maximize)their expectations, that is, the most geographically prox-imate dry port.

Third, the concept of dry ports, especially the per-ception of their roles and functions, and thus the ration-ale behind their construction, is fundamentally differentfrom Western, advanced economies. The philosophybehind the establishment of dry ports in these econom-ies, notably the United States and the European Union,is either as a mechanism to relieve seaport congestion(like the Port of Los Angeles in California) (SLACK,1999; ROSO, 2008; ROSO et al., 2009; RAHIMI et al.,2008), as part of the development process in comple-menting seaport regionalization, facilitating inland dis-tribution and better control of supply chains (HESSE

and RODRIGUE, 2004; NOTTEBOOM and RODRIGUE,2005; ROSO et al., 2009), or as part of the migrationprocess of the logistics industries providing value-added logistics services (ISLAM et al., 2005; NOTTE-

BOOM and RODRIGUE, 2007), or from urban to moreremote areas due to expensive land prices and land-userestrictions (CHAPMAN and WALKER, 1991;DUCRUET, 2007; HESSE, 2007) citing environmentaland social issues. Clearly, existing works indicate thatdry ports established in these economies are closelylinked to the supply chains (hereinafter called ‘supplychain-oriented’ dry ports). As a consequence, the estab-lishment of dry ports is highly specific and predomi-nantly demand driven; they are required to providehigh-end, value-added services, and are thus neededto locate in intermediate sites proximate to both portsand production centres, with efficient transport connec-tions in between. Thus, apart from simple (de-)consoli-dation centres, at the same time dry ports are required toprovide high-end, value-added services so as to comp-lement such trends, while they also need to be situatedin intermediate locations which are close to both sea-ports and production centres with highly efficient trans-port connections between different nodes. QuotingNORTH (1974), such functional changes serve as pullfactors to encourage dry ports moving to more periph-eral areas so as to serve new geographical markets.

Such specific demand for dry ports in the case studyregions is clearly lacking where dry ports are simplyregarded as transport and consolidation centres ratherthan components of multimodal supply chains.Indeed, apart from these conventional functions, theIndian government also expects dry ports to play cataly-tic roles in stimulating economic development of thesurrounding areas (UNESCAP, 2006), and hence theirestablishment is based on the perception that dry portscan be self-sustained and bring in perpetuity effects(BUTTON, 2004) to surrounding regions. This clearlyindicates that their establishment is based on the percep-tion that the provision of supply can stimulate demands,


Dow

nloa

ded

by [

Hon

g K

ong

Poly

tech

nic

Uni

vers

ity]

at 2

3:40

29

May

201

2

especially in view of the proposed implementation ofestablishing special economic zones throughout India.Such philosophy has explained the major motivationbehind the incredible number of dry ports beinginstalled throughout different parts of the countrywithin a short period of time. They are often establishedwithout the guarantee of genuine demands or clear rolesand functions, contributing to their high degrees ofsimilarities and immobility, not to mention theirinabilities to provide value-added services to shippers.As confirmed by interviewees, the lack of clearlydefined objectives has resulted in the establishment ofnumerous simple, generic (in terms of layout, facilitiesand services provided) dry ports throughout thecountry. The lack of demand, the high degree of simi-larities and the immobility of dry ports have paralysedthe competitive strength of those which are notlocated near the production bases because they areunable to improve their attractiveness through spatialrearrangement or product differentials (HOTELLING,1929; HOOVER and GIARRATANI, 1985) in the faceof the non-financial benefits that shippers can enjoyby using local dry ports as mentioned above.

Last but not least, the establishment of Indian dryports from simple warehousing and consolidationnodes to logistics centres is also hindered by the exist-ence of abundant, but mainly medium and small-sized, shippers within the country. For example,within the European Union the process of port regiona-lization, and thus the establishment of dry ports, is notnecessarily motivated by seaports, but also by logisticsdecisions and subsequent actions of (large) shippersand third-party logistics providers, especially in therecent trend of the development of logistics parks sur-rounding the dry port areas (NOTTEBOOM and RODRI-

GUE, 2005; HESSE, 2007). In contrast, given their sizesand the nature of their manufactured merchandize, theprerequisites for stimulating the growth of value-addition industries, or integrating the key business pro-cesses from the end user to the original suppliers asrequired by efficient supply chain management, simplydo not (at least explicitly) exist, and thus unsurprisingly,unlike their counterparts in advanced economies, ship-pers within the case study regions find it difficult (ifthey ever attempt) to play more innovative roles tobuild up the centrality of dry ports, thus further raisingthe difficulties in overturning the status quo. Aspointed out by SIMON (1959) and later by GUDGIN

(1978), such satisfying behaviours reflect the limitations,as well as the tendency to minimize risks, of shipperswhen they make dry port choices. Hence, the viewthat installing dry ports at carefully chosen locationsthat can automatically enable them to evolve fromsimple warehousing functions to multimodal distri-bution centres12 seems to be just an inaccurate assess-ment/lack of a thorough understanding of the realisticsituation. Thus, the threshold in self-sustaining econ-omic growth has not been achieved.

To sum up, the existence of significant variationsbetween the simulated optimal and realistic locationshave highlighted the importance of centrality in thedecision of shippers to use dry ports within the casestudy regions, where the pulling force of intermediacyis virtually non-existent. The inability of non-local dryports in attracting business also confirms the propositionthat the existence of facilities by themselves cannotstimulate regional economic development. Hence,rather than being ‘supply chain-oriented’, as exempli-fied by the case study regions and the works undertakenby NG and GUJAR (2009a) on Southern India, dry portsin developing economies seem to be more ‘cluster-oriented’, reflecting their dynamic interrelation withthe industrial clusters surrounding them. Such phenom-enon should not be too surprising, given the usual lackof efficient supply-chain management within develop-ing economies as discussed above. These results alsosupport Hall’s arguments where diversifications existthrough a process of regional institutional transform-ation (HALL, 2003, 2004).

Thus, in terms of policy implications, the case studyregions offer evidence in support of the propositionthat the Indian government’s strategy of stimulatingregional economic development by establishingsimple, generic dry ports around the country fails tolive up to its expectations. To enable them to becomethe catalysts for regional economic development, gov-ernments should undertake initiatives to enhance theirattractiveness by stimulating demands for such facilitiesand define the objectives and roles for each dry port,rather than just the physical installation of similar facili-ties. In this respect, to minimize the wastage ofresources, when deciding whether/where dry portsshould be established, decision-makers should firstaddress an important question – whether theircountry/region possesses the necessary conditions toestablish supply chain- or cluster-oriented dry ports.

CONCLUSIONS

By investigating dry ports in two case study regions, thispaper investigates the locational characteristics of inlanddistribution centres in North-west India, a globallyimportant developing economy. By applying the CoGmodel, the optimal dry ports in two major industrialregions within North-west India have been investi-gated, with special attention paid on whether the simu-lated locations are matching the current solutions and, ifnot, investigating the major reasons for such anomaly.

The results indicate that dry ports which locate closeto the industrial clusters are often preferred by users,despite the fact that the optimal dry ports are notalways proximate to the industrial clusters. This is instark contrast to the case inWestern, advanced econom-ies, where it is observed that dry ports are mainly locatedat intermediate (rather than central) places, in many cases


Dow

nloa

ded

by [

Hon

g K

ong

Poly

tech

nic

Uni

vers

ity]

at 2

3:40

29

May

201

2

along transport corridors (RAHIMI et al., 2008). Themain reasons behind this include cargo costs, high-quality connectivity, the provision of value-added ser-vices provided by the operators (such as procurement,assembly, production, reparation and reverse logistics,which are lacking in Indian dry ports), and, most impor-tantly, a different understanding of dry ports. In otherwords, the results seem to indicate that dry ports indeveloping economies are, in fact, cluster (rather thansupply chain) oriented, given that efficient supplychain management is often lacking within theseeconomies.

The results have also confirmed the main propositionthat where the spatial patterns between different types ofeconomies can be diversified due to fundamental geo-graphical differences and paces of regional development.As highlighted by GERTLER (2001) and NG and PALLIS(2010), specific backgrounds, culture and local interestscan lead to different interpretation of even the sameconcept (in this case, the understanding of dry ports),and thus has warned against the attempt of imposinggeneric solutions onto different regions even whenaddressing a similar problem. This course is agreedupon by BADYINA and GOLUBCHIKOV (2005) andBERKOZ and TURK (2008) who note that local practicescan lead to outcomes which contrast with what shouldbe expected, as witnessed in this study’s analytical out-comes. Quoting ZIMMER (1996), rather than just phys-ical settings (such as establishing simple, generic dryports around the country by the Indian government),an ideal terminal should also provide services meetingdedicated business requirements within specificmarkets. More importantly, the results also challengethe claims by AMIN (2000) on the relative importanceof relational/organizational over geographical proximi-ties, where the proposition that interactions workbetter when stakeholders are geographically proximatestill applies to developing economies.

This implies that the locational characteristics of dis-tribution centres are the outcome of interaction andcompromises between competing forces, where theirsurvival and competitiveness reflect the dynamicsbetween different exogenous factors and stakeholders.Citing BEHRANS et al. (2007), both ‘transport’ and‘non-transport’ costs always have certain impacts onthe spatial pattern of infra- and superstructures, includ-ing distribution centres. As illustrated in the case studyregions, competitive dry ports are proximate to theindustrial clusters reflecting the social nature of whichmarkets are constructed and governed (POLANYI,1944), of which it, at least, partially satisfies the influenceand competition between different forces other thanmonetary costs, notably the value of cargoes, flexibilityand better control, as well as different philosophy inthe roles, functions and orientation of dry ports.

Of course, this paper is by no means fully compre-hensive in generalizing the whole picture of theIndian dry port sector, not to mention developing

economies as a whole. Nevertheless, the authorsbelieve that it has provided insight into the forces thataffect the spatial characteristics of distribution centresand has added significant value in enhancing an under-standing of the concepts of location, the interactionbetween different forces in affecting the how suchcentres should be located, as well as the policy impli-cations. Although this study makes no attempt to quan-tify the values of non-economic parameters, it hasplayed a significant role in making this happen in theforeseeable future by identifying such parameters, theirnature and significance, and establishing comparableresults based on empirical evidences from differentcase study regions, thus further calibrating and improv-ing the model during the process. These results alsoaddress the debate about whether there is a develop-mental need, or pure coincidence, that developingeconomies have generally an uneven distribution ofeconomic masses. Perhaps most importantly, by challen-ging the conventional wisdom on the supply chain-oriented nature of dry ports which does not necessarilywork for developing economies (where the case studyregions have illustrated), this paper has enriched the lit-erature by highlighting the risks in attempting to imposegeneric solutions to different regions due to geographi-cal diversification and paces of development. That said,it is also important to bear in mind that the optimallocations of industries can shift through time(CHAPMAN and WALKER, 1991) due to changing cir-cumstances, and so do the roles, functions and orien-tations of dry ports. In other words, the pullingstrength of centrality and intermediacy is dependenton different developmental stages within particulareconomies, implying the significance of how strategiesand policies are implemented by decision-makers.

In terms of policy implications, the study indicatesthat the notion for governments just to establish distri-bution centres (and facilities) throughout the countryand then to expect them automatically to become thecatalyst in promoting regional economic developmentis inadequate, as illustrated by the case study regionswhere non-local, ‘intermediate’ dry ports often foundit difficult even in attracting cargo flows. Although thesignificance of an efficient transport system in catalysingeconomic growth is well noted (HAYUTH, 1987), it ishighly doubtful whether the establishment of a (chain-oriented) dry port itself can actually bring significanteconomic benefits to the surrounding regions, asdoubted by HESSE (2007) on the extent of correlationbetween the development of logistics industries andregional economic growth. As suggested from the casestudy regions, the potential of distribution centres canbe fully realized only when they are supported bygenuine demands, possessing specific roles and functionswith relevant planning and management structures, aswell as being fully inscribed within the supply chain soas to execute specific, laid-down objectives. Hence,rather than providing a generic solution, the Indian


Dow

nloa

ded

by [

Hon

g K

ong

Poly

tech

nic

Uni

vers

ity]

at 2

3:40

29

May

201

2

government should take the initiative and create afavourable economic environment, followed by theestablishment of dry port(s) with well-defined rolesand functions in fulfilling such specific demands.Indeed, it is sensible to argue that a dry port is a facilita-tor (rather than a dynamo) in regional development.

This paper can serve as a platform for further researchon similar topics in other developing economies, say,China and Brazil, to name but a few, as well as furtherinvestigations on other important issues. For example,while this study identifies the potential significance ofgovernment policies (of various levels) on the spatialmorphology of dry ports, further research is necessaryto investigate their correlations (particularly the degreeof significance). Together with previous similar worksconducted by the first author (NG and GUJAR 2009a,2009b), as well as further research, it enables one toaddress the influences of non-economic factors in affect-ing the spatial dynamics of dry ports, as well as develop-ing a general theory explaining the nature and evolutionof distribution centres in developing economies. Thiswould shed light on whether such choices show consist-ency, regardless of the developmental stage of the econ-omies, or, as the theory predicts, different attributeswould diversify the spatial dynamics depending ondifferent developmental stages.

Acknowledgements – This study was supported by theHong Kong Research Grant Council (Project Code A-PC1A). The authors would like to thank Dr César Ducruet(Université de Paris I Sorbonne, France), Dr Oleg Golubchi-kov (University of Oxford, UK), Dr Peter V. Hall (SimonFraser University, Burnaby, BC, Canada), Mr Jimmy Pun(The Hong Kong Polytechnic University, China), theEditor, and the anonymous referees for their useful commentsand advice. The usual disclaimers apply.

NOTES

1. Until 2008, The World Bank classified India as a devel-oping economy, as well as an emerging market(WORLD BANK, 2008).

2. The relevance of this statement is confirmed by conduct-ing various interviews with dry port stakeholders in India.According to anecdotal information based on the samegroup of interviewees, one reason for such an approachis the Indian government’s strategy to dominate themarket so as to control the growth of foreign-based term-inal operators.

3. This is not ignoring the existence of various worksaddressing similar issues in developing economies (forexample, DERAKHSHAN et al., 2005; ISLAM et al., 2005;LEE et al., 2008). In all these studies, however, the analysisfocuses on port or multimodal transport and/or supplychain development, while dry ports only occupy periph-eral roles.

4. An illustrative example can be found in the Indian state ofAssam, where Amingaon dry port (Inland ContainerDepot Amingaon) undertakes various critical functionssustaining the export of Indian tea (notably ‘Assam Tea’)and its competitiveness against major competitors (likeChina, Kenya and Sri Lanka), especially in view of thedownward pressure in global tea price and the rise of com-petitors (UNITED NATIONS FOOD AND AGRICULTURAL

ORGANIZATION (FAO), 2007), for example, consolida-tion, palletizing, stuffing, custom clearance, documen-tation, etc., as well as being pivotal in sustaining socialstability, given the high number of people involved dueto the industry’s labour-intensive nature. Indeed, accord-ing to industrial information, in 2009 the tea industry gen-erated over 1 million jobs annually within North-eastIndia, thus illustrating the potential contribution of dryports to the region’s economy.

5. For a comprehensive review of location models and theirapplications, see SEPPALA (2003), KLOSE and DREXL

(2005), and ZHU et al. (2010).6. Apart from shippers (rather than shipping lines and sea-

ports) being the main customers of dry port services, thetwo case study regions are also located very far awayfrom each other. Moreover, connections between thetwo case study regions are rather difficult, as any move-ments between states would trigger considerable border-crossing taxes (cf. NG and GUJAR, 2009a), not helped byexisting poor transport infrastructure (a good example isthe lack of efficient freight rail, and so only road connec-tion exists) connecting these two regions. Understandingthis, any potential competition, or ‘strategic interactions’,between SICD and NICD can be safely discounted.

7. The information was provided by interviewees. See thefourth section.

8. In India dry ports are commonly known as an inland con-tainer depot (ICD) or a container freight station (CFS).These terms very much reflect the nature of Indian dryports, that is, handling containerized cargoes.

9. It is assumed that one TEU carries 14 metric tonnes ofcargo. Based on industrial information, each train mustcarry at least 90 TEUs, that is, 1260 metric tonnes, toenable such a service to become economically viable.Thus, assuming that one train runs per day (360 daysper year), there should be at least 32400 TEUs, that is,453600 metric tonnes, available annually for shipment.

10. Although ROSO et al. (2009) briefly describe the locationsof Indian dry ports, they do notmake any attempt to inves-tigate or explain the reasons behind its spatial pattern.

11. Similarly, ISLAM et al. (2005) also point out the lack ofICDs close to industrial clusters as an obstacle to supplychain integration and multimodal transport inBangladesh. However, they did not make any attemptto explain the necessity for ICDs to be located close toproduction centres.

12. According to industrial information, from relativelysimple warehousing functions at the start, and mostother business activities situated elsewhere, many inlanddistributions centres in Europe have during their exist-ence been developed towards multifunctional establish-ments of their respective companies, providing morespecialized services.


Dow

nloa

ded

by [

Hon

g K

ong

Poly

tech

nic

Uni

vers

ity]

at 2

3:40

29

May

201

2

REFERENCES

AMIN A. (2000) Organizational learning through communities of practice. Paper presented at the Millennium Schumpeter Con-ference, University of Manchester, Manchester, UK, 2000.

BADYINA A. and GOLUBCHIKOV O. (2005) Gentrification in Central Moscow – a market process or a deliberate policy? Money,power and people in housing regeneration in Ostozhenka, Geografiska Annaler 87B(2), 113–129.

BALLOU R. H. (1999) Business Logistics Management Planning, Organizing and Controlling the Supply Chain 4th Edn. Prentice-Hall,Englewood Cliffs, NJ.

BARNES T. J. (2003) The place of locational analysis: a selective and interpretive history, Progress in Human Geography 27(1), 69–95.BEHRANS K., LAMORGESE A. R., OTTAVIAON G. I. P. and TABUCHI T. (2007) Changes in transport and non-transport costs: local vs.

global impacts in a spatial network, Regional Science and Urban Economics 37, 625–648.BERESFORD A. K. C. and DUBEY R. C. (1990) Handbook on the Management and Operation of Dry Ports. United Nations Conference

on Trade and Development (UNCTAD, Geneva.BERKOZ L. and TURK S. S. (2008) Determination of location-specific factors at the intra-metropolitan level: Istanbul case, Tijdschrift

voor Economische en Sociale Geografie 99(1), 91–114.BIRD J. (1973) Of central places, cities and seaports, Geography 58, 105–118.BLACK W. R. (1997) New approaches to modelling travel, Journal of Transport Geography 5(1), 46.BREMMER I. (2005) Managing risk in an unstable world, Harvard Business Review June, 51–60.BUTTONK. J. (2004) Economic development and transport hubs, inHENSHERD. A., BUTTONK. J., HAYNES K. E. and STOPHER P. R.

(Eds) Handbook of Transport Geography and Spatial Systems, pp. 77–95. Elsevier, Oxford.CANEL C. and KHUMAWALA B. M. (2001) International facilities location: a heuristic procedure for the dynamic incapacitated

problem, International Journal of Production Research 39(17), 3975–4000.CAREY H. C. (1858) Principles of Social Science. J. B. Lippincott, Philadelphia, PA.CARROTHERS G. A. P. (1956) A historical review of the gravity model and potential concepts of human interaction, Journal of the

American Institute of Planners 22, 94–102.CASEY H. J. (1955) Applications to traffic engineering of the law of retail gravitation, Traffic Quarterly 9(1), 23–35.CETIN I. B. and CERIT A. G. (2008) The potential of Turkey as a logistics centre between Far East and Europe: an application in

electronics industry, in Proceedings of the International Association of Maritime Economists (IAME) Conference, DalianChina, 2–4 April 2008.

CHAKRAVORTY S. and LALL S. (2005) Do local economics matter in cluster formation, Environment and Planning A 37, 331–353.CHAPMAN K. and WALKER D. F. (1991) Industrial Location, 2nd Edn. Blackwell, Oxford.CHRISTALLER W. (1933) Central Places in Southern Germany. Prentice Hall, London (reprinted in English, 1966).CLARKSON R. M., CLARKE-HILL C. M. and ROBINSON T. (1996) UK supermarket location assessment, International Journal of Retail

and Distribution Management 24(6), 22–33.CONCOR (2009) Container Corporation of India Ltd. [Home Page] (available at: http://www.concorindia.com) (accessed on 10

October 2009).COUGHLIN C. C., TERZA J. V. and ARROMDEE V. (1991) State characteristics and the location of foreign direct investment within

the United States, Review of Economics and Statistics 73(4), 675–683.DERAKHSHAN A., PASUKEVICIUTE I. and ROE M. (2005) Diversion of containerized trade: case analysis of the role of Iranian ports in

global maritime supply chain, Trasporti Europei 30, 61–76.DICKINSON R. E. (1961) The West European City. Routledge, London.DUCRUET C. (2007) A metageography of port–city relationships, in WANG J., OLIVIER D., NOTTEBOOM T. and SLACK B. (Eds) Ports,

Cities and Global Supply Chains, pp. 157–172. Ashgate, Aldershot.DUCRUET C., NOTTEBOOM T. and DE LANGEN P. (2009) Revisiting inter-port relationships under the new economic geography

research framework, in NOTTEBOOM T., DUCRUET C. and DE LANGEN P. (Eds) Ports in Proximity: Competition and Coordinationamong Adjacent Seaports, pp. 11–27. Ashgate, Aldershot.

FLEMING D. K. and HAYUTH Y. (1994) Spatial characteristics of transportation hubs: centrality and intermediacy, Journal of TransportGeography 2(1), 3–18.

FOUST J. B. (1975) Ubiquitous manufacturing, Annals of the Association of American Geographers 65, 13–17.FRIEDMAN J., GEDRLOWSKI D. A. and SIBERMAN J. (1992) What attracts foreign multinational corporations? Evidence from branch

plant location in the USA, Journal of Regional Science 32(4), 403–418.FUJITA M. and KRUGMAN P. (1999) The Spatial Economics: Cities, Regions and International Trade. MIT Press, Cambridge, MA.GERTLERM. S. (1995) Being there: proximity, organization and culture in the development and adoption of advanced manufactur-

ing technologies, Economic Geography 71, 1–26.GERTLER M. S. (2001) Best practice? Geography, learning and the institutional limits to strong convergence, Journal of Economic

Geography 1, 5–26.GUDGIN G. (1978) Industrial Location: Process and Regional Employment Growth. Saxon House, Farnborough.HALL P. V. (2003) Regional institutional convergence? Reflections from the Baltimore Waterfront, Economic Geography 79(4),

347–363.HALL P. V. (2004) Persistent variation: flexibility, organization, and strategy in the logistics of importing automobiles to the United

States, 1980–99, Environment and Planning A 36, 529–546.HAYUTH Y. (1987) Intermodality: Concept and Practice. Lloyd’s of London Press, London.


Dow

nloa

ded

by [

Hon

g K

ong

Poly

tech

nic

Uni

vers

ity]

at 2

3:40

29

May

201

2

http://www.concorindia.com

HEAVER T. D. (2002) The evolving role of shipping lines in international logistics, International Journal of Maritime Economics 4, 210–230.

HESSE M. (2004) Land for logistics: locational dynamics, real estate markets and political regulations of regional distribution com-plexes, Tijdschrift voor Economische Sociale Geografie 95(2), 162–173.

HESSE M. (2007) Logistics and spatial planning: regional development outcomes, spatial planning needs. Technical Note presentedat the ‘Interim’ Workshop at the University of Applied Sciences (TFH) Wildau, Wildau, Germany, 13 September 2007.

HESSE M. and RODRIGUE J. P. (2004) The transport geography of logistics and freight distribution, Journal of Transport Geography 12(3), 171–184.

HOOVER E. M. and GIARRATANI F. (1985) An Introduction to Regional Economics 3rd Edn. McGraw-Hill, New York, NY.HOTELLING H. (1929) Stability of competition, Economic Journal 39, 41–57.HOTTES R. (1983) Walter Christaller, Annals of the Association of American Geographers 73(1), 51–54.INDIAN PORTS ASSOCIATION (2009) Indian Ports Association [Home Page] (available at: http://www.ipa.nic.in) (accessed on 7

October 2009)ISLAM D. M. Z., DINWOODIE J. and ROE M. (2005) Towards supply chain integration through multimodal transport in developing

economies: the case of Bangladesh, Maritime Economics and Logistics 7, 382–399.KLOSE A. and DREXL A. (2005) Facility location models for distribution system design, European Journal of Operational Research 162

(1), 4–29.KRUGMAN P. (1991) Increasing returns and economic geography, Journal of Political Economy 99, 482–499.KRUGMAN P. (1996) Geography and Trade. MIT Press, Cambridge, MA.LARSEN T. S. (2001) Logistics in the Öresund Region after the Öresund Link. Report Number 3120/2001. Copenhagen Business

School, Ulf Paulsson & Sten Wandel, Lund University, Lund.LEE S. W., SONG D.W. and DUCRUET C. (2008) A tale of Asia’s world ports: the spatial evolution in global hub port cities,Geoforum

39, 372–385.LEE Y. and HOBDAY M. (2003) Korea’s new globalization strategy: can Korea become a business hub in Northeast Asia?, Manage-

ment Decision 41(5), 498–510.LOSCH A. (1954) The Economics of Location. Yale University Press, New Haven, CT.LU C. S. and YANG C. C. (2006) Comparison of investment preferences for international logistics zones in Kaohsiung, Hong Kong,

and Shanghai ports from a Taiwanese manufacturer’s perspective, Transportation Journal 45(1), 30–51.LUNDVALL B. A. and JOHNSON B. (1994) The learning economy, Journal of Industry Studies 1, 23–42.MACCORMACK A. D., NEWMAN L. J. and ROSENFIELD D. B. (1994) The new dynamics of global manufacturing site location, Sloan

Management Review 35(4), 69–78.MACHARIS C. and VERBEKE A. (1999) The optimal location of intermodal terminals. Paper presented at the NECTARConference,

Delft, the Netherlands, 20–23 October 1999.MARTINSONS M. G. and HEMPEL P. S. (1998) Chinese business process re-engineering, International Journal of Information Management

18, 393–407.MIKKONEN K. and LUOMA M. (1999) The parameters of the gravity model are changing – how and why?, Journal of Transport

Geography 7, 277–283.MINISTRY OF TEXTILES OF THE GOVERNMENT OF INDIA (2009) Annual Report 2008–09. Ministry of Textiles, Government of India,

New Delhi.NG A. K. Y. (2006) Theory and structure of port competition: a case study of container transhipment in North Europe. Unpub-

lished DPhil thesis, University of Oxford, Oxford.NG A. K. Y. and GUJAR G. C. (2009a) The spatial characteristics of inland transport hubs: evidences from Southern India, Journal of

Transport Geography 17(5), 346–356.NG A. K. Y. and GUJAR G. C. (2009b) Government policies, efficiency and competitiveness: the case of dry ports in India, Transport

Policy, 16(5), 232–239.NG A. K. Y. and PALLIS A. A. (2010) Port governance reforms in diversified institutional frameworks: generic solutions, implemen-

tation asymmetries, Environment and Planning A 42(9), 2147–2167.NG A. K. Y. and TONGZON J. L. (2010) The transportation sector of India’s economy: dry ports as catalysts for regional develop-

ment, Eurasian Geography and Economics 51(5), 669–682.NORTH D. J. (1974) The process of locational change in different manufacturing organizations, in HAMILTON F. E. I. (Ed.) Spatial

Perspectives in Industrial Organization and Decision Making, pp. 213–244. Wiley, London.NOTTEBOOM T. E. and RODRIGUE J. P. (2005) Port regionalization: towards a new phase in port development, Maritime Policy and

Management 32, 297–313.NOTTEBOOM T. E. and RODRIGUE J. P. (2007) Re-assessing port–hinterland relationships in the context of global commodity

chains, in WANG J., OLIVIER D., NOTTEBOOM T. and SLACK B. (Eds) Ports, Cities and Global Supply Chains, pp. 51–66.Ashgate, Aldershot.

OUM T. H. and PARK J. H. (2004) Multinational firm’s location preference for regional distribution centres: focus on the NortheastAsian region, Transportation Research E: Logistics and Transportation Review, 40, 101–121.

POLANYI K. (1944) The Great Transformation. Rinehart, New York, NY.RAHIMI M. ASEF-VAZIRI A. and HARRISON R. (2008) An inland port location–allocation model for a regional intermodal goods

movement system, Maritime Economics and Logistics 10(4), 362–379.REILLY W. J. (1931) The Law of Retail Gravitation. G. P. Putnam’s Sons, New York, NY.


Dow

nloa

ded

by [

Hon

g K

ong

Poly

tech

nic

Uni

vers

ity]

at 2

3:40

29

May

201

2

http://www.ipa.nic.in

RODRIGUE J. P., DEBRIE J., FRÉMONT A. and GOUVERNAL E. (2009) Functions and actors of inland ports: European and NorthAmerican dynamics, Journal of Transport Geography 18(4), 519–529.

ROSO V. (2008) Factors influencing implementation of a dry port, International Journal of Physical Distribution and Logistics Manage-ment 38(10), 782–798.

ROSO V., WOXENIUS J. and LUMSDEN K. (2009) The dry port concept: connecting container seaports with the hinterland, Journal ofTransport Geography 17(5), 338–345.

RUTTEN B. C. M. (1998) The design of a terminal network for intermodal transport, Transport Logistics, 1, 279–298.SCHNEIDER M. (1959) Gravity models and trip distribution theory, Papers and Proceedings of the Regional Science Association 5, 51–56.SCHOENBERGER E. (1997) The Cultural Crisis of the Firm. Blackwell, Cambridge, MA.SEPPALA U. (2003) An Evolutionary Model for Spatial Location of Economic Facilities. Report Number IR-97-003. International Insti-

tute for Applied Systems Analysis, Laxenburg.SHEN G. (2004) Reverse-fitting the gravity model to inter-city airline passenger flows by an algebraic simplification, Journal of

Transport Geography 12, 219–234.SHERALI D. H., LOUGHANI A. I. and SUBRAMANIAN S. (2002) Global optimization procedures for the capacitated Euclidean and LP

distance multi facility location–allocation problem, Operations Research 50, 433–488.SIMON H. A. (1959) Theories of decision-making in economics and behavioural science, American Economic Review 49, 253–283.SLACK B. (1985) Containerisation, inter-port competition and port selection, Maritime Policy and Management 12(4), 293–303.SLACK B. (1999) Satellite terminals: a local solution to hub congestion?, Journal of Transport Geography 7(4), 241–246.STEWART J. Q. (1948) Demographic gravitation: evidence and applications, Sociometry 11, 31–58.THAI V. V. and GREWAL D. (2005) Selecting the location of distribution centre in logistics operations: a conceptual framework and

case study, Asia Pacific Journal of Marketing and Logistics 17(3), 3–24.TOBLER W. R. (1970) A computer model simulation of urban growth in the Detroit region, Economic Geography 46(2), 234–240.TODARO M. P. and SMITH S. C. (2005) Economic Development 9th Edn. Addison Wesley, Harlow.ULLMAN E. L. (1941) A theory of location for cities, American Journal of Sociology 46(783), 853–864.ULLMAN E. L. (1956) The role of transportation and the bases for interaction, in THOMASW. L. (Ed.)Man’s Role in Changing the Face

of the Earth, pp. 862–880. University of Chicago Press, Chicago, IL.UNITED NATIONS ECONOMIC AND SOCIAL COMMISSION FOR ASIA AND THE PACIFIC (UNESCAP) (2006) Promoting Dry Ports as a Means

of Sharing the Benefits of Globalization with Inland Locations. Report Number E/ESCAP/CMG(3/1)1, UNESCAP, Bangkok.UNITED NATIONS ECONOMIC COMMISSION FOR EUROPE (UNECE) (2001) Terminology on Combined Transport. UNECE, Geneva.UNITED NATIONS FOOD AND AGRICULTURAL ORGANIZATION (FAO) (2007) Tea-Market Developments and its Short Term Prospects.

FAO, Rome.VICENTE J. and SUIRE R. (2007) Informational cascades versus network externalities in locational choice: evidence of ‘ICT’ clusters’

formation and stability, Regional Studies 41(2), 173–184.WALKER J. L. (2006) Opening up the black box: enriching behavioral models of spatial and travel choices, Journal of Transport

Geography 14, 396–398.WEBBER M. J. (1972) Impact of Uncertainty on Location. MIT Press, Cambridge, MA.WEBER A. (1909) Theory of the Location of Industries. University of Chicago Press, Chicago, IL (reprinted in English, 1929.WEIGEND G. (1958) Some elements in the study of port geography, Geographical Review 48, 185–200.WORLD BANK (2008) Global Development Finance 2008. World Bank Group, Washington, DC.ZHU Z., CHU F. and SUN L. (2010) The capacitated plant location problem with customers and suppliers matching, Transportation

Research Part E: Logistics and Transportation Review, 46(3), 469–480.ZIMMER N. R. (1996) Designing Intermodal Terminals for Efficiency. Transportation Research Circular Number 459. Transportation

Research Board, Washington, DC.ZIPF G. K. (1946) The P1P2=D hypothesis: on the intercity movement of persons, American Sociological Review 2, 677–686.


Dow

nloa

ded

by [

Hon

g K

ong

Poly

tech

nic

Uni

vers

ity]

at 2

3:40

29

May

201

2

Locational Characteristics of Dry Ports in Developing Economies: Some Lessons from Northern India

Documents