Companies often make location decisions based on the economics of transportation

The Geographic Information System: A Tool for Locational Decision-Making Support

Locational decision making addresses the problem of finding one or more suitable sites for some activity. Inputs for solving this problem include skilled analysts, georeferenced attribute data, models for evaluating relevant data, and an ability to construct maps displaying the data and enabling identification of potential sites. Many of these tasks can be efficiently and effectively executed with the help of a geographic information system (GIS). This computer software tool supports locational decision making by furnishing an attribute data viewer table, data manipulation facilities, and graphics capabilities for displaying results as layers on a map. A GIS helps guide a decision-making progress—albeit operational, tactical, or strategic— by furnishing timely access to attribute data, speedy data manipulations, and rapid map updates. The focus is on decision support, not replacement of a decision maker. GIS, in its spatial analysis mode, is applicable to this spatial decision-making context.

3 Location and Growth

Locational decisions are usually not stand-alone choices of a firm, but are part of a broader set of firm decisions, e.g., on market areas, marketing channels, technology to be used, image, etc. Against this background transport—and thus trade—plays an important role. This fact was already recognized in the well-known Hecksher–Ohlin theory on international (or interregional) trade flows, where trade intensity between two countries is dependent on the comparative advantage of the use of factor inputs. Clearly, also the degree of homogeneity of products to be traded plays a role here. Based on the assumptions of monopolistic competition, Dixit and Stiglitz (1977) have tried to make the Hecksher–Ohlin theory more realistic, inter alia by allowing for product differentiation. In this way, both inter- and intra-industry trade flows and locational patterns can be analyzed. This integration of trade theory and location theory forms one of the main characteristics of modern growth and trade theory.

In this context there is also a need for a more dynamic perspective to spatial equilibrium phenomena. This presupposes a more integrated view on locational decisions and emerging market patterns, thus necessitating an economic growth perspective. Consequently, trade and location are endogenously determined in modern economic growth theory, as developed inter alia by Krugman (1991). This also means that factor mobility, transportation costs, and transaction costs become an integral part of modern theories on location and urban or regional growth.

In a recent study by Fujita et al. (1999), the authors offer an attempt to re-track economic geography (and regional science), while they also aim to build a ‘new economic geography’ based on a few vigorous economic principles. The authors signal a regrettable dividing line between mainstream economics and the economics of location. They aim to build a bridge on the basis of a few simple (certainly not universally valid) concepts, in particular the imperfect competition model marked by increasing returns to scale (originating from Dixit and Stiglitz 1977). In an open (multiregion or multicountry) system various types of spatial agglomeration patterns may emerge, depending inter alia on transport costs, forward and backward linkages, and immobility of resources. The authors offer a new perspective in their effort to bring general economics closer to regional economics. They illustrate their arguments by referring to core-periphery phenomena, agriculture, urban systems, city size, transportation, international trade, and industrial clustering. The book forms a clear manifestation of the rigour of solid economic analysis for the explanation of the spatial patterns and evolution of economic activity.

The two pillars of regional economics are certainly formed by agglomeration economies and generalized transportation costs. Much emphasis is made in the book on the economic analysis of urbanization and scale advantages, but less on transportation costs. In our era of ICT development where many economists advocate the ‘death of distance,’ it would have been necessary to pay more attention to both psychological and virtual distance costs, and their implications for the spatial organization of our world (see Sect. 6).

It is clear that a further integration of location theory, transportation science, trade theory, and industrial organization seems to be a necessary condition for a rigorous way forward.

Spatial search is recognized as an important construct within behavioral geography because it is the preliminary, information-acquisition phase of most locational decision processes. Spatial search is the task of identifying and investigating spatially distributed choice alternatives which serve as the basis for a locational decision. Spatial search processes range from purely mechanical, space covering procedures characteristic of remote sensing methods through complex, multistage searches drawing on indirect information sources as well as direct experience—the search for a new residence or a new place of employment, for example. No less important is the ongoing acquisition of geographic information which occurs as people move about their environment, consciously and unconsciously creating a practical knowledge base which they draw upon to inform the myriad locational decisions made during the normal course of a day—way finding and shopping decisions, for example.

4 Complete Models

While residential choice models are the most important and perhaps the most interesting, a complete model as described at the outset would be a model suite containing numerous submodels. First, land development and its uses are determined by way of locational decisions that in turn are strongly influenced by transport considerations. Putman and Echenique have taken the lead over the last three decades in driving this lesson home—at least to the transport planning community. We will assume that any complete land-use model will embody or interface with an appropriate transportation model and use a forecast of transportation system changes. Second, even the most advanced residential models suffer from some of the general difficulties discussed below, and in particular, models of residential housing supply are weaker than those of demand. Third, models are needed for all land uses, most particularly manufacturing industry and trade and services. With the increasing shift in the overall economy toward business and consumer services, these sectors and their representation in models need thorough review.

There is a modest number of complete model suites publicly available worldwide (see Wegener 1994 for a review). Putman's ITLUP and EMPAL models have been extended to METROPOLIS, which is widely used, and maintains the Lowry tradition of noneconomic constraints. Its employment location is somewhat descriptive in style. Echenique has developed an elaborate input–output framework to generate goods traffic as well as personal trips, with extensive economic constraints. Anas (1987) is continuing to develop a range of economically rigorous and ambitious models, most recently for the New York Metropolitan Transportation Commission (in progress). New work is being undertaken which we cannot elaborate here, but special mention must be made of de la Barra, who follows Echenique in the Americas, and Waddell (2000), who is approaching residential location from a microeconomic viewpoint in direct collaboration with state planning and transportation studies. Most model sets are provided on a made-to-order basis, which raises their cost and severely limits their availability.

In our judgment, the characteristic nonresidential models have various weaknesses in almost all model sets. The manufacturing sector is undergoing changes that are not yet fully understood and which may upset conventional wisdom. Business services have barely begun to be studied in the locational literature, let alone embodied in models. Retail trade is the best understood of these three fields, but inadequate use is being made of existing knowledge.

2 Antecedents and Conceptual Approaches

Formal analysis of locational conflict by American geographers began in the early 1970s as an extension of classical economic location theory. Classical theories of industrial and residential location presented a model of locational choice by autonomous units (firms or households) seeking to maximize individual utility functions within budget constraints. Political geographers such as Kevin Cox and his students at Ohio State University challenged the assumption of autonomous locational decision-making by recognizing that individual utilities are not independent of the externality effects produced by the locational decisions of others (Cox and McCarthy 1980). Conflict ensues when the utility of a locational choice is diminished by the negative externalities generated by the locational choices of others. This extension of traditional location theory shifted analysis away from individual locational choice towards a focus on the manipulation of the spatial distribution of externalities, that is, on attempts to exclude negative externalities and attract positive ones. While classical location theory addressed individual locational decision-making within a framework of economic rationality, locational conflict theory examines collective strategies for the protection and enhancement of neighborhood quality within a framework of collective political action. As a result, the focus of analysis moved from the economic to the political arena, from individual choice to collective action, and from locational decision-making to locational conflict as a politics of turf.

The view of locational conflict as turf politics, however, retains an axiomatic assumption of traditional location theory. The central assumption in this behavioral approach is that conflict arises when a locational decision perceived as beneficial by some is perceived negatively by others. While focus has shifted to collective action, the impetus for political engagement still resides in the individual's perception of positive or negative consequences associated with a proposed locational decision. The active agent is the autonomous individual whose behavior is guided by a unique calculus of perception, whether locational conflict is situated within an individual choice framework of conflicting utilities or in the pluralist politics of neighborhood change.

Situated at the level of conflicting perceptions, the behavioral approach to locational conflict rarely considers why changes occur in local areas such that conflict arises over their perceived positive or negative consequences. A structuralist approach views locational conflicts as symptomatic of fundamental contradictions situated within the basic structure of society. Locational conflicts, in this view, are merely surface manifestations of deep-seated conflicts inherent in the extant system of social organization (Cox and McCarthy 1980). Within a structural mode of explanation, negative externalities are not viewed simply as evidence of market imperfections subject to correction but rather are understood as necessary and inevitable consequences of class relations within the process of capital accumulation.

Identifying the particular structural contradictions that become manifest as locational conflict, however, is itself a subject of conflict and disagreement. In one view, the inherent contradiction between labor and capital generates contradictory orientations to neighborhoods, such that labor's attachment to place based on the neighborhood's use value conflicts with capital's investment interest based on the neighborhood's exchange value. Locational conflict, in this framework, arises from the fundamental antagonism between labor and capital with respect to neighborhood change. In another view, locational conflict is situated within the incessant competition between factions of capital, and results when the expansionary interests of mobile capital conflict with the exclusionary interests of capital already fixed in place (Plotkin 1987). For example, existing investors seeking to exclude potential new competitors may resist a shopping center developer's search for new investment sites. Bridging these two approaches is a third view in which labor's attachment to neighborhood erects a barrier to continued accumulation of capital. In response, capital encourages long-running societal processes—the homogenization of space, pervasive ideologies of materialism and consumerism, media manipulation and mass education, and increasing residential mobility, among other contributing factors—which succeed in transforming neighborhoods from communities into commodities and, consequently, transforming labor's attachment to community into an orientation based on protecting neighborhood exchange value (Cox 1981). Now locational conflict occurs when the exchange value interests of capital and labor fail to coincide as, for example, when a land developer's proposal for a shopping center generates traffic and pollution that threaten to reduce the resale value of surrounding homes.

More recently, a post-structuralist approach has emerged that situates locational conflict within the antagonistic discursive or representational strategies of contending groups vying for control over the use of space, where such control is an expression of political power. Rejecting the view of space as simply a container for action, this approach, influenced by the work of Lefebvre (1991) and others, considers the social and political process through which the meaning and use of space are constructed in particular instances. Now locational conflict is not primarily about the spatial distribution of activities or land uses nor is it simply about contrasting preferences for various activities in particular locations. Rather, locational conflict is symbolic conflict over the social distribution of power to assign meaning and uses to space (Mitchell 1992). Locational conflict symbolizes contention over whose values have standing within the political process and whose values, therefore, become expressed in the landscape. Recursively, control over the use of space symbolically constitutes the controlling group as a legitimate actor within the broader political process. Establishment of a squatter settlement within an affluent residential neighborhood despite the opposition of existing residents, for example, provides a site for housing impoverished families. Also, and perhaps more importantly, it discursively represents squatter families as legitimate occupants of space and, therefore, legitimate participants in the political process.

The evolution of locational conflict theory from classical to behavioral, to structural, and post-structural formulations parallels theoretical developments in the field of geography as a whole. Underlying each of these theoretical positions is a set of often incommensurate assumptions regarding the relative autonomy of individual actors, the primacy of market processes, the nature of the public interest, and the respective roles of structure and agency in creating geographic landscapes. In addition, the various theoretical understandings of locational conflict point to substantially different routes for its amelioration or resolution. A behavioral model of conflict based on contrasting perceptions suggests that resolution is attainable through public information and education designed to bring perceptions into agreement. A model of conflict based on the unequal spatial distribution of negative externalities suggests the use of compensation to equalize outcomes. Understanding conflict as symptomatic of deep underlying structural relations requires the elimination of structural contradictions. A focus on locational conflict as the symbolic expression of rights to create the meaning of space requires the deconstruction of opposing claims and their reconstruction in more equitable forms. Far from being merely an academic exercise, one's theoretical approach to locational conflict implicates significantly different avenues for its resolution, as discussed in the following section.

2.2 ‘-Isms,’ Schisms, and ‘Turns’

Although some promoted this ‘new geography’ as surpassing its predecessor, there was no complete revolution to a new hegemonic paradigm in the Kuhnian sense. Not all geographers were won over, and some continued to teach and research in their established ways—sustaining the traditions of historical, cultural, and regional geography which were foundations for later responses to the ‘shock of the new.’ Those responses criticized spatial science on two main grounds. The first attacked location theories for narrow economism, for privileging profit-maximizing (translated geographically as transport-cost-minimizing) as a determinant of decision-making and so denigrating the wide range of cultural and other influences on behavior. Spatial science, it was claimed, removed free will by assuming that humans react to stimuli in predetermined ways while its elevation of space as the predominant influence on locational decisions led to a derogation of place. Some critics explored alternatives to positivism—such as phenomenology, idealism and existentialism—in work categorized as humanistic geography.

The second set of criticisms concerned spatial science's approach to explanation. Location theories, it was argued, could not account for the major contours of the geography of development and underdevelopment at various spatial scales, of inequality and injustice, wealth, power, and discrimination. Explanation required understanding how economic systems, especially capitalism, work and involved more than adding the costs of transport and communications to neoclassical economic models: such explanation could be achieved through Karl Marx's writings. Harvey's (1973) collection of essays initiated a series incorporating space to Marxian thought (notably Harvey 1982) and stimulated others during a period of social tensions over civil rights and the Vietnam War. Their work became known as radical geography.

Although both critiques attracted adherents, many were as dissatisfied with each as they were with spatial science. As Gregory (1978) demonstrated, whereas some spatial scientists strayed very close to naive economic determinism, much humanistic geography approached another extreme of voluntarism, tearing the individual out of context, while some radical geography offered little more than an alternative economic determinism. Much effort was spent in the late twentieth century integrating aspects of humanistic and radical geography.

2.2 The Behavioral Challenge

The descriptive analysis of urban residential characteristics was not the only type of work in social geography done during the 1960s and early 1970s. There were those who, despite their hopes for a positivist approach to the development of theory within geography, were dissatisfied with the approaches to studying human behavior that were used widely at that time.

In the 1960s, much research in human geography drew its inspiration from models in economic geography that posited that people possessed perfect information and were rational and optimizing in their behavior. Dissatisfaction with these assumptions prompted some researchers to turn to psychological theory concerned with cognition and decision-making in order to investigate what information people had available to use in spatial decision-making and how they used it. The principal method employed in these studies was the questionnaire survey, frequently incorporating some variety of attitudinal and personality tests drawn from social psychology. The research carried out within this broad framework is known as ‘behavioral geography.’

Work in behavioral geography was concerned with the locational decision-making of firms, individuals' decisions over where to live, people's everyday movements, and their information and ideas about space and place. Gould's work on mental maps was particularly influential (Gould and White 1974). His initial research examined American undergraduates' views of each US state as a place in which to live after graduation. The results showed that individuals had a strong preference for their home area and that this preference was superimposed upon an image of a national space with marked variations in preferences. The idea of an incomplete, biased and personal vision of the spatial world conjured up by the term ‘mental map’ became part of geographers' ideas about how people made decisions about both everyday movement and migration. Information about space and place was also recognized as important within studies of residential mobility and of people's everyday travel behavior. In the former, researchers were influenced by the early work of the sociologist Peter Rossi whose 1955 book Why Families Move was subtitled A study in the social psychology of residential mobility. A large number of studies of residential mobility in different cities and countries resulted. In the latter, diary data on travel behavior were used to develop inductive theories about the way people choose and order travel destinations. Travel behavior and migration research explored how people searched for information in space and developed strategies for choosing locations.

Research using the perspectives of behavioral geography—while part of social geography—remains marginal to the field because of its focus on the individual rather than on people as interacting with and partly constituted by their social milieu. However, the impossibility of such an individualist approach is evident in much of the work actually carried out. Researchers often examined contrasts between individuals drawn from different social groups (for example, different income, class, age and, occasionally, gender groups). Others examined the influence of the resources available to individuals on their spatial cognition and so ended up investigating differences between social groups. The research is also significant to social geography because criticisms of its individualistic approach were important to the development of alternatives.

1 Introduction

Facility location, location analysis, location theory, locational decisions and siting are terms used interchangeably for the same purpose. They address a well-known classic problem referring to the placement of at least one facility (e.g. a resource or server) among several existing facilities (e.g. demand points) to serve them (Farahani & Hekmatfar 2009). Facility location is one of the very first and prominent strategic decisions that has a profound effect on tactical and operational decisions in any organization. It has applications in various areas such as industry, services, politics, business and economics, to name just a few. Facilities can be anything which needs to be located such as hospitals, fire stations, bus stops, train stations, truck terminals, fuel stations, blood banking centers, retail outlets, urban districts, libraries, parks, post offices, airports and waste disposal sites (Daskin 2008).

Facility Location Problems (FLPs) are mainly solved by using various quantitative and qualitative techniques from operations research (OR), management science and operations management. Depending on the nature of the facility to be located, various objective functions may be considered. Among them minimizing travel distance, maximizing service level, minimizing waiting time, maximizing coverage, minimizing transportation costs or avoiding placement next to hazardous facilities are the most popular.

Some scholars believe facility location can be traced back to Pierre de Fermat, Evangelista Torricelli (1608-1647) and Bonaventura Francesco Cavalieri (1598–1647) who independently proposed (and some say solved) the basic Euclidean spatial median problem early in seventeenth century (Drezner & Hamacher 2002, Farahani & Hekmatfar 2009). Research studies on facility location formally date back to 1909 when Alfred Weber considered the location of a warehouse with the objective of minimizing the total travel distance between the warehouse and its customers (Weber 1909). An important turning point for FLPs was in 1964 when Hakimi (1964) attempted to locate (a) switching centers in a communication network and (b) police stations on highways. Hakimi (1964) promoted location theory by proving several important basic theorems, especially in network space and for Median and Centre problems.

Facilities in location problems are characterized over a wide spectrum from manufacturing facilities at one end to service facilities at the other end. It is difficult to find a pure service or manufacturing facility. In other words, most of the products offered by organizations are a mix of services (intangible products) and goods (tangible products). However, if the share of service in the delivered product is significantly higher than the good itself, we can call it a service facility. Restaurants and retailers are examples of service organizations that deliver something tangible while banks, insurance companies and schools are examples of service organizations.

The majority of the research studies published before the 1970 s focused on either manufacturing facilities or the movement of tangible goods. However, readers interested in facility location problems may refer to the following references to learn that this area has a rich OR-oriented literature: Handler and Mirchandani (1979), Love, Morris and Wesolowsky (1988), Mirchandani and Francis (1989), Francis, McGinnis and White (2015), Daskin (1995), Drezner (1995), Owen and Daskin (1998), Plastria (2001), Drezner and Hamacher (2002), Hale and Moberg (2003), Nickel and Puerto (2005), Klose and Drexl (2005), Snyder (2006), Boffey, Galvãob and Espejob, 2007, Şahin and Süral (2007), Alumur and Kara (2008), Melo, Nickel and Saldanha-da-Gama, 2009, Church and Murray (2009), Farahani and Hekmatfar (2009), Farahani, SteadieSeifi and Asgari, 2010, Arabani and Farahani (2012), Campbell and O'Kelly (2012), Farahani, Hekmatfar, Arabani and Nikbakhsh, 2013a, Farahani, Hekmatfar, Fahimnia and Kazemzadeh, 2014 and Farahani, Rashidi Bajgan, Fahimnia and Kaviani, 2015. Some of the earliest turning points in this trend were the publication of Toregas, Swain, ReVelle and Bergman, 1971 and Church & Revelle (1974) that introduced real-life applications of facility location in emergency departments. In fact, at least two aspects of their work distinguished them from previous publications: (1) the application to a fire department that is a service facility and (2) the application in urban areas that was different from the common industrial areas at that time.

This paper follows this trend with a particular focus on service facilities that are operating in “urban” areas. We are interested in the “service” or tertiary sector since in relation to enterprise turn-over and gross domestic product (GDP) of nations, service industries are playing an increasingly more important role than their manufacturing counterparts, especially in developed countries. For example, in 2012 service industries contributed to producing 79.7% of the GDP of the United States while the rest came from the agriculture and manufacturing industries. Services’ share of the GDP of the UK, France, the US, Japan and Germany were 79.6%, 79%, 77.6%, 72.2% and 69.1%, respectively (CIA World Fact-book, 2015). We highlight urban areas since according to our survey, unlike manufacturing facilities, the majority of service facilities are located within urban areas.

Urbanization, considered to be the process of people migrating from rural to urban areas, is a common phenomenon in developed and developing countries. Interested readers may refer to Knox & McCarthy (2012) to learn about the basics of this field. There are positive and negative effects in urbanization. Some of the positive effects can be employment opportunities, quality of educational systems and access to health services. There are also negative effects associated with urbanization such as air, water and noise pollution, waste-disposal, high energy consumption (Jones 1991, Sadorsky 2013), traffic congestion, high population density, lack of infrastructure, housing provision, slum development (Vij 2012) and crime. The service facilities located in urban areas play an important role in all the above-mentioned causes. From the literature we identify some unexplored gaps in the urban service facility location (USFL) area compared with manufacturing facilities. Cities have become a significant contributor to the quality of life of individuals as well as to the overall economy. Currently, more than half of the world's population live in urban areas and about 1.3 million people are added to cities every week; this rate is 3 million people per week in developing countries (World Urbanization Prospects, 2015). With this trend, it is estimated that by the year 2050, about 66 percent of world's population will be living in an urban environment (World Urbanization Prospects, 2015). This rapid growth rate in urbanization has resulted in a growing demand for different services in cities. Cities generate 80 percent of the global GDP (Dobbs et al. 2011). As the engine of economies, cities must have an efficient transportation system to facilitate ease of mobility and avoid time, productivity, fuel and pollution costs. According to the American Society of Civil Engineers in 2013, the estimated annual costs of wasted time and fuel due to traffic congestion in the US is about $101 billion (ASCEReport Card for America's Infrastructure 2013). Cities must have sufficient capacity for an increase in residence and the necessary infrastructures to provide required services such as education, healthcare, entertainment and emergency facilities. These facilities must provide acceptable levels of quality and appropriate costs by considering major concerns such as maximizing service and coverage and minimizing traffic congestion and waiting times.

This paper is motivated by the significance of the USFL. In particular, we intend to elaborate on the types of decisions, location models, objective functions and constraints when focusing on the location of service facilities in the current literature with the aim of providing a general framework of problems for both practitioners and scholars, as well as identifying gaps, current trends and providing future research directions.

The rest of this paper is organized as follows: in Section 2 the research method and scope used in this survey is explained. Sections 3–8 analyze the papers that fall within the scope of this research in terms of type of decisions, location space, main assumptions, input parameters, objective functions and constraints. Section 9 introduces real-life applications in each of applications in USFL models. Section 10 concludes the paper by introducing areas for further research on the USFL.