1. Map Data Overview

1.1. Background

Mapping is an ancient craft, with two early map makers, Mercator and Ortelius, providing much of the foundation for the mapping profession.

Over the last two decades, mapping has been extensively transformed by advances in computers and information systems. Today, map data, also referred to as spatial data (data that reference location) are the fuel for location-enabled applications. Good data—precise, accurate, and reliable—are the essential foundation for useful information. With robust map data underpinning their software applications, users can view maps and do much more:

• Plot a route to a destination—the fastest way, the most scenic way, avoiding freeways, and so on
• Sort categories, such as businesses, by time and distance from the user's location
• Dynamically modify a route to avoid obstructions, such as heavy traffic
• Pinpoint the locations of fleet vehicles, and dispatch the nearest vehicle or the one most likely to arrive first

Correct decisions often rely on complete, accurate, reliable, and up-to-date map data for navigation, traffic, point of interest (POI) lookup, social networking, entertainment and leisure, business applications like geomarketing, and industry vertical types of applications like telecom, mobile resource management, and the like.

Map data consist of topologically consistent vector or raster data that can be organized in various ways, such as in a relational database. These data are not only essential for calculating routes, avoiding traffic, and finding a nearby destination or POI, but are also useful for displaying or printing maps or for other types of geographic analysis.

In simplest terms, spatial data consists of points, lines, and/or polygons. Each of these data elements can be associated with one or more attributes that define a quality or characteristic of the element. The Britannica online dictionary defines an attribute as "A quality or characteristic inherent in or ascribed to someone or something." In a map context, an attribute can be anything that can be ascribed directly or indirectly to a location at a point, about a line, or about a polygon. These attributes include both static (relatively stable) dimensions, such as roads, signs, stores, landmarks, and rivers, as well as dynamic (changing frequently) dimensions, such as traffic and weather conditions.

1.2. Map Data Management

For database management, these points, lines, polygons, and associated attributes have data structures, additional data elements, and/or processing methods that make their data processing faster, less resource intensive, and easier to manage/update in order to complete their transition to map data. These include:

• Spatial topology – consists of spatial or geographic data structuring and maintaining line features such as road networks, polygon features such as district and county boundaries, and point features such as city and hospital locations.
• Geocoding – assigns a map position to an address record. The process matches and links records in two databases: an address database (without map position information) and a reference street map, thus tagging the address with the correct map position, such as latitude-longitude coordinates.
• Linear reference – keeps data on utility and transportation networks using a measurement from a fixed reference point along a route. Methods used include complete route mileage, geographic coordinates, or GPS (Global Positioning System) coordinates.
• Dynamic segmentation – locates a point or line segment by interpolating the distance between two known points and allows the recording of information along linear features, typically using GPS.
• Overlaying – Combines two or more maps within the same geodetic reference.. Historically, overlaying was done by transferring two maps to clear sheets and literally overlaying them on a light board. Digital mapping makes this process much more powerful by overlaying multiple layers of data and integrating them with other related business information. Two categories of map data best handled with layering are:
  • Routing network data, which are stratified into layers of successive detail, by functional class and speed category, to optimize data retrieval for fast route calculation
  • Cartographic data, such as water features, railroad networks, city limits, and political boundaries, which are stratified into layers to simplify and speed the data retrieval for rendering maps and for collecting and maintaining map data.

An example of overlaying a number of data layers is shown below.

2. The Core Process to Build a Map

2.1. Overview

Maps are built through sourcing, field collection, and continuous validation of the data.

2.2. Acquire Source Material

Sourcing is an efficient way to begin growth of coverage. NAVTEQ draws on thousands of data sources as its raw map data material, such as local, regional and national government agencies, postal services and private companies through partnerships and/or acquisitions. Key dimensions to successful sourcing include:

• Thousands of sources
• Robust supplier quality management
• Dedicated large scale production facility
• Automated processing tools

2.3. Enhance Through Field Validation

Sourcing is a start, but is not good enough. Sourcing has its limitations, including:

• Sources are notoriously variable in quality/accuracy
• Sourcing cannot capture navigable attributes

That is why you need a local field staff, to develop and use:

• Local presence to create pride of ownership
• Relationships with local sources
• Sophisticated collection technology

Mercator made his maps on bed sheets; modern mapmakers use slightly more sophisticated tools and techniques. Key dimensions to successful field validation include:

• Experienced field teams
• Advanced tools and technology
• Committed database investment
• Local ownership
• Eyes on the road

NAVTEQ has a dedicated software toolkit and a very comprehensive specification.

With over 200 attributes, map makers need tools to detect possible errors. Hundreds of individual validations occur in the toolkit to ensure the data is input correctly. Hundreds more occur before products are extracted from the NAVTEQ database.

2.4. Audit the Build Process

In Mercator’s day, required precision could be measured in miles if not tens of miles. Today, NAVTEQ is building maps that are relatively accurate down to 1 meter. This level of precision requires a significant investment in the ability to measure performance and understand the nuances of tracking.

To enable continuous improvement in map reliability, NAVTEQ:

• Establishes metrics
• Measures performance on those metrics
• Identifies root cause for poor performance
• Fixes and re-evaluates map data

Other key dimensions to successful build process auditing include:

• Certified evaluation methods
• Dedicated teams, tools and processes
• Ubiquitous use for continual improvement
• Integral part of the map data provider culture

• Dedicated NAVTEQ teams drive in the field to compare maps to the real world
• NAVTEQ compares its performance to benchmark performance levels
• NAVTEQ uses results to prioritize maintenance

2.5. Validate That It Works

At the end of the day, what matters is the accuracy of the map data in the applications in which it is used. Field validation and build auditing are essential parts of obtaining this accuracy, but the final “acid test” is seeing how the data are actually used in applications, from the eyes of the user. NAVTEQ spends tens of thousands of hours and drives hundreds of thousands of miles each year in this end-user experience part of the map build process.

2.6. Summary

Building a digital map is a continuous process of sourcing, driving, and validations to ensure we deliver the most accurate, high quality map to ground truth reality.

3. Structure and Attributes of Navigable Maps

3.1. Overview

NAVTEQ has a number of key map data dimensions of great interest to developers, including:

• Unique, powerful structure
• Unique geodesic reference system (WGS 84) with longitude and latitude in decimal degrees (10-5)
• Contiguous country borders
• Cross-border road networks
• Over 200 attributes

The structure and selected attribute examples within each structure category of the NAVTEQ map database are shown below.

NAVTEQ Database Worldwide Structure and Attribute Categories

Appendix A.1 - NAVTEQ Overview

Background

Founded in 1985 in Silicon Valley, California, NAVTEQ has a unique and eventful history rooted in technology, geography, hands-on research, and an infectious entrepreneurial spirit. From the beginning, NAVTEQ has been focused on capturing the reality of the road network to enable dynamic turn-by-turn routing. The company began by collecting detailed data for large metropolitan city areas. After Philips Electronics signed on as an early investor, the company grew strategically, establishing its first European office in 1991. The company’s significant growth led to the opening of offices in Yokohama, Japan in 1996.

Today, NAVTEQ, headquartered in Chicago, Illinois, USA has approximately 2,100 employees worldwide, and has major production facilities in Fargo, North Dakota, USA, and a support center in Yokohama, Japan. NAVTEQ has achieved ISO 9001: 2000 certification of all of its main operating locations.

NAVTEQ provides products and services that address several parts of the location-based services (LBS) value chain. At its core, it provides the digital map data content that forms the heart of all location-based services. NAVTEQ provides this data in a variety of formats directly to customers as well as to geospatial platform providers such as Autodesk and deCarta who in turn offer developers various tools to develop their LBS applications.

NAVTEQ also provides a variety of technical, business, and application development support services including:

• Business development support
• Navigation advisory services
• Product development support
• Testing services
• Channel development services
• End-user services

More information on these services is available at:
http://developer.navteq.com.