Embed Size (px)
Citation preview
Review of the use of Geographical Information Systems in theMarketing and Planning of Logistics Services
Mike Forster
Christian Salvesen Logistics Research Paper no. 3
September 2000
Executive Summary
Geographical Information Systems (GIS) enable storage, manipulation, analysis anddisplay of geographically referenced data. The rate of growth in the GIS industry hasaccelerated in the 1990s as businesses have adopted GIS to relate different sources ofinformation to one another through a common geographical reference.
The aim of this paper is to examine opportunities for GIS to add competitive advantage inmarketing and planning for companies on the supply side of logistics. It draws togethermaterial from fourteen interviews with software vendors and logistics service providersand from secondary desk-based research.
The value of GIS lies in enabling users to integrate different sets of data through acommon geographical reference system such as latitude and longitude, eastings andnorthings or a common pre-defined geography such as U.K. postcodes. Users can theninterrogate these data based on their geographical relationships and display the results on amap, in a table or on a chart.
Techincal barriers to the widespread use of GIS in business have been eroded in the 1990sby five factors: reduced cost of computing power; increased availability of digital mapdata; availability of software component technology; integration with corporate databases;and growth in use of the Internet for sharing software and data.
Spatial data processing functions are now available under a wide range of different termsincluding ‘desktop mapping’, ‘spatial information’. ‘spatial resource planning’ and ‘spatialdecision support’ as well as geographical information systems. These functions can befound as part of wider corporate decision support systems.
There are numerous opportunities to exploit geographical relationships in data withinactivities in the value chain and to support different levels of logistics decision-making.
Amongst companies providing logistics services the term ‘geographical informationsystems’ (GIS) is either not recognised or considered to include any software capable ofdisplaying digital maps.
The greatest use of software packages with an element, or component, of GIS technologyis at an operational level e.g. routing, scheduling, tracking, tracing or navigation. Lack ofuse of GIS packages to support higher-level logistics decision-making may be for a varietyof reasons: a lack of involvement of contract distribution companies in these decisions; theavailability of centralised resources for planning; difficulty in justifying the cost of buyingand supporting a mapping package; the perception by software companies that logisticsservices is not a target market.
The greatest use of GIS packages appears in two areas: companies with large numbers ofcustomers, large networks of facilities and a large geographical spread e.g. express parcelscompanies. Also logistics consultants involved in strategic levels of logistics decision-making.
Given the inherently geographical nature of much of the analysis that supports logisticsdecision-making it is likely that the use of these packages will grow within this sector.
Acknowledgements
This project has been financed by research sponsorship for the study of transport and logisticsissues from Christian Salvesen plc. The views expressed in the paper are solely the responsibilityof the authors.
I am very grateful to all those managers who participated in the survey.
Contents1. INTRODUCTION........................................................................................................................... 1
1 WHAT ARE GEOGRAPHICAL INFORMATION SYSTEMS? ................................................. 2
1.1 DEFINING GIS ............................................................................................................................... 21.2 FUNCTIONS OF A GIS ..................................................................................................................... 3
2 GIS IN BUSINESS.......................................................................................................................... 6
2.1 THE DEVELOPMENT OF THE GIS MARKET........................................................................................ 62.2 OPPORTUNITIES FOR GIS IN LOGISTICS ........................................................................................... 82.3 RESEARCH METHOD .................................................................................................................... 11
3 OPPORTUNITIES FOR GIS IN LOGISTICS SERVICE PROVIDERS................................... 12
3.1 THE ROLE OF LOGISTICS SERVICE PROVIDERS IN DECISION-MAKING ............................................... 123.2 MANAGEMENT OF DISTRIBUTION RESOURCES ............................................................................... 123.3 INVOLVEMENT IN HIGHER LEVEL DECISION-MAKING ...................................................................... 143.4 LOGISTICS SERVICE PROVIDERS AND MARKETING DECISION-MAKING ............................................. 203.5 AVAILABILITY OF STAFF AND TRAINING ........................................................................................ 20
4 CONCLUSIONS ........................................................................................................................... 22
APPENDIX: SOURCES OF INFORMATION ON GIS...................................................................... 23
TRADE ASSOCIATIONS: ......................................................................................................................... 23PEER REVIEWED ACADEMIC JOURNALS: .................................................................................................. 23TRADE PRESS (AVAILABLE IN THE UK): .................................................................................................. 23WEB BASED MAGAZINES:....................................................................................................................... 23UK CONFERENCE WEB SITES: ................................................................................................................ 23
REFERENCES....................................................................................................................................... 24
Figures and Tables
FIGURE 1 AUTOMATED MAPPING ................................................................................................................ 3FIGURE 2 THEMATIC MAPPING .................................................................................................................... 3FIGURE 3 MAP OVERLAY MODELLING ......................................................................................................... 3FIGURE 4 SPATIAL QUERY........................................................................................................................... 4FIGURE 5 RASTER AND VECTOR DATA (AFTER WORBOYS, 1995) .................................................................... 5FIGURE 6. PARAMETERS FOR GIS .............................................................................................................. 19
TABLE 1 GIS IN THE VALUE CHAIN. SOURCE: HENDRIKS, 1998...................................................................... 9TABLE 2 A TAXONOMY OF LOGISTICS DECISIONS.SOURCE: MCKINNON 1998.................................................. 9TABLE 3 GIS OPPORTUNITIES TO SUPPORT LOGISTICS DECISIONS. ................................................................ 10TABLE 4 ACORN™ GEODEMOGRAPHIC CLASSIFICATION (©CACI LIMITED)................................................ 16
1
1. Introduction
Information technology (IT) can play a major role in improving efficiency and effectiveness
in logistics processes. Through these improvements IT can provide a competitive
advantage to both users and providers of logistics services. Logistics is inherently concerned
with geographical information: from board level decisions about the location of
manufacturing and warehousing to operational level decisions about the best route for a
vehicle. IT that is specifically designed to work with geographical data offers interesting
possibilities for both users and providers of logistics services.
Geographical Information Systems (GIS) enable storage, manipulation, analysis and display
of geographically referenced data. The GIS industry has developed over the past 30 years
from specialist academic and government roots in cartography, photogrammetry and remote
sensing into a $1.2 billion industry world-wide1. The rate of growth in the GIS industry has
accelerated in the 1990s as businesses have adopted GIS to relate different sources of
information to one another through a common geographical reference.
The aim of this paper is to examine opportunities for GIS to add competitive advantage in
marketing and planning for companies on the supply side of logistics i.e. the companies that
provide logistics services such as contract distribution, freight forwarding, express parcels,
international haulage, short sea and deep sea shipping and all cargo aircraft.
2
1 What are Geographical Information Systems?
1.1 Defining GIS
GIS are commonly defined by the processes that are carried out: “… computer system[s] for
capturing, storing, checking, integrating, manipulating, analysing and displaying data related
to positions on the Earth's surface”2. The physical components required to carry out these
processes are illustrated in figure 1. The software is divided into four parts: DBMS
(database management system; graphics system; interface; and operating system). There is
specialist hardware - scanner, plotter and digitiser – in addition to the computer and the
graphics workstations. There are data, spatial (map) and attribute (company). Finally, the
system may be linked through a communications network.
Figure 1. The physical components of a GIS. Source: AGI
In practice, the term GIS is often used to mean just the software product that contains the
functions necessary to carry out some, or all, of these processes. Used in this sense,
geographical information systems are primarily commercial products for producing maps.
Their value lies in enabling users to integrate different sets of data through a common
geographical reference system such as latitude and longitude, eastings and northings or a
common pre-defined geography such as U.K. postcodes. Users can then interrogate these
data based on their geographical relationships and display the results on a map, in a table or
on a chart. Well-known GIS products in the UK include ArcInfo™ and ArcView™ from
ESRI, GeoMedia™ from Intergraph, MapInfo™ Professional from MapInfo Corp and
GeoConcept™ from GeoConcept SA (distributed by Kingswood Ltd in the UK).
3
1.2 Functions of a GIS
The processes of integration and interrogation can be broken down into a number of general
categories. Laurini and Thompson3 have identified ten major tasks for spatial information
systems listed below (figures 1-4 are derived from Laurini and Thompson).
1. Automated mapping: replicating paper maps on computer.
AutomatedMapping
Paper map
Digitized or scanned
Computer plotted map
Figure 1 Automated Mapping
2. Thematic mapping: for instance using customer information and demographic data.
ThematicMapping
Statistical areas Number of businesses Symbol map+
Figure 2 Thematic Mapping
3. Map overlay or composite mapping: producing a map from several layers of data.
Map OverlayModelling
Map thematiclayers
Combination
Figure 3 Map Overlay Modelling
4
4. Spatial querying: obtaining information from a database in response to identification of
particular conditions.
Spatial Query
A
B
C
D
A 101 2
B 54 5
C 496 3
D 44 1
Figure 4 Spatial Query
5. Spatial browsing: exploring the contents of a database in response to identification of
particular conditions.
6. Spatial problem solving: for example deducing inclusions of points in polygons, or for
spatial decision making incorporating both spatial and logical deductive reasoning.
7. Analysis of spatial data: tasks which deal with the attributes of entities, like the average
size of sales territories or the degree to which product sales are related to weather
conditions.
8. Creating spatial statistics: tasks that require measurements of spatial properties of
phenomena, like the total distance travelled by a vehicle on a road network.
9. Analysis of spatial statistics: tasks which treat spatial properties as attributes, for
example the correlation between highway network connectivity and levels of economic
development.
10. Spatial analysis: encompassing tasks, including simulation, which use a variety of tools
of spatial statistics and location-based problem solving.
These tasks define ways in which users make use of spatial data. Tasks are carried out on
spatial and attribute data sets held in a database.
Geographical or spatial digital data is generally in one of two forms: vector or raster (see
figure 5). Vector data is made up of points, lines and polygons (on the right in figure 5).
The location of customers for an express parcels operator could be stored as a set of points,
the road network as a set of lines and the boundaries of depot service areas as a set of
polygons. Raster data takes the form of a grid in which each cell stores a particular value
(on the left in figure 5). Satellite and aerial photographs are in raster format, or paper maps
5
that have been digitally scanned. Figure 5 shows that either format can be used to represent
the same data. Both formats have advantages and disadvantages in terms of computer
processing speed and storage space. Most GIS products are able to make use of data in
both formats. The attributes of points, lines, polygons or cells are typically a mixture of
numerical and textual data.
Figure 5 Raster and Vector Data (after Worboys, 19954)
Typically users will buy, or licence, standard geographical, boundary and attribute data sets
and combine these with their own data in a GIS. Major suppliers of digital map data in the
UK include the Ordnance Survey, Bartholomews and the Automobile Association (AA).
There are also a number of companies offering specialist products such as Navigation
Technologies which provide street level maps with navigational attributes or Cities
Revealed which provide aerial photography. A typical set of vector data will include
different layers for the motorway and trunk road network, for towns and cities of different
populations, ports, airports rail way stations, etc. Boundary data sets are supplied from a
number of sources. They can be in the form of hierarchies like electoral districts and wards
(from ONS) or postcode areas, districts and sectors (from Royal Mail), or a single
geography such as television boundary regions. There are also a number of general and
specialist point data sets available such as golf courses, billboards, retail outlets. The
suppliers of these data sets generally distribute through one or more official re-sellers such
as Geoplan, Kingswood, ESRI’s DataStore or MapInfo’s Data Products.
6
2 GIS in business
2.1 The development of the GIS market
Until the early 1990s geographical information systems were complex, they used proprietary
database management systems and the components were expensive. The major commercial
uses were to maintain land inventory records for local and national government departments
and utility companies. While there were many proponents of the use of GIS in business cost
and complexity were effective barriers to all but the largest companies. Since the early
1990s these barriers have been eroded by five major enabling factors resulting in the
growing use of GIS, or GIS related software, in business.
First, reductions in the cost of computer hardware and networking technology have given
any company the possibility to operate GIS software. Second, detailed digital map data
have become available for most Western countries. These are complemented by a wide
range of digital data sets available from national statistical organisations and commercial
market research companies (although the cost of these data sets varies from country to
country). Third, software component technology, in particular for Microsoft operating
systems, has enabled the major GIS vendors to supply individual mapping functions in a
form that can be easily incorporated with other packages increasing the number of people
exposed to the potential of mapping software. At the same time the vendors have been able
to integrate their products with established software products providing other functions.
Fourth, companies producing the software used to manage large databases have introduced
products that assist integration of existing corporate databases with GIS. Finally, this
process of integration has been further facilitated by the growth in the use of the Internet as
a common network for sharing software and data.
GIS products have also evolved and new companies have entered the market. GIS vendors
have also recognised the need to reduce, or mask, the complexity of GIS software in order
to appeal to a wide range of business users. This has led to the introduction of software
packages with a limited set of functions, with guidance on how to complete standard tasks
and often integrated with standard sets of map and attribute data5. Most recently Microsoft
have entered the mapping market. Microsoft MapPoint™ is a part of the Microsoft
Office™ suite. It provides users with the ability to combine their data with some commonly
used geographical data sets and produce maps. Maps can be incorporated in document,
7
spreadsheets and presentations. Microsoft are offering ease-of-use, key data sets and
limited functionality at a price that is considerably lower than the major desktop mapping
packages for the Microsoft Windows™ platform.
As the commercial market has developed so vendors of GIS products have moved away
from the original term, ‘GIS’, and created a range of alternative descriptions for software
products, such as ‘desktop mapping’, ‘spatial information’. ‘spatial resource planning’ and
‘spatial decision support’, in an attempt to differentiate their products from their
competitors and to appeal to new markets. International Data Corporation (IDC), a U.S.
market analysis company, tracks GIS products within what they refer to as the ‘spatial
information management’ market6.
The analytical tasks performed within GIS products involve non-spatial processes
particularly statistical analysis, optimization, network routing and dynamic positioning.
Software companies can attempt to address a particular business requirement from strength
in spatial or non-spatial processing. They can either buy-in or write additional software to
supplement their main area of expertise. Rather than trying to include all functions in one
package software vendors offer modules, or add-ons, with extra functions or provide users
with the ability to write their own functions in either proprietary or standard programming
languages. There are many products that provide some of the tasks described in section 1.2
and so fall between the extremes of comprehensively featured GIS products and simple
mapping packages. These products often share a common purpose: users need to provide
information to support business decisions7. This shared purpose covers a broad range of
applications such as Enterprise Resource Planning (ERP) or Decision Support Systems
(DSS).
Generally these applications involve a number of software products linked together.
Corporate data are held in database, or data warehouse, products such as Oracle, Informix
or DB2. They are entered either through an interface to support a business process e.g.
orders, or generated from processing within a package e.g. orders delivered on time. This
interface could be a package designed for a particular database, for any database or a set of
custom-built software routines written specially for a company in a particular programming
language8. The interface may be linked to the database on the same physical machine, on a
Local Area Network (LAN) or company intranet, or across the Internet. A user at a
8
workstation can retrieve data through software located on the workstation, served across a
LAN, intranet, or the Internet. These data can be analysed and the results presented in the
form of a table, chart or map. The results can be printed in hard copy or disseminated
across a LAN, intranet, or the Internet. The spatial processing tasks constitute a small part
of this system. GIS vendors address the different permutations of system structure by
offering a range of products. For instance MapInfo sell, in addition to their original product
MapInfo Professional™ : MapX™ - a software component to integrate mapping in other
products; SpatialWare™ - software to integrate with database products from Oracle,
Informix and IBM; MapXtreme™ - a map server for the Internet; and MapXsite™ -
software to provide ‘Where’s the Nearest’ capabilities in company web sites.
The result of these developments is that a broad range of software now exists incorporating
some spatial information functionality. Trying to define a GIS market is increasingly
difficult, and, perhaps, increasingly meaningless. A more practical approach to
understanding the potential value of GIS technology is to look at how spatial data can
support decision-making in different areas of business.
2.2 Opportunities for GIS in logistics
An often-cited statistic in promotional literature on GIS is that 80% of business data has a
geographical element. The implication is that geographical information systems have
relevance and can add value in almost any area of business. Hendriks9 uses using Porter’s
concept of the value chain to examine where geographical, or spatial, information systems
fit within the information strategy of organisations. He suggests that companies can identify
opportunities within the five primary and four supporting activities of the value chain. His
indication of likely opportunities is shown in table 3.
9
Table 1 GIS in the value chain. Source: Hendriks, 1998
ADMINISTRATION & INFRASTRUCTURE: GIS as a tool for strategic plann ing; as a spatial decision support toolfor asset managementHUMAN RESOURCES MANAGEMENT: Flexib le workforce management based on project locationPRODUCT / TECHNOLOGY DEVELOPMENT: Examination of effects of spatialization in process/productPROCUREMENT: fleet management, supply management
INBOUNDLOGISTICS:optimization ofwarehouse usage;logistics modelling
SALES &MARKETING:GIS as a marketanalysis tool;simu lation ofdispersion of newproducts;target marketingand advertising
SERVICES:route planning;dealer networkmaintenance;customercomplaints;dispatch;maintenanceforecasting
OPERATIONS:enhancing thespatial conten tof process orproduct
OUTBOUNDLOGISTICS:route planning;fleet management;deliveryassessment
There is a strong logistics focus including route planning, optimisation, modelling, network
maintenance, fleet management and delivery assessment. The ‘spatialization’ of process or
product might be taken to mean the way in which demand varies geographically. Equally,
enhancing the spatial content of process or product might refer to customisation or
postponement operations although this is not clear. It is possible to divide the potential
logistics opportunities according to the level of decision-making that they support.
McKinnon10 provides a taxonomy of logistics decision making consisting of four levels
shown in table 2. This taxonomy was originally devised to help assess how logistics
decisions affect the demand for road freight transport. Changing ‘transport’ to
‘distribution’ at the bottom level is sufficient to include other functions besides transport.
Table 2 A taxonomy of logistics decisions. Source: McKinnon 1998
Level Description
Logistics structures Numbers, locations and capacity of factories,warehouses and terminals
Pattern of trading links Created by commercial decisions on sourcing,sub-contracting and distribution, and manifest asa freight network linking a company'’ premises tothose of its trading partners
Scheduling of product flow The programming of production and distributionoperations translate trading into discrete freightflows. Adherence to a just-in-time (JIT) regime,for example, usually requires frequent delivery ofsmall orders
Management of transport resources Within the framework defined by decisions at theprevious three levels, transport managers stillhave discretion over the use of transportresources.
10
Table 3 places the opportunities that Hendriks sees for GIS in logistics at the level of
decision-making that they support. In addition, alternative descriptions of opportunities,
based on descriptions of existing software, have been added. The table is not intended to be
definitive but to illustrate the fact that opportunities for GIS exist at all levels of decision-
making. For a comprehensive list of software packages to support logistics processes see
the European Logistics Software Guide (Institute of Logistics and Transport, Arthur
Andersen and Cambridge Market Intelligence) now in its 13th edition.
Table 3 GIS opportunities to support logistics decisions.
Level Decision support software
Logistics structures Strategic planning,Distribution network planning,Asset management,Performance measurement
Pattern of trading links Supply management,Demand analysis and measurementDealer network maintenance
Scheduling of product flow Logistics modellingGoods receipt and despatchDispatchDelivery assessment
Management of transport resources Routing and schedulingDriver and vehicle performanceFleet management,Maintenance forecasting,Optimization of warehouse usageWarehouse management
Articles in US trade journals suggest that companies are using GIS in logistics. Proctor and
Gamble used MapInfo in combination with optimization software to re-design their facility
location11. Federal Express used ArcInfo™ and GenaMap™ to measure service standard
across the U.S.12 and have built a warehouse location model using the Visual Basic
programming language together with ESRI’s Map Objects™ mapping components13. A
recent special edition of the magazine Business Geographics discusses the role of GIS in
supply chains14. Articles in the U.K. are scarce. However, a recent article in Distribution
Business hints at growing interest in GIS in logistics, suggesting that: “A quiet revolution is
under way in the use of computer-based mapping and geographic information systems in the
logistics world”15
11
2.3 Research Method
It is against this background of potential opportunity that a series of interviews was
undertaken to investigate the use of GIS amongst logistics service providers. Given the
range of different possible applications a random survey did not seem appropriate. The first
step was to talk to a four of GIS vendors to establish how they viewed the potential for GIS
in logistics. These companies were also asked if they could provide introductions to
customers who were using GIS for logistics applications. Where possible these
introductions were followed up. At the same time ten logistics service providers and
logistics consultancies were approached for interview separately. While these interviews do
not constitute a sample from which statistical generalisations can be made they do provide
valuable insight into the level of understanding of GIS amongst logistics service providers.
Where companies are making use of GIS the interviews shed light on some of the barriers to
using GIS and benefits that can accrue. The remainder of the paper discusses the
opportunities described above and draws together material from the interviews and from
secondary desk-based research.
12
3 Opportunities for GIS in Logistics Service Providers
3.1 The role of Logistics Service Providers in decision-making
Companies providing logistics services are operating as part their clients’ logistics systems.
In terms of the value chain shown in table 1 they are part of their clients’ inbound and
outbound logistics. From the perspective of a client’s value-chain the opportunities for GIS
to support inbound and outbound logistics depend on the involvement of the service
provider in the clients’ logistics decisions. From the perspective of the service provider’s
value-chain, administration, human resources, technology development, procurement, sales
and marketing, services and operations are adding value to the inbound and outbound
logistic services.
The role of logistics service providers has been limited to the lowest level of decision-
making, the management of distribution resources. De-regulation of freight transport,
particularly by road, stimulated the market for outsourcing of freight transport and
warehousing services. Companies that did not see transport as a core activity could remove
the costs of transport assets and labour from their balance sheets. However, outsourcing
also helped to stimulate the market for IT products to support these functions.
3.2 Management of Distribution Resources
There is now an extensive range of software, available for different hardware and operating
systems to support anyone managing warehouses or transport operations. One source in
the UK lists over 100 different software packages that are able to provide routing and
scheduling functions16. While there may be some overlap there are clearly a large number of
routing products that are not classified as GIS. Vendors of routing and scheduling
packages have developed their products, and their geographical interfaces, independently of
the main GIS vendors.
The ability to calculate drive times is a key element of much GIS analysis in business. The
cost benefit analyses for routing and scheduling products should be straight forward to
present in terms of savings in numbers of vehicles and reduced mileage. The cost savings in
the area of operation addressed have to be greater than the cost of purchasing and using the
system. A recent estimate suggests that computerised vehicle routing and scheduling
packages can cut transport costs and distance travelled by between 5 and 10 percent17.
13
Despite this, companies have been slow to accept the use of routing and scheduling
packages18. While much effort has been focused on the technical side of routing
applications much less effort appears to have been spent in considering how people
integrate these packages into existing business processes. The management implications
and the impact on other areas of the business are often of greater concern than the use of
the package19. Interviews with logistics service providers confirm that these are still
important issues to be considered in the adoption of new software, particularly with
tracking software.
Research in GIS places routing applications under the umbrella of GIS-T meaning GIS for
transportation. Waters20 provides a detailed description of what a GIS-T software package
might include in addition to ‘standard’ GIS functions: matrix handling; shortest path
analysis; spatial interaction and gravity models; trip generation-trip attraction. While he
suggests that there are GIS-T software packages that include some, or all, of these
functions (in particular he mentions the TransCad™ package produced by Caliper
Corporation) he also points out that the package is a combination of GIS and other
software. He also refers to the links between GIS-T and automatic vehicle location systems
(AVLS). AVLS covers a range of related applications that use some form of
communications technology, such as the Global Positioning System (GPS), to track the
location of a vehicle. According to articles in the logistics trade press these systems are
growing in popularity as they offer opportunities for dynamic re-scheduling of vehicles to
take advantage of additional loading opportunities or to avoid areas of congestion21. They
also offer management greater visibility over how their assets are being used. GIS can
provide a visual display of the location of the vehicle. In the UK Isotrak offers a range of
business services to fleet managers based on a combination of technologies that includes
uses the GeoConcept™ GIS package. Several of the major GIS vendors market modules
or add-ons for their products that provide tracking functionality. However, GIS rarely
forms a part of the marketing literature for tracking products themselves. The mapping
functions are ‘embedded’ in the package and the visual interface is taken for granted22.
Where logistics appears in GIS vendor marketing it is generally in combination with a
routing and scheduling package or a tracking and tracing system. MapInfo produced a
promotional white paper on Desktop Mapping and Transportation in the U.S23. The main
thrust of the paper is on the links between mapping and routing. More recently ESRI have
14
launched a product in the U.S. called ArcLogistics™ which appears to be a combination of
routing and mapping functions. According to their publicity material it is derived from
work that they carried out with the Sears Group in the US which included the use of GIS to
assist in warehouse operations. The extent to which new entrants to the routing software
market can succeed may depend on their ability to tap into the accumulated experience of
existing software providers in the practical problems of applying routing and scheduling
software in a business environment and of integrating their products with other business
software. In the U.K. Kingswood Ltd. market GeoConcept™ for use in logistics. They are
also distributors of the Truckstops™ routing package and re-sellers of AA and NavTech™
road data sets.
This marketing focus is perhaps responsible for the common assumption amongst the
logistics service providers interviewed that the terms GIS, geographical information systems
and desktop mapping refer to routing and scheduling or tracking and tracing systems.
Overall, however, there is a general lack of awareness of GIS products. The terms
mentioned above are not generally recognised. A comprehensive survey might reveal that
any software that includes some form of digital map would fall into a common-sense
classification as a GIS. This lack of familiarity of GIS amongst logistics service providers
may be because they are seen by vendors of GIS and Desktop Mapping software as a
market for routing and tracking packages not for site analysis and marketing packages.
Another reason may be because software aimed at supporting this level of decision-making
is functionally oriented. The generic spatial processing capabilities of GIS packages may
not appeal unless they are presented with a functional orientation as in the case of ESRI’s
ArcLogistics™ .
3.3 Involvement in higher level decision-making
Service providers do become involved in higher level decision-making in their clients in
certain circumstances. First, in some cases the logistics service provider may build, or lease,
a new facility for a client in which case they will have been involved in decisions at the
logistics structures level. Second, the large contract distribution companies have created
internal consultancy departments to respond to invitations for tender and to support on-
going contracts. Third, there is a trend for clients to use one service provider, a ‘lead’
company, to manage relationships with other service providers24. This may mean that the
main service provider becomes involved in decisions affecting scheduling of product flow
15
and the pattern of trading links. In these cases the logistics service provider is becoming
involved in distribution network planning.
The nature of decisions at this level is that they involve assessing multiple criteria and
multiple scenarios. Issues of site location are also intrinsically geographical. Senior
management want to assess the performance of the existing distribution network, simulate
the performance of new configurations of the network and consider the impact of new
configurations on their operations. Different logistics service markets have different
requirements. The national freight transport companies, including express parcels, couriers
and freight forwarders, need to run a network of offices to supply their market. The
contract distributors need to understand how their operations fit into the distribution
networks of their clients to produce competitive responses to tenders and to manage change
during the course of a contract. Again, there are over 100 packages already available to
help companies with these tasks25. However, at the level of strategic site analysis there is a
close link to the use of GIS in the business-to-consumer sector.
Site analysis has been one of the major applications of GIS amongst retail businesses. For
introductions to the use of GIS as a market analysis tool see Birkin et al26, or Grimshaw27.
Retailers are concerned with the potential market that an outlet will serve. Location
decisions may be based on experience and personal judgement of an area28. However, the
availability of a large variety of data sets derived from the National Census or business data
on consumers and business derived from market research and credit rating companies
provides an opportunity to test personal intuition or to help focus attention on particular
areas.
For instance, a retail organisation has purchased a set of postcode districts and the
population data for each of those districts. Within the GIS it will define a catchment area
for an outlet usually based on a drive time from the outlet. The company could also define
a catchment in terms of a crow-fly distance or in terms of a standard set of geographical
boundaries such as a postcode. They can then overlay the catchment area on the population
data (a map overlay task) and retrieve information about the total population included in
that catchment (a spatial query task). The GIS examines the boundary of the catchment
area to see which postcode districts fall completely within it and which fall partly within it.
The software returns the total population figures for all of the enclosed and intersected
16
postcode districts. The company may then want to see which customers already fall within
the catchment area. If they have the postcode for each of their customers they can link each
customer to a geographical reference point, in this case perhaps the geographical centre of
the postcode in which they are located. This process is known as ‘geocoding’. This layer
of customer information can then be super-imposed on the catchment (a map overlay task).
This gives a visual impression of how many existing customers fall within the new
catchment area of the new store.
The company can also make use of geodemographic data. These data sets are derived from
statistical analyses of census data and aim to divide the population into a number of
categories of shared interests, affluence or need. The categories are linked to areas through
electoral or postcode geography. The most popular of these are the ACORN™
classification from CACI and MOSAIC™ from Experian. The principal value of
geodemographic classification in business is that it provides a link between existing
customers living in an area and the total population living in that area. For instance, the
ACORN™ classification is split into 6 Categories, 16 Groups (shown in table 4) and 56
different Types.
Table 4 ACORN™ geodemographic classification (©CACI Limited)
Category A Thriving Group 1 Wealthy Achievers, Suburban Areas
Group 2 Affluent Greys, Rural Communities
Group 3 Prosperous Pensioners, Retirement Areas
Category B Expanding Group 4 Affluent Executives, Family Areas
Group 5 Well-Off Workers, Family Areas
Category C Rising Group 6 Affluent Urbanities, Town & City Areas
Group 7 Prosperous Professionals, Metropolitan Areas
Group 8 Better- Off Executives, Inner City Areas
Category D Settling Group 9 Comfortable Middle Agers, Mature Home Owning
Areas
Group 10 Skilled Workers, Home Owning Areas
Category E Aspiring Group 11 New Home Owners, Mature Communities
Group 12 White Collar Workers, Better-Off Multi-Ethinc Areas
Category F Striving Group 13 Older People, Less Prosperous Areas
Group 14 Council Estate Residents, Better-Off Homes
Group 15 Council Estate Residents, High Unemployment
Group 16 Council Estate Residents, Greatest Hardship
Group 17 People in Multi-Ethnic, Low-Income Areas
In a given area a business has a certain number of customers in each of these 56 types. The
total population for that area can also be broken down into each of the 56 types. By
17
comparing the ACORN™ profile of the customers with the ACORN™ profile of the total
population it is possible to see if the customers are over or under represented in particular
ACORN™ types. If the business infers from this comparison the ACORN™ types most
likely to be its customers it can also see how many potential customers are in that area and
also focus its marketing on the same ACORN™ types in other areas. A map is not
necessary for geodemographic profiling but enhances the value geodemographic analysis by
helping the business visualise the results.
These are only a few simple examples of the processes that a company might go through in
using a GIS to support a decision to open a new site close an existing site. In practice,
these simple tasks are repeated many times using a mixture of commercial and company
data. A number of modelling techniques, such as gravity modelling and location-allocation
can be used to derive new data that can be mapped. The procedures can be automated so
that a number of simulations can be run perhaps to suggest a number of new locations, or
examine the effects of changing key criteria such as floor space or product range. Target
marketing activities such as planning direct mail or leaflet drop campaign focus on the
individual rather than the site. However, at the heart of target marketing is the need to
profile and segment both existing and potential customers. Companies carry out these
activities using a mixture of demographics, geodemographics and company data.
For the logistics service industry there is little reason to use geodemographics. There have
been attempts to try and create geodemographic classifications for the business-to-business
sector but the characteristics of businesses are not directly related to geographical location
in the same way that they are for consumers. Companies that are possible exceptions to this
are those involved in home delivery operations or house moving. These companies can
assess likely demand for their services through the customer profiling although whether they
have access to data about the people they are delivering to is not clear.
Another set of GIS related applications has developed around the concept of territory
management where the location of sales staff or service engineers changes. Here the aim is
to create sales territories or service areas that are balanced according to chosen criteria such
as potential sales or clients or equal drive time. The mapping elements of these products are
linked to optimization algorithms. With all of these applications – territory management,
retail site location and target marketing – the value of the spatial tasks is in providing a
18
visual interface and enabling spatial query and browsing. In the case of business-to-
consumer marketing additional value is added through the use of additional demographic
and geodemographic datasets. It is the availability of these data sets that has helped to
stimulate the use of GIS as a market analysis tool.
In the course of the interviews companies using GIS packages confirmed that the visual
interface and the ability to integrate different data sets are the key benefits of a GIS package
for strategic logistics analysis. However, most of the data used by these companies is
internal. Unlike the retail sector there does not appear to be an opportunity to develop
value-added data products from existing national data sets. Alternatively, it may be that no
one has yet tried to develop any such data sets. Where GIS packages are used it is in
conjunction with other packages rather than in direct competition with them. The
availability of GIS functions as software components may improve an original custom-built
map interface in the same way that reporting functions such as Seagate’s Crystal Reports
improve can improve reporting capabilities for a range of software products. For example,
the Cast-DPM™ strategic logistics-modelling package from Radical now includes mapping
using ESRI’s MapObjects™ .
In the retail sector the decision to use a GIS requires an assessment of the potential loss in
making a bad site location decision, the extent to which balancing sales territories will lead
to improved sales or better service, or the number of new customers gained through a
marketing campaign. This has to be weighed against the cost of the application, the data
used within it, which often exceeds the price of the software many times, and the cost of the
user to operate the system. The economics favour companies with a large number of retail
outlets e.g. supermarkets, car dealerships, retail arms of banks, a large numbers of
customers or a large number of sales or service staff. This is partly because in smaller
companies it may be as cost effective to carry out some of these tasks manually. It is also
because, typically, the larger companies are able to dedicate staff to the use of the software.
These simple parameters are shown in figure 4.
19
Figure 6. Parameters for GIS
In summary, the value of GIS lies in the ability to integrate disparate sets of data, visualise
them in the form of a map and link to software that provides non-spatial functions, such as
optimisation routines. This value is enhanced by the availability of commercial data sets
that enhance internal company data and provide information about potential consumers.
The greater the numbers of customers, sites or service staff the greater the potential value
of using GIS to support business functions. A recent article in Distribution Business alludes
to the value of using a mapping package to help visualise the results of logistics analysis29.
However, it is hard for companies to put a value on the ability to map corporate data.
In the case of software used to support strategic level logistics decisions the cost-benefit is
difficult to calculate. The cost of making a bad decision must be weighed against the price
High potential forNetwork PlanningLow potential forMarketing
Low potential forNetwork PlanningMarketing
High potential forNetwork PlanningLow potential forMarketing
High potential forMarketingLow potential forNetwork Planning
Large
Small
Small Large
Number of customers
Num
ber o
f site
s
High potential forNetwork PlanningLow potential forTerritory Planning
Low potential forNetwork PlanningTerritory Planning
High potential forNetwork PlanningTerritory Planning
High potential forTerritory PlanningLow potential forNetwork Planning
Large
Small
Small Large
Number of sales staff / engineers
Num
ber o
f site
s
High potential forMarketingLow potential forTerritory Planning
Low potential forMarketingTerritory Planning
High potential forNetwork PlanningMarketing
High potential forTerritory PlanningLow potential forMarketing
Large
Small
Small Large
Number of sales staff / engineers
Num
ber o
f cus
tom
ers
20
of the software and data, and the labour cost in training and use of a piece of software.
Such a calculation may be made harder because the process is likely to involve other
software as well as the GIS. In the case of site location it is easy to suggest that the cost of
a bad decision may run into millions of pounds compared to the cost of a software package
costing thousands of pounds. However, the cost of a bad decision is only a potential loss
not an actual cost as in the case of running 10 trucks when only nine are needed. Choosing
to use a GIS package requires a belief in the overall benefits that may accrue rather than a
straightforward cost benefit analysis.
3.4 Logistics service providers and marketing decision-making
Logistics service providers need to market and sell their services. The way in which they go
about this depends on the nature of the service that they are offering. Peters et al30 offer an
interesting insight into the sales and marketing process in third party logistics, or contract
distribution, companies. They provide a ranking of the most effective ways of finding new
clients based on responses from the CEOs of major European third party logistics
companies to a survey. Direct mail, the form of marketing most commonly associated with
GIS, is ranked only eighth. The top three most effective ways to find new clients are:
reference from a customer; presentation at a professional meeting; and sales call by a
representative. In the same survey 75% of the CEOs reported that they had fewer than 100
customers. It is hard to see the need for GIS to support for the marketing function in
contract distribution companies. Where the number of customers is large the potential is
greater. Domestic and international express parcels companies, freight forwarders and large
hauliers all tend to have large customer bases. It has been suggested that the largest express
operators have gained market share through branding of services and the effective
marketing of these brands31. Companies such as Federal Express, DHL and TNT have
almost become generic terms for express parcels, mail and courier services. One of the
companies that participated in the interviews is a large express distribution company. They
make use of GIS principally for network planning. However, they also act as a resource for
other departments, including marketing, to use as they need. This maximises the benefits
gained from the use of the package for the company by focusing use on a small group of
expert users.
3.5 Availability of staff and training
Although mapping packages are becoming easier to use they are still by no means simple
software. The arrival of Microsoft MapPoint™ may change this in the medium to long
21
term. However, where GIS or desktop mapping packages are used as analysis tools within
logistics service companies or consultancies, it is by ‘experts’ who have had an opportunity
to learn to use the product and make regular use of it in their work. In one case it was
suggested that it can take up to three months for a user to become familiar with a package
and effective in its use. The packages may be broadly aimed at a particular function but
they also require creative thought. Users need to understand the concepts behind mapping
packages to be able to use them effectively. It would also be helpful if they understand
some of the basic design issues in the creation of maps. Monmonier32 provides an
interesting perspective on the role of maps as a means of disseminating information in his
book ‘How to Lie with Maps’. At least as important, users need to understand the
corporate data that is used in the analysis. Understanding what criteria to use in an analysis
of site profitability or performance comparison is critical to providing meaningful support.
The software vendors run their own training programmes to support particular products.
There are also a growing number of courses in GIS at both undergraduate and postgraduate
level at universities in the UK. As yet there is no UK equivalent to the course advertised by
the Management Development Centre at the University of Tennessee in the use of GIS in
logistics and supply chain planning.
22
4 Conclusions
Geographical information systems are available in a variety of forms, generally designed to
appeal to particular industrial markets and for particular functions. The term ‘GIS’ is used
in a loose fashion to refer to the market for mapping software products and also to a
particular sub-set of this market, complex, high-specification and expensive mapping
software.
Amongst companies providing logistics services the term ‘geographical information
systems’ (GIS) is either not recognised or considered to include any software capable of
displaying digital maps. The greatest use of software packages with an element, or
component, of GIS technology is at an operational level e.g. routing, scheduling, tracking,
tracing or navigation. Lack of use of GIS packages to support higher-level logistics
decision-making may be for a variety of reasons:
• a lack of involvement of contract distribution companies in these decisions;
• the availability of centralised resources for planning;
• difficulty in justifying the cost of buying and supporting a mapping package;
• the perception by software companies that logistics services is not a target market.
Companies may make greater use of GIS tools as: they take on more responsibility for
clients e.g. by becoming ‘lead’ logistics companies; or are exposed to consultants using
these tools. The greatest use of GIS packages appears in two areas: companies with large
numbers of customers, large networks of facilities and a large geographical spread e.g.
express parcels companies (these companies may have marketing strategies which
encourage the use of GIS for analysis). Also logistics consultants involved in strategic
levels of logistics decision-making.
The ability to map data held in corporate databases assists decision-making but is hard to
value. The cost of software and data is outweighed by importance of proficient users.
However, logistics service providers will become increasingly exposed to GIS technologies
in a variety of forms. Given the inherently geographical nature of much of the analysis that
supports logistics decision-making it is likely that the use of these packages will grow
within this sector.
23
Appendix: Sources of Information on GIS
Trade Associations:
Association for Geographical Information (AGI) (http://www.agi.org.uk/)
Peer reviewed academic journals:
International Journal of Geographical Information Science (IJGIS)Transactions in GIS
Trade press (available in the UK):
GIS EuropeNew Perspectives (marketing oriented)
Web based magazines:
GEOPlace.com (http://www.geoplace.com/)Directions (http://www.directionsmag.com/) (marketing oriented)
UK Conference web sites:
Digital Mapping Show (http://www.digitalmappingshow.com/)GIS 99 (http://www.gisexpo.com/)
24
References
1 Mapping Awareness, 1999a, ‘GIS Industry’, Mapping Awareness, March 1999.2 AGI, 2000, Association for Geographic Information web-site http://www.geo.ed.ac.uk/agidexe/term?271.For a discussion of definitions of GIS see Cowen, D.J., ‘GIS versus CAD versus DBMS: what are thedifferences?’ in Introductory readings in Geographic Information Systems D.J. Peuquet and D.F. Marble(eds.), Taylor & Francis, London 1990 and Chrisman, N.R., ‘What does ‘GIS’ mean?’ Transactions in GISvol. 3, no. 2, March 19993 Laurini, R. and Thompson, D, 1992, Fundamentals of Spatial information Systems Academic Press,London. p55-574 Worboys, M.F., 1995, GIS A Computing Perspective, Taylor & Francis, London5 For an interesting perspective on how this fits a model followed by all software packages see UserInterface Engineering, Inc. ‘Market Maturity’ Eye For Design Jan/Feb 1997, UIE, Massachusetts (alsoavailable at http://world.std.com/~uieweb/market.htm)6 IDC, 1999,7 Sonnen, D., 2000, http://www.directionsmag.com/article.asp?articleID=328 Sonnen, D., 2000, http://www.directionsmag.com/article.asp?articleID=409 Hendriks, P.H.J., 1998, Information Strategies for Geographical Information Systems. InternationalJournal of Geographical Information Science vol. 12, no. 6: 621-63910 McKinnon, A.C., Logistical Restructuring, Freight Traffic Growth and the Environment, chapter 5 inTransport Policy and the Environment (ed. David Banister), Routledge, London, 199811Blending OR/MS, Judgment, and GIS: Restructuring P&G’s Suppy Chain. Interfaces No. 27 Jan-Feb.1997, pp 128-14212 Anonymous, 1994, Putting logistics activity on the map, Traffic Management, May, Denver13 Sahoo, S.N., A GIS Approach towards Development of a Warehouse Location Model, ESRI UserConference 1998 Proceedings(http://www.esri.com/library/userconf/proc98/PROCEED/TO450/PAP419/P419.HTM)14 Business Geographics, 1999, supply chain – Exorcising an operational evil, Buisness Geographicshttp://www.geoplace.com/bg/1999/0999/999chn. Adams Business Media, inc15 Distribution Business, Geographic Analysis edges further on to the logistics agenda, DistributionBusiness, issue 7, Oct ‘9916 Institute of Logistics (Andersen Consulting & CMI) ‘Logistics Software Guide 1999 12th Edition’,Cambridge Market Intelligence, London, 199917 National Economic Research Associates (NERA) ‘Motors or Modems.’ NERA London, 199718 McKinnon, 1989 Physical Distribution Systems chapter 7, Routledge, London19 Eibl, P. 1996 Computerised vehicle routing and scheduling in road transport, Avebury20 Waters, N.M., 1999, Transportation GIS:GIS-T in Longley P., Goodchild M.F., Maguire D.J., RhindD.W. (eds) Geographical Information Systems: Principles, Techniques, Management and Applications, 2nd
edition, Harlow, Longman/New York, john Wiley 7 Sons inc. vol. 2: 827 - 82421 Distribution Business, Tracking and the importance of using the data it provides issue 1, Jan/Feb 200022 Op. Cit. 1423 MapInfo, 1995, Desktop Mapping: Transportation. Mapinfo Corp.24 HIDC, 1998, Worldwide Logistics , The Future of Supply Chain Services, Holland InternationalDistribution Council, The Netherlands25 Op.Cit. 1526 Birkin, M., Clarke,G., Clarke,M. and Wilson,A., 1996, ‘Intelligent GIS: Location Decisions andStrategic Planning’. GeoInformation International, Cambridge.27 Grimshaw, D.J., 1999, Bringing Geographical Information Systems into Business, 2nd edition28 Clarke, I and Rowley, J., 1995, A case for spatial decision-support systems in retail location planning,international Journal of Retail & Distribution Management vol. 23 no. 3: 4-1029 Distribution Business, ITTs – are you winning enough?, issue 1, Jan/Feb 200030 Peters, M., Cooper, J., Lieb, R.C., Randall, H.L., 1998, The Third-Party logistics Industry in Europe:Provider Perspectives on the Industry’s Current Status and Future Prospects. International Journal ofLogistics vol. 1, no. 1: 9-2631 Peters, M.J. and Wright, D., 1999, A research report on international air express distribution, CranfieldUniversity32 Monmonier, M., 1996, How to Lie with Maps, 2nd edition. University of Chicago Press, Chicago
