Executive Summary

Advanced Traveler Information System (ATIS) is one of several Intelligent Transportation System (ITS) technologies that offers users integrated traveler information before and during travel, thereby providing a wider range of choices about how, when, and where to travel based on individual interests and needs. One of the major reasons ATIS services has garnered public and professional interest is the concern created by the continuing disparity between the growth in surface transportation travel demand and the relatively minor addition of travel capacity. This combination has resulted in increased regional roadway congestion, greater uncertainty in travel time estimates, and higher real or perceived costs in safety and productivity. Increasing transportation capacity by building roads and other related infrastructure is not a feasible solution in many urban areas due to the high costs as well as environmental and associated societal concerns. Alternative solutions are necessary.

To be effective, ATIS systems must work with a broad set of source data and information, combine and qualify the information to yield better traveler information, and disseminate the information when needed by travelers. One component of this complex process is data fusion.

The purpose of ATIS data fusion is to combine data (in the broadest sense of the term) to estimate or predict the state of some aspect of the surface transportation world. These estimates may include statements about current or future vehicular speeds, mean speeds, vehicular classifications and volumes on selected roadway segments, environmental information, transit system performance, and similar topics of interest to travelers.

The overall effectiveness of data fusion needs to be evaluated in a systems context, taking into consideration the overarching system mission and purpose, architectures, data processing capabilities, data validation and verification, human-system interface, and institutional arrangements. A study was completed, examining these issues, the findings of which can be found in this report. The process of the study was three fold. First, a literature review was conducted of the ATIS and data fusion fields to examine current practices. The review also included an examination of relevant field case studies and discussions with selected ATIS practitioners to determine the extent and direction of their data fusion interests and applications. Second, an appropriate ATIS data fusion model was developed, along with guidelines to enable a multitude of source data to be fused to create ATIS products and services. The model describes ATIS data fusion using five distinct levels of functional activities. Third, appropriate metrics were identified that describe quantitatively and qualitatively how data quality can be verified, modeled, and processed so that traveler information products can be considered more reliable and useful.

The major findings that arose from this study include:

A general awareness of ATIS data fusion and its purpose exists.
The general public, and travelers in particular, are predisposed to use better travel information, although it is unclear under what business model and range of products and services.
Data fusion conducted at transportation and environmental public agencies involves basic fusion functions such as spatial or temporal alignment of input data. Assessments of ATIS services at selected Metropolitan Model Deployment Initiative sites and Field Operational Test sites indicate positive support and public benefits.
Third-party ISPs perform additional data fusion activities, but the specific techniques, accuracy, and usefulness are not readily discernable due to proprietary restrictions.
Two general perspectives exist for ATIS data fusion: a data-centric view and a model-centric view. These perspectives are not mutually exclusive.

Agencies are migrating to the use of the NTCIP and associated data element and message set standards, but will likely initiate their use based on industry acceptance and agency confidence in their utility and longevity.
Data quality policies and procedures are not comprehensive. Practices are limited primarily to “find and fix” tactics on the more egregious data elements.

The findings point out the need for a more comprehensive ATIS data fusion methodology that would allow for increased cross-disciplinary communication and research sharing. A proposed ATIS data fusion model, based on the JDL process model, was offered to help bridge this gap. Moreover, specific data fusion techniques, appropriate for the ATIS context were identified and qualitatively assessed using multiple criteria such as ease of implementation and potential usefulness.

General guidelines for data fusion architectures are presented in the report. The wide variety and combination of ATIS fusion applications and associated architectural components do not allow for a prescriptive, detailed definition of architectural components and fusion techniques. This prescription is best handled through a more structured, system engineering (SE) process involving all stakeholders and design experts. Key elements of the SE process are outlined.

Input data quality continues to be a hindrance to the offering of more advanced ATIS services. Current practices focus on “fix and find” methods without long-term, systemic attention to data quality issues. One of the key issues is the different perspective held by stakeholders on the level of satisfaction with the existing data quality and the associated remedies and costs to make improvements. Greater awareness and understanding of the issues are needed before prescribing remedial action, if any. Resolution of data quality issues will require partnerships among the data owners and users to reach a shared solution.

The report concludes with proposed future actions for enhancing ATIS data fusion practices, summarized into three categories: technological, institutional, and economic opportunities.

Glossary

ATIS Advanced Traveler Information System

AVI Automatic Vehicle Identification

AVL Automatic Vehicle Location

CATV Cable Television

CVO Commercial Vehicle Operations/Operator

CORBA Common Object Request Broker Architecture

DATEX-ASN.1 DATa Exchange in Abstract Syntax Notation One

DMS Dynamic Message Signs

FHWA Federal Highway Administration, USDOT

FOT Field Operational Test

FTP File Transfer Protocol

GPS Global Positioning System

HAR Highway Advisory Radio

HTML HyperText Markup Language

IEEE Institute of Electrical and Electronics Engineers

IM Incident Management

ISP Internet Service Provider

ITS Intelligent Transportation Systems

IVR Interactive Voice Recognition

JDL Joint Directors of Laboratories (U.S. government laboratories)

MMDI Metropolitan Model Deployment Initiative

NTCIP National Transportation Communications for Intelligent Transportation System Protocols

PDA Personal Digital Assistant

RAID Redundant Array of Inexpensive Disks

RFID Radio Frequency Identification

SDO Standard Development Organization

SE Systems Engineering

SNMP Simple Network Management Protocol

SQL Structured Query language

STMP Simple Transportation Management Protocol

TFTP Trivial File Transfer Protocol

TMC Traffic Management Center

USDOT United States Department of Transportation

VMS Variable Message Signs

W3C World Wide Web Consortium

WIM Weigh In Motion

WWW World Wide Web

XML Extensible Markup Language

Section 1 Introduction

Figure 1﷓1 Greater Roadway Congestion Has Generated Increased Interest in Advanced Traveler Information Systems (ATIS) Many issues will affect the performance of the 21^st century surface transportation systems, particularly highways and transit systems, in the United States and in similar industrialized nations. A paramount concern is the growing congestion on highways and roads created in part by increased travel demand coupled with a modest increase in travel capacity. The effects of this disparity are captured in a number of measures and perceptions, including visible and consistent roadway congestion, the loss of personal and professional time, environmental degradation, and general traveler frustration. An often-cited report of this phenomenon is the Texas Transportation Institute’s report on urban mobility, which estimated the total cost due to roadway congestion at $78 billion for the 68 largest urban areas in 1999[1]. A recent national satisfaction survey conducted by the Federal Highway Administration notes 65 percent of those surveyed are satisfied with the major highways they travel most often. However there is greater dissatisfaction due to heavier traffic flows and delays, especially circumstances caused by workzones and roadway incidents[2]. Substantial debate has occurred about the proper course of action to address these concerns and trends.

Advanced Traveler Information Systems (ATIS) is one of the many components of Intelligent Transportation Systems (ITS). The purpose of ATIS is to provide practical and timely help to travelers in an integrated, multi-modal environment using the goals, principles, and practices of the National ITS Architecture[a]. Effective traveler information would support the needs of many travelers. The information would assist users in selecting their mode of travel (car, train, bus, etc.), route, and departure time, as well as provide supplemental information about the weather, congestion indicators, and other issues affecting their travel.

1.1 ATIS Concepts

Effective traveler information services support many types of information[b] requests and categories of travelers, and combine multi-modal information in an effective and timely manner. Information may be provided in a number of ways, including pre-trip (static) information and real-time information. Static information comes from such sources as transit schedules, planned workzones, and known road closures. Dynamic information comes from a variety of sources including roadway-based sensors, surveillance equipment, and driver information. The information assists travelers in selecting their mode of travel, route, and departure times. Figure 1-2 illustrates the range of data sources, processing and uses of ATIS information. The figure depicts the various sources of data (left-hand side) that are collected and centrally processed (central part of figure) to yield integrated information about the current and future travel conditions, such as roadway congestion and transit schedules. This information is broadcasted or disseminated to travelers, allowing them to make informed choices about when, where and how to travel.

Figure 1-2 Informal Illustration of Sources and Uses of ATIS Information[3]

The National ITS Architecture has identified nine market packages that incorporate ATIS user services [4]. These packages can be further broken down into static or dynamic services. For the purposes of this report, the six dynamic related services will be emphasized.

Dynamic, Traffic Related Market Packages	Definition and Key Points
Broadcast Traveler Information	§ Disseminates near real time traffic information over a wide area through existing infrastructures and low-cost user equipment such as radio or cellular phones § Information flow is one way
Interactive Traveler Information	§ Responds to a user’s request with tailored information, using wide range wireless and wireline communication systems
Dynamic Route Guidance	§ Provides advanced route planning and guidance that is responsive to current conditions § Relies on a digital receiver, map databases and a variety of in-vehicle computational systems and devices
ISP-Based Route Guidance	§ Similar to Dynamic Route Guidance, but it moves the route planning function from the user device to a service provider § The user has the option of equipping their vehicle with the map databases and location determination capability
Integrated Transportation Management/ Route Guidance	§ Used by both public consumers and traffic management centers § Traffic management centers use it to optimize traffic control § Consumers benefit from advanced route planning and guidance based on current conditions
In-Vehicle Signing/Message Exchange	§ Based on communication between roadside equipment and in-vehicle devices § Roadside equipment communicates with the traffic management subsystem in order to provide traffic and travel advisory information to the in-vehicle device

Figure 1-3 The Range of Dynamic Traveler Information Services

In addition, the National ITS Architecture can be categorized by two types of basic services: i) traffic and road condition information; and ii) location, navigation, and route guidance information.

1.1.1 General ATIS Model

A number of individuals, organizations, technologies, and processes must be assembled to develop, implement, and sustain effective and valued ATIS services. Appropriate sensing and surveillance equipment is required. Public-private partnerships are needed to gather and disseminate timely, useful traveler information based on public and private data sources and data processing. Multiple vendors and technologies necessitate the use of accepted standards and protocols to enable interoperability and functionality.

The following diagram offers a generalized functional model of ATIS. Read from left-to-right, Figure 1-4 indicates the wide range of available information sources, the data fusion activity, the opportunity for value-added services from public or private agencies, and the dissemination of traveler information through multiple means and mediums. Data collection has traditionally been conducted by public agencies (e.g., highway and transit agencies) primarily to meet their agency objectives for management and operation within their service areas and responsibilities. Recently, private agencies have supplemented public agency data to provide more complete coverage of a region or subcorridor. Data fusion, in general, refers to the process of combining information from a variety of sensors and processing the data to yield better estimates describing the state of the transportation system. The value-added function may include a variety of activities, such as repackaging basic traveler information for consumers in a form more available (e.g., cellular phone, websites, or mass media) or understandable (e.g., graphical displays or site-specific congestion metrics). Moreover, additional content may be added (or fused in the earlier stage) to enhance basic traveler information, such as confirmation of incidents, better microscale weather information, geo-location of events, and recommendations on alternative routes or departure times. The resulting information can be distributed to consumers through a variety of media, with the opportunity for specialized equipment and software allowing the receiver to customize the traveler information.

Figure 1-4 A Generalized Model of ATIS Functions and Services4

The simplicity of this ATIS model is complicated by a variety of factors, including institutional and regulatory issues, legal concerns, partnership agreements (among all combinations of public and private organizations), consumer expectations, contracts for the types and quality of traveler information delivered, and the changing technology base underpinning this service delivery[c].

1.2 General Concepts of Data Fusion

In general, the purpose of ATIS data fusion is to combine data (in the broadest sense of the term) to estimate or predict the state of some aspect of the surface transportation world. These estimates may include statements about current or future vehicular speeds, mean speeds, vehicular classifications and volumes on selected roadway segments, environmental information, transit system performance, and similar topics of interest to travelers.

Within the graphic depiction of the data fusion function illustrated in Figure 1-4, a complex set of activities is occurring and will be elaborated upon in subsequent report sections. Major data fusion functions include:

Raw Data Collection Transmitting and receiving error-free[d] data from field sensors or other locations

Data Identification Matching the sensed data with the source or adjusting for missing data values

Data Alignment Configuring identified sensor data to a common spatial and temporal reference/origin, as well as transforming data into compatible representations and/or languages (e.g., XML[e])

Data Combination Performing various association analyses (e.g., statistical correlations, pattern recognition, etc.) to improve detection, classification, and tracking of entities of interest (e.g., cars, surface temperature readings, etc.)

State Estimation Predicting the kinematic (time and/or spatial) performance of an entity of interest

Performance Assessment Applying techniques to assess fused data quality and fusion processes.

1.3 Opportunities and Challenges of ATIS Data Fusion

ATIS data fusion is an emerging and evolving field. Some of the basic benefits of ATIS have been garnered through the savvy, cost effective design of regional ITS architectures and the value-added application of market-proven techniques to meet customers’ needs. However, there are opportunities for greater ATIS data fusion applications. Prospects include the increased collection of usable data from sources other than the installed sensor and surveillance networks owned and operated primarily by public agencies. Wireless technologies, coupled with the increased acceptance of data standards and protocols, will offer the potential for easier reporting and access to customized ATIS information. Technological and data processing advances in affiliated scientific and engineering disciplines, such as database management and web-based commerce, coupled with cost reductions in computer and telecommunication equipment, can provide a foundation for greater ATIS data fusion applications and value-added services. Moreover, public agencies can contribute substantially to ATIS through their internal systems for monitoring and improving the performance of the transportation system, such as traffic signal control, incident management, bus fleet performance or the analysis of ADUS[f] information. However, several challenges may hinder the accelerated use of data fusion for ATIS. These obstacles include the improvement of institutional/organizational collaboration, the timely establishment and use of standards and protocols that serve the widest set of potential ATIS customers, concerns about data quality, and establishing a delivery model that provides real-time, quality ATIS services in an environment when basic traveler information is free.

1.4 Section 1 References

Section 2 Study Approach

2.1 Study Objectives

It has been suggested that improved ATIS data fusion techniques and processing will improve the overall quality, timeliness, and usefulness of traveler information. In particular, with increased use of multiple sources of data, properly combined, fused, and quality-controlled, a more reliable, real-time set of traveler information can be produced, which will be valued more than the information services currently available to the traveler. Increased attention to data fusion will also better inform agency planning and guide ITS architecture development and deployment.

The purpose of this study was four fold. First, conduct a literature review of the ATIS and data fusion fields in order to summarize current ATIS data fusion practices. The review also included an examination of relevant field case studies and discussions with selected ATIS practitioners to determine the extent and direction of their data fusion interests and applications. Second, develop an appropriate ATIS data fusion model[g] and guidelines to enable a multitude of source data to be fused to create ATIS products and services. The model should be able to account for the challenges of multiple sources of data, varying types of quality, institutional impediments, and evolving standards and practices associated with the National ITS architecture. Third, identify appropriate metrics that describe quantitatively and qualitatively how data quality can be verified, modeled, and processed so that traveler information products can be considered more reliable and useful. Fourth, provide general guidelines on the development of an ATIS data fusion system.

As a result of the study, a phased model and guidelines is available to assist agencies with ATIS data fusion considerations. These considerations include the development of specific ATIS data fusion goals and subsystems in the context of an overall ITS mission and supporting architecture. Moreover, agencies will likely gain an increased awareness of ATIS data fusion capabilities, limitations, and resources for further inquiry.

This study was conducted during December 2000 until August 2002. During this period substantial changes were occurring nationally and internationally in three fields closely related to ATIS data fusion, namely telecommunications, computing, and web-based commerce. These factors are mentioned since the annotated literature listed in the appendix presents a potentially divergent picture of growth, opportunity, challenges, and retreat, which may confuse the reader without an explicit mention of the study period.

2.2 Study Approach

Data collection was the first study task and involved a literature review, examination of case studies, and discussion with ATIS experts. The literature review was conducted using web-based searches, reference list back-chaining, a search of relevant transportation databases, and discussions with knowledgeable individuals to identify key documents and source materials. The information was screened for relevance and then organized and sorted based on assigned keywords, such as data fusion, data quality, etc. A summary of the major findings is presented in Section 3. The appendix contains an annotated bibliography of the sources.

Case studies, primarily from the Metropolitan Model Deployment Initiative (MMDI) evaluation program, were reviewed for ATIS features and applications. The cases were examined for state-of-the-practice and specific data fusion activities performed by either public agencies or private firms. In the case of private sector firms, little or no information was available regarding data fusion techniques, as these were considered highly proprietary and competitor-sensitive.

Structured interviews were conducted with representatives or individuals knowledgeable about ATIS deployments. The interviews were designed and the data collected not to be generalizable, but instead to document the practices and attitudes of some of the major public sector ATIS practitioners. The interviewer elicited a description of ATIS services provided by the agency/organization, identification of fusion techniques (as appropriate), metrics on data quality, difficulties in implementing ATIS services, the extent of conformance or use of ITS standards, and remarks on future activities, especially for data fusion. Interviews were conducted with individuals from Seattle (Washington), San Francisco (California), Los Angeles (California), Houston (Texas), I-95 Corridor Coalition (Virginia/Maryland representatives), Hampton Roads-Smart Travel Center/I-81 (Virginia), and Transcom (New York, New Jersey, and Connecticut). The aggregate findings are reported in Section 3.