Geographic Information System and Location Analytics for Business and Management
Summary and Keywords
A geographic information system (GIS) is a system designed to capture, store, organize, and present spatial data, which is referenced to locations on the Earth. Locational information is of value for a wide range of human activities for decision-making relating to these activities. As spatial data is relatively complex, GIS represents a challenging computer application that has developed later than some other forms of computer systems. GIS uses spatial data for a region of the Earth; such regional data are of interest to a wide range of users whose activities take place in that region, and so many users in otherwise disconnected domains share spatial data. The availability and cost of spatial data are important drivers of GIS use, and the sourcing and integration of spatial data are continuing research concerns. GIS use now spans a wide range of disciplines, and the diversity created is one of the obstacles to a well-integrated research field.
Location analysis is the use of GIS for general-purpose analysis to determine the preferred geographic placement of human activities. Location analytics uses spatial data and quantitative spatial models to support decision-making, including location analysis. The growth of location analytics reflects the increasing amounts of data now available owing to new data collection technologies such as drones and because of the massive amounts of data collected by the use of mobile devices like smartphones. Location analytics allow many valuable new services that play an important role in new developments such as smart cities. Location analytics techniques potentially allow the tracking of individuals, and this raises many ethical questions, however useful the service provided; therefore, issues related to privacy are of increasing concern to researchers.
Geographic or spatial data are data connected to a location, a place on the Earth, and such data have been important for society since the earliest times. For instance, property owners were interested in the boundaries of their holdings, and the area or value of a property was frequently used as a basis for taxation. Geographic information was important for travel, and an accurate spatial model of the world was especially critical for nautical navigation. Geography was relevant to military operations, so governments found it beneficial to understand the geography of their territory to control it better.
As people came to have a better understanding of the world around them, a variety of representations of geographic data have been introduced. The best known of these representations is the map, which provides a two-dimensional depiction of the relative position of geographic objects. The science of graphically presenting the Earth through maps is known as cartography, which was a well-developed field long before the computer age. For centuries cartographic representations were one of the most developed forms of visualization used for decision-making. Maps initially concentrated on providing a usable representation of the relative location of geographic and man-made features. The effective mapping of larger areas required an understanding that the Earth is a spherical body that is being represented on a two-dimensional sheet of paper. A map projection is a systematic transformation of locations from the surface of a sphere into locations on a plane. As a globe cannot be peeled off and laid out flat, map projection systems invariably involve compromise as a sphere cannot be represented on a plane without distortion. Different projections have different characteristics, so maps that are suitable for one purpose may be less suitable for another area of application.
As survey techniques and cartographic representation improved, maps in the modern sense became available, an early example being the comprehensive map of France, the Carte géométrique de la France, published in 1789. Initial national mapping projects of this type required initial labor-intensive surveying of the entire territory, but this comprehensive survey then provided a foundation for the further use of geographic information in the territory concerned. In the 19th century, the availability of comprehensive survey data allowed the production of sophisticated maps, and those maps were then used to plan the routes of new canals and railways. These transport projects, and many military applications, took advantage of data on the elevation of land as well as its relative location.
With the availability of more spatial data, maps began to move beyond the mere representation of the locations toward the presentation of geographic information for decision-making. The “Atlas to accompany the second report of the Railway Commissioners” was published in Ireland in 1837 (Harness, 1837). This report was concerned with predicting the routes on which railways, then a very new technology, might be built. Thematic maps in the atlas used lines of different thicknesses to represent existing traffic flows by road and canal; Figure 1 is an early example of a flow map.
In 1854, John Snow, an English doctor, used a map to attempt to understand the mode of dissemination of the then-widespread cholera disease. His work was influenced by the earlier work of Charles Picquet, who had produced a map in 1832 visually representing the number of cholera cases in each district in Paris. Snow used a map of London to plot the locations of the homes of cholera fatalities, and he also marked the location of water pumps on the map. The visualization of the patient locations clearly showed that cholera patients were clustered around a specific pump in Broad Street, and this supported Snow’s theory that cholera was spread by polluted water. This map was a clear example of how spatial data can be visualized to identify a spatial pattern. Without the spatial plot of cholera cases, the pattern was less clear, as the disease affected young and old, men and women alike.
This long tradition has meant that cartography later provided useful guidelines for the development of other forms of visualization using information technology (IT) (DeSanctis, 1984). Cartographic concepts introduced in the pre-computer age (e.g., the flow map in Figure 1) form the foundation for modern geovisualization techniques. Less complex geovisualizations are now widespread in mainstream business visualization software (e.g., Tableau), with specialist GIS software providing more complex geovisualizations using specialist spatial techniques.
Geographic Information Systems
In the 1950s, computers started to be used for business data processing. As the first business computers were not powerful, these early applications of computing were necessarily relatively simple. Nevertheless, despite the high relative cost of computing at this time, significant cost reductions could be achieved by this automation of the routine processes required for the day-to-day business operations. Compared to business applications, geographic applications involved much larger volumes of data and more computationally intensive calculations, such as the calculation of map projections. Consequently, the use of computers for geographic data lagged behind less demanding areas of application. Additionally, early developments in GIS and quantitative geography challenged the computational ability of the available technology (Nagy & Wagle, 1979).
In the late 1950s, computers began to be used in North America for the automation of geographic calculations, but in a much less widespread way than for business data processing. In the mid-1960s, the Canadian Land Inventory project introduced the term “geographic information system” (GIS). This large-scale project was a multilayer land-use/planning map to analyze the areas in use or available for use for activities such as forestry, agriculture, or recreational activities. As a large area of around 1 million square miles (2.6 million km2) was involved, the computational ability of the computer significantly improved manual processing approaches, similar to the way the use of data processing improved productivity in business domains.
Later in the 1960s, further projects in developed countries used computer technology for automated mapping. Initially, the attraction of automated mapping lay in the productivity improvements that computing made possible. Computers facilitated the storage and editing of maps in a similar way to the improvement in business processes brought about by the use of IT. In business, when these basic functions had been computerized, further gains became evident from the greater flexibility provided by IT. Similar productivity gains from computerization also occurred in GIS. As technology became available, complex maps could be represented on high-resolution monitors as well as through output devices such as plotters. This improved technology allowed the mass update of printed maps and facilitated early experiments in different forms of geographic information provision, although the data volumes and complex calculations required still stretched the performance of the available computer technology at that time.
GIS requires the use of suitable computers coupled with ancillary data capture and output devices and appropriate software to allow these devices to be used. GIS software has a distinct capability in the provision of a spatial database, which provides the capability to reference and query data using a geographically referenced coordinate system. At a global scale, we use the latitude and longitude system, which models the curvature of the Earth. In practice, most maps which are concerned with only a small part of the Earth and use other local coordinate systems simplify or ignore issues arising from the curvature of the Earth. One advantage of modern GIS is that the computer can quickly translate coordinate systems when required, calculations that challenged mapmakers in earlier periods of history. While GIS capabilities originated in specialized software, general database systems such as Microsoft SQL Server and Oracle have now been extended to handle spatial data. These enhanced systems incorporate spatial indexes and special spatial database operations which utilize these spatial operations.
Spatial functionality is more demanding on computer performance than most other computer applications; therefore, it was the 1980s before GIS developed in a widespread way. In the 1980s, GIS software typically ran on expensive computer workstations, as the regular personal computers of the day were not sufficiently powerful. GIS software in the 1980s was still at a relatively immature stage of development, and the focus was on the creation and representation of maps. Because of this limited focus, GIS software in the 1980s was unwieldy and lacked the flexibility to be an adequate basis for decision-making applications. The continuing improvement in computer performance since the early days of GIS means that GIS software in the 21st century will run on commonly available computers and increasingly also on handheld devices. Commercial GIS software has gone through several generations of development, as illustrated by the software from Esri, the market leader. Newer software provides additional functionality and also an increased ability to interact with other software, which provides improved flexibility to build customized applications. In addition to commercial software, open-source software projects have also now reached comparative maturity, for instance QGIS.
In addition to the improved performance of user devices, GIS has benefited from rapidly improving data capture technology, especially the availability of unrestricted global positioning system (GPS) signals since 2000. GPS allows simple devices, smartphones being a typical example, to identify their position on the Earth; this facilitates both data capture and the use of applications on those devices. One valuable source of information is remote sensing, the observation of an object from a distance. The main categories of remote sensing are aerial photography and the use of satellites to observe the Earth, but drones are also now playing an increasing role in data capture. Remote sensing can use a variety of electronic sensors, using technologies such as radar or infrared in addition to conventional photography. These techniques allow the collection of large amounts of data, but this then presents a challenge in the identification of the objects recorded. Modern machine learning techniques can assist in this identification process, especially as most maps are being updated rather than surveyed for the first time.
As GIS uses spatial data that is of interest to an extensive community of users, advances in network technology, especially the Internet, have facilitated widespread GIS use. The large volumes of spatial data needed in some spatial applications mean that cloud computing has been a valuable technology (Yang et al., 2011). The widespread use of mobile devices has facilitated the widespread presentation and collection of geographic information and has formed the basis of many of the developments in the last two decades. Over time, the range of potential GIS applications was greatly expanded by the combination of large spatial databases in the cloud, connected to custom apps on mobile devices that organize that general spatial data to meet a specific user requirement, which can then be displayed on the mobile device at a location of the user’s choice.
Spatial data are data connected to a location, usually by a coordinate that indicates the position of the spatially referenced object on the Earth’s surface (see Table 1). Modern spatial databases typically hold spatially referenced information on three basic types of object; points, lines, and areas (also known as polygons). Points represent a single location and are typically used to identify the existence of objects whose actual dimensions are not of interest in the map in question. On a detailed map a point might represent a utility pole, while on a map of an entire country a point may represent a town.
Lines are used to represent linear objects, both natural features such as roads or rivers and man-made features such as utility networks. Each line is composed of several segments so that lines can appear to be curved. A closed boundary line that forms an enclosed shape is known as an area or polygon. Many natural features are represented in this way, for example lakes or mountains. Areas are also an important form of representation for administrative regions and other political structures and can also represent business concepts such as a service territory.
Table 1. Types of Spatial Data
Information technology use in business moved from initial simple data processing applications to the later storage of business data in databases. These databases then formed the basis of management information systems (MIS) and decision support systems (DSS). While these diverse systems were used in the same organization, the users of these management systems in many cases were in a different department of the company from the operational data processing, for instance in the finance department. Similarly, as technology improved, the use of spatial data also moved from operational applications (e.g., creation of printed maps) to database applications that could support more specialist decision-making. However, while business data were usually only used within the same organization, digital spatial data have from the beginning been used beyond the organization that collected the data because many different types of organization, public and private, share the same geographic space represented by that data.
Government mapping agencies were an important group of early adopters of GIS. These agencies initially wanted to improve their productivity in the production of traditional maps. However, once the spatial data became available in an electronic form, it made specialist applications economically feasible. Consequently, a market developed for spatial data, which connected spatial data providers with GIS users in different domains who required that data. In general, this concept of a market for digital data was relatively unusual in the early days of GIS, although digital marketplaces have become more common in the period since then.
Internationally, there has been a notable difference between the United States, where public mapping data was made available at little or no cost, and other developed countries, where public bodies wanted to recover the cost of digital spatial data from users. When mapping agencies outside the United States first made spatial data available, it was often expensive and better suited to display as a map than further processing. As the number of users has increased, it became possible to recover government cost spread over a greater number of licenses, which has led to government agencies dropping the cost of their spatial data over time. The high cost of government data in many countries encouraged the emergence of competing private-sector sources of spatial data, and these private sector data providers designed their spatial databases for use in a GIS from the beginning, as GIS users were their primary customer base. In addition to national mapping agencies, other useful data are available from other public organizations such as local government or utility providers. However, there remains a significant challenge in integrating data from providers from different sectors, public and private. As a consequence, there have been spatial data infrastructure (SDI) initiatives internationally to better catalog spatial data and provide standards that allow spatial data from different sources to be easily integrated (Budhathoki & Nedovic-Budic, 2008).
The GIS research environment is influenced by its history and the communities that have initially contributed to its development. GIS software is a combination of appropriate spatial data storage, spatial data processing, and visualization. GIS is a data-intensive environment with specialized processing and storage techniques, and these techniques have required specific computer science research. The cartography research community is long-established, and this has formed the basis of geovisualization research, which remains connected to other forms of computer visualization. Such systems then formed the basis for successful commercial products that greatly extended the use of GIS, and this increased use brought GIS to other domains and therefore raised new research questions. The core geography community chose to introduce the term geographic information science (Goodchild, 1992) for the body of core research on techniques for processing geographic information. The abbreviation GIS is also used to represent geographic information science, causing some terminological confusion, so the longer expression “GIScience” is often used, and the abbreviated term GISc is sometimes favored.
Consequently, GIS research is influenced by the information systems (IS) field (Keenan & Miscione, 2015), by ongoing technological developments and by a distinct body of methods drawn from the GIScience community (Goodchild, 2015). Initially, GIS drew from the IS community in its progression from structured spatial data processing toward more flexible specialist applications that could be used for decision support. GIS was always concerned with large amounts of data on society in general, outside organizational boundaries. The IS field paid less attention to such issues in the 1990s, but the advent of “big data” has led to new links between mainstream IS research and the GIS research community.
An important development in spatial data availability has been the augmentation of government and commercial data with open source crowdsourced data, known as volunteered geographic information (VGI) (Goodchild, 2007). These developments were made possible by the widespread availability of unrestricted GPS data and the inclusion of GPS receivers in commonly used devices, notably smartphones. The most important VGI project is the OpenStreetMap project, which has over 5 million contributors. These crowdsourced free data have become available for parts of the world where updated comprehensive data are not available or where data are too expensive, substantially extending the usefulness of GIS. OpenStreetMap and other crowdsourced projects are continually being extended, and quality errors are gradually being eliminated (Haklay, 2010).
Because of these developments, crowdsourced data is now useful for many GIS applications (Goodchild & Li, 2012), and the international standardization of OpenStreetMap facilitates the implementation of spatial applications throughout the world. Specialized VGI data initiatives either collect data of interest to particular groups (e.g., on bird or animal habitats) or focus on a particular region, especially when disasters occur (Elwood, Goodchild, & Sui, 2012). A large amount of spatial data is now potentially available between government, commercial, and VGI sources, but quality issues remain in the integration of data, and the mere existence of data does not mean that it can easily be used by prospective users (Keenan & Miscione, 2017). Consequently, the integration of spatial data from different sources has remained an important issue in the GIScience research community.
Another significant development has been ambient or “involuntary” VGI (Fischer, 2012). Spatial information can arise directly when commercial organizations receive spatial data from many users who have subscribed to their services. For instance, mobile phone companies must know the location of a device in order to route calls to it, and this information allows them to plan their network. Smartphone app users report their position to get an accurate travel time to their destination from an app, and this data on multiple users then allows the app provider to improve its calculations as travel speeds can be estimated from the changing position of users. This form of spatial data acquisition is the basis of successful business such as Waze. Spatial data also arise directly from social media use when people choose to geotag their tweets or Instagram posts. However, spatial data also arise indirectly, for instance when people discuss geographic locations in their tweets, and machine learning techniques such as text mining can be used to extract this.
While most GIS applications are two-dimensional, reflecting their origin in paper-based maps, there is also continuing interest in three-dimensional (3D) applications. Elevation data are routinely included in GIS, allowing the calculation of slopes and the representation of variations in the landscape, and many potential decision-making applications can use these data. 3D GIS has distinct visualization requirements that link it to other forms of visualization, such as virtual reality (Huang, Sun, & Li, 2016). GIS has also expanded to represent indoor locations, including the spatial movement of people within airports, shopping malls, sporting venues, and university buildings. These indoor applications link GIS research to other current work in areas such as sensor nets, the Internet of Things (IoT), and smart buildings. Indoor GIS has a valuable application in modeling emergency evacuation from buildings (Tashakkori, Rajabifard, & Kalantari, 2015). GIS techniques are also relevant to virtual reality and gaming applications that represent virtual landscapes.
Spatial Decision Support
In the IS domain, business applications moved from general-purpose business reporting to more specific DSS applications designed to plan for the future. Early GIS use was largely based on the recording of existing information and its representation in a standard form based on a map. However, early GIS researchers quickly appreciated the decision support potential of the technology. While the concept of DSS originated in the mid-1970s, the demanding nature of spatial applications meant that equivalent spatial systems did not originate until the mid-1980s when the term spatial decision support system (SDSS) was introduced by Hopkins and Armstrong (1985). The SDSS concept achieved recognition within the GIS field after the inclusion of SDSS as a chapter (Densham, 1991) in the definitive 1991 research compendium on GIS (Maguire, Goodchild, & Rhind, 1991). As was the case for many GIS applications in that period, building early SDSS was challenging as GIS software of the period lacked flexibility and the capability for smooth interaction with other software and SDSS was demanding on the hardware capabilities of that period. As it matured, SDSS came to address decisions in which spatial data played a role but where other aspects of the decision were also important, and this meant that SDSS was of interest to academic and practitioner communities far beyond the core GIScience community. Since its early days, SDSS has developed into a wide range of application areas, with new areas developing as technology improved and the necessary spatial datasets became available. However, the increasingly wide range of SDSS applications often have a limited research connection to general DSS research or the core GIScience community (Keenan & Jankowski, 2019).
The heterogeneity of SDSS applications illustrates the wide range of issues that currently arise in the GIS field. One group of early adopters of SDSS were sectors where there had been long use of decision modeling but with limited spatial information included in the decision. For instance, traditional operations research (OR) had long modeled transport and routing applications, but with very simplified spatial information. The use of GIS greatly improved the quality of spatial information, allowing the models to provide better solutions. The addition of decision models to GIS provided interactive SDSS that allowed a decision-maker to build appropriate solutions. Spatial information is valuable in problems such as noise or pollution minimization, which play an increasing role in transport planning. Sectors like public utilities started using GIS as a facilities management tool to record the location of their resources but then saw that better planning could take place with the enhanced spatial information provided. Significant public concerns such as critical infrastructure protection have a substantial spatial element, and GIS plays a valuable role in resilience planning for transport and utility networks.
One important group of SDSS users are those from disciplines related to geography who adopted GIS at an early stage and who have subsequently seen the potential to use GIS to plan and make decisions about the future. One such discipline is planning, which is intrinsically future-oriented and concerned with very widely spatially distributed activities. The planning discipline has introduced the concept of a planning support system (PSS) (Pettit et al., 2018) that exploits the many data sources in the modern smart city and makes information available to a broad range of interested users. As planning is of interest to a large and possibly diverse population who live in the area concerned, it is beneficial to involve that population in the decision-making process. Widespread involvement in planning decisions is achieved by public participation systems that have a significant spatial element and represent an extension of the group decision support system (GDSS) concept to the general population (Keenan & Jankowski, 2019).
The last category of SDSS users is those who are relatively new to both GIS and modeling. This group took advantage of the increased availability of inexpensive or free spatial data relevant to their discipline and novel convenient technology to use it, for instance GPS-enabled tablets. There are examples of new users in business, in particular in marketing, where customer location has become an important determinant in marketing campaigns. However, much of the recent growth in GIS has come from outside the traditional areas of application, often in the environmental and ecological disciplines. GIS facilitates measuring the environmental impact of developments and the modeling of ecological concerns such as the dispersion of plant or animal species. Precision agriculture is an example of a new environmental development enabled by spatial technologies (Tayari, Jamshid, & Goodarzi, 2015). In precision agriculture, farmers customize their approach using a precise spatial location, selecting crop varieties, fertilizers, and pest control strategies reflecting the slope of the land, type of soil, and other variables at that precise location.
The use of spatial systems has been transformed by the availability of GPS signals and inexpensive devices, such as smartphones, to receive them. Smartphones have allowed the development of location-based services, which identify an individual’s location and then offer a service tailored to that specific location. It is estimated that more than half the world’s population now have smartphones, which provides an enormous market for location-based systems and provides a rich source of ambient spatial data collection. The location-based service provided might be as simple as the presentation of a map centered on the user’s location or might involve spatial processing to identify the nearest service or business of interest to the user. For instance, a public transport app can identify the nearest bus stop and offer information on the buses arriving at that stop in the next few minutes. Other widely used systems offer route navigation from a user’s current location to a destination of their choice. Such systems can be considered as a simple form of SDSS made available to the general public. Location-based systems play an important role in the modern “sharing economy” with new business models like Uber or Lyft based on the availability of the location of the service providers and the customers (Cohen & Kietzmann, 2014). Location-based marketing offers services tailored to customers who visit certain locations, encouraging them to avail themselves of a service or visit a retail location, this can either be based on potential customers nearby at that point in time or based on regular visitors to a location. Nearby services are often of interest to customers, and they are often willing to tolerate receiving promotional information about places nearby, although they would regard information on places away from their typical routes as a nuisance. However, many people have privacy concerns about location-based systems, and this is reflected in public regulation of such systems.
Figure 2 represents a summary of the influences on GIS. In characterizing GIScience, Goodchild (2010) noted the importance of three distinct poles to the field: society, the human, and the computer. Figure 2 extends this approach, recognizing that GIS has additional influences beyond GIScience alone. In the figure, the technical influences are shown on the right, both from IT field generally and from GIScience. The society element is connected to the IS field, in addition to GIScience, including data marketplaces and group systems. SDSS is influenced by both the GIScience and the DSS field generally, as GIS software combines spatial techniques combined with a wide range of domain-specific decision approaches.
In earlier years, the GIS field could be distinguished from the IS field generally by the vast amounts of data involved and the extensive nature of that data. However, as technology improved in the 21st century, the IS field generally has used larger datasets, culminating in the trend known as “big data.” These larger business datasets have been made accessible to users in the form of business intelligence (BI), which summarizes large quantities of data to make it easier to understand by decision-makers. The availability of more data and of faster computers has led to the growth of business analytics, which uses quantitative approaches to analyze the data and provides forecasting models to propose future actions. Similarly, big data often contain spatial references, providing big geospatial data (Lee & Kang, 2015), and location analytics exploits these big geospatial data. These systems are more general than SDSS, which focuses on a particular specialized problem.
There is a clear spatial aspect to this data-rich environment. Spatial data are readily available for many datasets as addresses can be identified spatially by a process known as geocoding and as other locations can be positioned using a GPS enabled device. For example, cargo can be tracked from its origin to its destination when vehicles or shipping containers are equipped with GPS devices. Standard BI tools now include simple geovisualization and basic summarization functions for spatial regions.
Like business analytics, location analytics makes use of quantitative techniques and spatial data, the combination of these approaches support many valuable business applications. Market analysis can identify different levels of custom in different areas and relate these to the demographic profile and the availability of service in that area. This information can then be used to propose new locations for facilities, which could be the location of shops or the location of warehouses to facilitate the shipping of physical goods. Another good example of location analytics is the positioning of mobile phone masts to reflect the topography of a region. Insurance companies can make use of location information to assess the risk of flooding. Traditional insurance approaches might have used text addresses and have applied increased premiums on a named street because of flood claims from that street. However, spatial techniques allow accurate spatial modeling of elevation so that the company can identify that only one end of the street floods and price their insurance accordingly.
Smart cities represent an example of the integration of location data with other forms of information technology (Pick, 2017). A smart city uses digital technologies to increase operational efficiency, share information with the public, and improve the quality of public services. In addition, smart city projects aim to bring together diverse sources of information from across the region of interest, using this for analysis to improve public services and providing information back to the public using multiple channels. Singapore has long been an example of a leader in the smart city concept, with multiple sensors monitoring traffic, pollution, and other issues of concern to the government. However, smart city projects now source data from indirect sources, for instance social media, and extract location information from this. For example, citizens might tweet about traffic congestion or litter on a specific street, and text analysis allows this information to be spatially referenced, allowing authorities to develop a spatial picture of public concerns.
Location analytics is also relevant in ecological and environmental applications outside the urban context. For instance, agricultural production in a region can be modeled with the use of meteorological data, which allows public bodies to advise farmers and estimate total agricultural production. In a similar way, production from extensive forestry can be modeled, recalling the early applications of GIS in extensive forests. Spatial techniques can help model environmental concerns such as pollution or the loss of animal habitats on a regional level, and such concerns are of increasing interest to the public.
Location data are relevant to many different types of activity, and so the GIS research community has extended into multiple academic domains. GIS software has developed from its early beginnings in the 1980s and has been used to create large spatial datasets. The availability of suitable software and relevant spatial data in these datasets has initiated the use of spatial techniques in multiple subdomains and provided real value to decision-makers.
The Future of GIS
While GIS and location techniques are now used in a wide range of applications, there is still great scope for increased use of spatial techniques. GIS use among less traditional users expands when the relevant spatial data become available at a reasonable cost and this process of building and improving spatial databases continues. New data collection techniques both increase the volume of data and improve its resolution. However, significant research challenges remain in automating the integration of data to ensure coherent high-quality spatial datasets. New applications will be spatially data-intensive; for instance, the move to driverless cars will exploit a variety of spatial data sources (Goodchild, 2018). New developments will be of considerable benefit to society, for instance allowing a faster response to hazardous location-based events like storms and other disasters.
The GIScience community has continued to research ways of representing the Earth, but these novel approaches have had limited influence on software. This situation echoes that found in other forms of technology; for instance, the spreadsheet continues to reflect design decisions that were reasonable when spreadsheets were first introduced and computing power was a small fraction of what it is today. The spreadsheet design would not be optimal if designed today, yet users continue with the current spreadsheet representation because they are familiar with it. Similarly, GIS software and spatial data representations reflect decisions that were reasonable three decades ago but might be suboptimal today. Nevertheless, our use of GIS today continues to reflect representations from the past as these are embedded in GIS software. The data design for spatial data mainly originated in the United States and reflected the needs of that environment. However, a data structure that suits the needs of the United States, with its gridiron city blocks, might be less suitable for another international location with a different addressing structure. Limited assumptions about the address structure underlie the software design of most commercial software and inhibit use in some locations. However, despite these limitations, it is difficult for research innovations to make their way into widely used software. The software industry caters to the largest demand, and the demand for innovation is not always obvious, as users often only formulate new applications when the appropriate software becomes available, so the extent of demand for a potential innovation can be difficult to assess in advance.
Privacy issues and the implications of the use of digital data are now of general concern in society, and spatial data are an integral part of these concerns (Zuboff, 2019). Consequently, the use and misuse of spatial data and the appropriate restrictions to place on this is going to be a key area of research investigation in the future. Geosurveillance is surveillance that uses spatial location and is a component of privacy research of particular interest to GIScience researchers (Swanlund & Schuurman, 2019). Location data allow inference on the habits of citizens; for instance, if they visit sports stadiums frequently, they might be presumed to be fans of that sport. Location data allow inference about the associates of citizens—if two people are in the same place at the same time, then they might well be together. Well-intentioned data collection, designed to be beneficial for citizens, can also have implications for the privacy of citizens, for instance, spatial data collected in the smart city context (Kitchin, 2015). While government and business may concentrate on research on how to best track individuals, academia can provide a counterbalance by pointing out the risks and discussing mechanisms to mitigate those risks.
While the availability of multiple data sources should mean that spatial data are available for all of the Earth, the actual provision remains somewhat uneven, and this hinders the use of GIS. The term digital divide is used to describe the gap in terms of access to and usage of information technology between different groups or regions. A digital divide also exists for spatial data. Even crowdsourced data collection will work best where people have appropriate skills, and projects such as OpenStreetMap have benefited from existing resources such as older maps. Germany and Britain are the most comprehensively mapped countries on OpenStreetMap—both locations already well mapped before this project. Less developed countries have a greater need for new spatial data but often have less crowdsourced spatial data collection as people there have less training and technical resources. The United Nations has several initiatives in global geospatial information management that attempt to address these issues. In developed countries, differing attitudes to privacy or public security may inhibit the collection, dissemination, or integration of spatial data; for instance, the widely used Google Streetview is not available in Germany, although it is available in neighboring countries.
GIS has now become widely used, and improvements in GIS technology and the spatial data it provides continues to drive new areas of application. However, this breadth of use has caused the research field to lose some of its connection to the GIScience community that spawned the concept. While GIS has drawn something from the IS field, there have been a limited number of connections in the reverse direction, and the IS field could benefit from a further examination of the implications of the widespread use of spatial technologies, especially as spatial methods become integrated into other forms of software. The diverse domains that use GIS could usefully learn from each other, as they have the spatial dimension of their problems in common, although their overall fields differ greatly. Some of the most innovative new applications of GIS are in environmental disciplines, which are significant for society but not traditionally of interest to the IS field, which needs to broaden its focus.
geographic information system (GIS):
system designed to capture, store, organize, and present data referenced to locations on the Earth
geographic information science:
discipline that studies techniques for processing geographic information
short for geographic visualization, refers to a set of techniques supporting the visualization of geospatial data
use of GIS for general-purpose analysis to determine the preferred geographic placement of human activities
uses spatial data and quantitative spatial models to support decision-making
systematic transformation of locations from the surface of a sphere into locations on a two-dimensional map
observation of an object from a distance to collect data, e.g., by aerial photography
spatial data infrastructure:
set of technologies and policies that facilitate the availability of spatial data
data referenced to locations on the Earth
database enhanced to store and organize spatial data
Longley, P. A., Goodchild, M. F., Maguire, D. J., & Rhind, D. W. (2015). Geographic information science and systems. Hoboken, NJ. John Wiley.Find this resource:
Pick, J. B. (2005). Geographic information systems in business. Hershey, PA: Idea Group.Find this resource:
Pick, J. B. (2004). Geographic information systems: A tutorial and introduction. Communications of the Association for Information Systems, 16, 307–333.Find this resource:
Farkas, D., Hilton, B., Pick, J., Ramakrishna, H., Sarkar, A., & Shin, N. (2016). A tutorial on geographic information systems: A ten-year update. Communications of the Association for Information Systems, 38(1), 190–234.Find this resource:
Shekhar, S., Xiong, H., & Zhou, X. (2017). Encyclopedia of GIS (2nd ed.). Boston: Springer.Find this resource:
Sugumaran, R., & Degroote, J. (2011). Spatial decision support systems. Boca Raton, FL: CRC Press.Find this resource:
Budhathoki, N. R., & Nedovic-Budic, Z. (2008). Reconceptualizing the role of the user of spatial data infrastructure. GeoJournal, 72(3–4), 149–160.Find this resource:
Cohen, B., & Kietzmann, J. (2014). Ride on! Mobility business models for the sharing economy. Organization & Environment, 27(3), 279–296.Find this resource:
Densham, P. J. (1991). Spatial decision support systems. In D. J. Maguire, M. F. Goodchild, & D. W. Rhind (Eds.), Geographical information systems, Vol. 1: Principles (pp. 403–412). Harlow, Essex, UK: Longman Scientific & Technical.Find this resource:
DeSanctis, G. (1984). Computer graphics as decision aids: Directions for research. Decision Sciences, 15(4), 463–487.Find this resource:
Elwood, S., Goodchild, M. F., & Sui, D. Z. (2012). Researching volunteered geographic information: Spatial data, geographic research, and new social practice. Annals of the Association of American Geographers, 102(3), 571–590.Find this resource:
Fischer, F. (2012). A new but delicate data-source: VGI as big data. Geoinformatics, 15(3), 46–47.Find this resource:
Goodchild, M. F. (1992). Geographical information science. International Journal of Geographical Information Science, 6(1), 31–45.Find this resource:
Goodchild, M. F. (2007). Citizens as sensors: The world of volunteered geography. GeoJournal, 69(4), 211–221.Find this resource:
Goodchild, M. F. (2010). Twenty years of progress: GIScience in 2010. Journal of Spatial Information Science, 1, 3–20.Find this resource:
Goodchild, M. F. (2015). Two decades on: Critical GIScience since 1993. Canadian Geographer/Le Géographe canadien, 59(1), 3–11.Find this resource:
Goodchild, M. F. (2018). GIScience for a driverless age. International Journal of Geographical Information Science, 32(5), 849–855.Find this resource:
Goodchild, M. F., & Li, L. (2012). Assuring the quality of volunteered geographic information. Spatial Statistics, 1, 110–120.Find this resource:
Haklay, M. (2010). How good is volunteered geographical information? A comparative study of OpenStreetMap and Ordnance Survey datasets. Environment and Planning B: Planning and Design, 37(4), 682–703.Find this resource:
Harness, H. D. (1837). Atlas to accompany the second report of the railway commissioners, Ireland. Dublin: Irish Railway Commission.Find this resource:
Hopkins, L. D., & Armstrong, M. P. (1985). Analytic and cartographic data storage: A two-tiered approach to spatial decision support systems. In Proceedings, Seventh International Symposium on Computer-Assisted Cartography. Washington, DC: American Congress on Surveying and Mapping.Find this resource:
Huang, W., Sun, M., & Li, S. (2016). A 3D GIS-based interactive registration mechanism for outdoor augmented reality system. Expert Systems with Applications, 55, 48–58.Find this resource:
Keenan, P., & Miscione, G. (2015). Desperately seeking the IS in GIS. Paper presented at the Pre-ICIS Workshop on Locational Analytics and Big Data, Fort Worth, TX.Find this resource:
Keenan, P., & Miscione, G. (2017). Spatial data: Market and infrastructure. In S. Shekhar, H. Xiong, & X. Zhou (Eds.), Encyclopedia of GIS (pp. 2003–2010). Cham, Switzerland: Springer.Find this resource:
Keenan, P. B., & Jankowski, P. (2019). Spatial decision support systems: Three decades on. Decision Support Systems, 116, 64–76.Find this resource:
Kitchin, R. (2015). Continuous geosurveillance in the “smart city.” Dis Magazine, 15, 64–76.Find this resource:
Nagy, G., & Wagle, S. (1979). Geographic data processing. Computing Surveys, 11(2), 139–181.Find this resource:
Lee, J.-G., & Kang, M. (2015). Geospatial big data: Challenges and opportunities. Big Data Research, 2(2), 74–81.Find this resource:
Maguire, D. J., Goodchild, M. F., & Rhind, D. W. (1991). Geographical information systems: Principles, techniques, management and applications. Harlow, Essex, UK: Longman Scientific & Technical.Find this resource:
Pettit, C., Bakelmun, A., Lieske, S. N., Glackin, S., Thomson, G., Shearer, H., . . . Newman, P. (2018). Planning support systems for smart cities. City, Culture and Society, 12, 13–24.Find this resource:
Pick, J. B. (2017). Smart cities in the United States and worldwide: A rich arena for MIS studies. Journal of Information Technology Case and Application Research, 19(3), 133–144.Find this resource:
Swanlund, D., & Schuurman, N. (2019). Resisting geosurveillance: A survey of tactics and strategies for spatial privacy. Progress in Human Geography, 43(4), 596–610.Find this resource:
Tashakkori, H., Rajabifard, A., & Kalantari, M. (2015). A new 3D indoor/outdoor spatial model for indoor emergency response facilitation. Building and Environment, 89, 170–182.Find this resource:
Tayari, E., Jamshid, A. R., & Goodarzi, H. R. (2015). Role of GPS and GIS in precision agriculture. Journal of Scientific Research and Development, 2(3), 157–162.Find this resource:
Yang, C., Goodchild, M., Huang, Q., Nebert, D., Raskin, R., Xu, Y., . . . Fay, D. (2011). Spatial cloud computing: How can the geospatial sciences use and help shape cloud computing? International Journal of Digital Earth, 4(4), 305–329.Find this resource:
Zuboff, S. (2019). The age of surveillance capitalism: The fight for a human future. New York: Public Affairs.Find this resource: