• System Description
• Type of Information Provided
• Information Sources
• Usage
• Effect on Trip-Making
• Cost
• Overall Assessment
• References

SYSTEM DESCRIPTION

Pre-trip travel information is an Advanced Traveler Information System (ATIS) user service. Its objective is to inform travelers of traffic and transit conditions, so they can best assess travel options before selecting a route, mode, time-of-day, or deciding whether to make a trip.

Traditionally, transit information (routes and posted schedules) has been supplied by local transit providers, while traffic information is reported by radio stations during commute hours. Pre-trip ATIS can enhance travel information by improving content and media. For example, posted routes and schedules are usually available in printed form, but brochures are often available only at selected locations. In some instances, travelers can access a transit service desk by telephone or Internet. Traffic information is targeted to a broad audience primarily through radio, which means that it is usually not detailed or timely enough to serve trip-planning purposes, except in the cases of major incidents.

Pre-trip ATIS makes use of ITS technology and infrastructure to deliver real-time, customized information to several different travelers. This review examines four different services: the Los Angeles SmartTraveler, the Boston SmarTraveler, TravInfo, and traffic information provision through the Internet of World Wide Web.

Los Angeles SmartTraveler:
The Los Angeles Smart Traveler Field Operational Test was deployed between January 1994 and January 1995. It offered information on real-time traffic conditions, transit schedules and route planning, and ridesharing services via public kiosks, telephones, and PC-modem links. It is important to note that this service was not continued beyond the period of the field operational test.

The test covered the Los Angeles Basin area affected by the Northridge Earthquake: the San Fernando Valley, the Santa Clarita Valley, Palmdale, and West Los Angeles. The test was based on a public-private partnership among the California Department of Transportation; the Los Angeles County Metropolitan Transportation Authority; the State of California Health and Welfare Data Center; Commuter Transportation Services, Inc.; Pacific Bell and Pacific Bell Information Services; IBM Corporation; and North Communications. Field test funding was provided by the State of California, through the California Advanced Public Transit System Program, and the Federal Highway Administration through earthquake relief funds.

Boston SmarTraveler:
The Boston SmarTraveler was initially planned as a one-year operational test, but it has been in operation continuously since January 1993. The system provides real-time, route-specific traffic and transit information to travelers in the Boston, Massachusetts, Metropolitan Area. It is accessible only by telephone (via a toll-free number) within a 1,400 square mile service area, including two million drivers. The findings reported here correspond to operations between March 1994 and December 1994. They are based on several surveys, including: a follow-up questionnaire of current users (sample size 447), a caller intercept questionnaire (sample size 547), and a marketing questionnaire (sample size 1,920). The Federal Highway Administration, the Massachusetts Highway Department, and the private sector provided SmarTraveler operational funding. Total funding was approximately $6 million through December 1994. Service is provided free of charge to the general public, excluding phone call costs. Since 1997, SmarTraveler projects have begun in Cincinnati, Philadelphia, and Washington, D.C.

TravInfo:
TravInfo is a regional traveler information system for the San Francisco Bay Area. It was a Caltrans and FHWA field operational test, implemented over a two-year period from September 1996 to September 1998 through a partnership of public agencies, research institutions, and private firms, under the Metropolitan Transportation Commission's leadership (i.e., the local MPO). Additional funding was obtained from TEA-21 funds, along with local area matches, for another five years.

Travel information is gathered from a variety of sources. The most important source is the California Highway Patrol Computer-Aided Dispatch (CHP CAD), which includes roadway incident reports and other records, such as: incident date, CHP CAD identification number, CHP CAD time stamp, incident type, and incident location. CHP obtains information about roadway incidents from a variety of sources, including 911 calls and Metro Traffic Airborne. When a CHP CAD incident is recorded, TravInfo TIC operators obtain it from their CHP CAD terminal. Then, TIC operators confirm the incident and make a decision whether or not to enter it into the TravInfo TIC computer system. At present, such incidents are entered manually, since there is not yet an electronic linkage between the CHP CAD system and the TIC system.

Travel information is available via telephone (or TATs) and from several private-sector partners who customize it for various consumer markets (these include some of the Internet-based providers discussed below).

Traffic/Traveler Information on the Internet:
Traveler information is provided by a number of private, for profit, companies. The industry has undergone some major changes in the last five years in the wake of "digital revolution". The information used to be predominantly traffic (as it still is) information provided on a regional level. Generally, the information was provided to radio and TV stations in exchange for air-time slots which are then sold to advertisers. This is still the case but with the recent explosion in media and communications devices (web, cellular, palm pilots, cable and satellite TV, soon in-vehicle satellite radio), information dissemination methods have multiplied and the traditional payment or revenue-generating schemes are also undergoing changes. Web-based information is generally supported by advertising on the providers' page (SmarTraveler) or is licensed (Metro/Etak) to information or entertainment web sites (such as web radios or TVs, on-line newspapers etc). Information provision is also becoming more accurate and increasingly personalized to users' needs and targeted to specific markets. The three main new traveler information provision markets are the internet, wireless devices (phone and palm pilots) and in-vehicle communications.

Travelers in several U.S. metropolitan regions can now access real-time traffic information via numerous web sites. Most of these sites have been designed in collaboration with local transportation authorities or nationwide traffic information providers. The information is usually displayed on a map, color-coded to identify network speeds, congestion levels, or incident locations. In addition, some of these sites include CCTV videos, updated as frequently as every few seconds, but more commonly every few minutes.

Internet sites provide a wealth of information: congestion, speed, and incident location maps, detailed incident information, mountain pass reports and current conditions, freeway photographs, transit routes and schedules, current and forecasted weather conditions, road construction updates, and sometimes even trip planning tools.

Web-based traveler information dissemination has developed very rapidly in the last few years along with the boom in all internet services. This appears to be a particularly attractive way to rapidly disseminate pre-trip traveler information. It will also fast become an important means of en-route traveler information and dynamic route guidance as cars and cell phones are increasingly equipped with internet connections.

The following are a few examples of Internet traveler information sites

• Smart Trek: Up-to-the minute multimodal traveler information provided by the Washington State Department of Transportation (project leader) and 24 private and public organizations.
• Metro Networks and Etak:Etak Inc. is a publisher and provider of digital mapping technology. Metro Networks is provider of traffic information and news. Etak, in partnership with Metro Networks, provides real-time information for 65 metropolitan areas. Both were private partners in the San Francisco Bay Area's TravInfo field operational test.
• SmarTraveler: Offers real-time traffic information as well as other multimodal information (transit, carpooling, parking etc.) for 18 major cities.
• TravInfo:this site describes the TravInfo projects, and lists all sources of traffic information that have derived from it.
• KPIX (San Francisco and Oakland): this is the CBS affiliate for the San Francisco Bay Area. It uses Etak maps and TravInfo information, supplemented with traffic information from Shadow Traffic.
• Maxwell Technologies: this site uses Caltrans and CHP freeway surveillance data to supply real-time traffic information for the Los Angeles / Orange Counties region.

Pre-trip information usually consists, in the case of transit, of posted schedules and routes. The Los Angeles SmartTraveler assisted with transit trip planning between any given origin/destination pair requested by a user. Neither of these services included real-time waiting times. Traffic information, however, was based on real-time conditions:

• The Boston SmarTraveler provides route-specific information on 21 highway segments, including descriptions of travel conditions, travel times, and incident location and expected duration. Information is continuously updated on weekdays between 5:30 A.M. and 7:00 P.M., and 12:00 noon to 7:00 P.M. on Sundays. Users access information by entering a three-digit code, unique for each highway section or transit service.

• TravInfo disseminates traveler information in three ways: 1) the interactive Traveler Advisory Telephone Systems (TATs), 2) TravInfo Web site, and 3) service providers who registered to participate in the project and tap into the center's database. TATs provides regularly updated information on current traffic conditions, carpooling, highway construction reports, bicycle programs, San Francisco International Airport ground transportation, and a direct connection to the region's transit and paratransit operations. During emergencies and special events, information is added. The TIC operates 24 hours a day, seven days a week.

• The Los Angeles SmartTraveler displayed a map of Los Angeles freeways, color-coded to indicate congestion level.

• Internet sites provide a wealth of information: congestion, speed, and incident location maps, detailed incident information, mountain pass reports and current conditions, freeway photographs, transit routes and schedules, current and forecasted weather conditions, road construction updates, and sometimes even trip planning tools.

INFORMATION SOURCES

Transit information was obtained from local transit authorities, the Los Angeles County Metropolitan Transportation Authority and the Bay Transportation Authority. Traffic information is obtained from several sources. In Los Angeles, Caltrans provides average freeway speeds derived from loop detector data. In Boston, information is obtained from live and slow-scan cameras located along several freeways, three fixed-wing aircraft, 200 vehicle probes (i.e., regular commuters who report travel information to SmarTraveler operators), and various state and regional transportation agencies (e.g., the State Police, the Bay Transportation Authority, the Port Authority, the Turnpike Authority, and the Highway Department).

Internet sites use a variety of sources, including the state's freeway surveillance systems and private traffic information providers, such as Shadow Traffic and Metro Traffic. Real-time information, including CCTV camera views, is usually updated every few minutes.

Boston SmarTraveler:
As of December 1994, the system received approximately 4,000 calls per month. Usage increases on bad weather days; indeed, it peaked during the winter of 1993-1994, due to a series of severe snowstorms. Cellular phone calls represent approximately 60% of all calls (customers of one local cellular phone service provider call free of charge). Usage has increased over time due to an increase in number of users and higher calling frequency per user. SmarTraveler has substituted other traffic information sources for cellular phone users, but not traditional phone callers.

A marketing study was conducted to understand what characteristics encouraged SmarTraveler use. Researchers found that users are more likely to make long trips: 90% of users make trips in excess of 12 miles, compared to 62% of non-users; and 95% of users make trips over 20 minutes in length, compared to 73% of non-users. Those who call SmarTraveler are more likely to make trips to areas where service coverage is extensive (i.e., 75% or more of the route).

Two-thirds of calls are made for work-related trips, with twice as many calls regarding evening versus morning commutes. This is because congestion is worse, closer to trip origins, in the evening than the morning; a call from a work is likely to be free; and convenient access to other traffic information sources is limited at work.

Two-thirds of SmarTraveler users have access to a cellular phone in their cars, and 50% of all calls are made from a vehicle. Moreover, cellular calls are twice as likely to be made for trips of at least 50 miles. Compared to the Boston Metropolitan Area population, users are disproportionately high-income males in the 35 to 54 age cohort.

Call frequency varies among users: 43% call whenever they are making a specific trip, while 25% call infrequently. Occasional users place calls when the weather is bad, they learn about an accident or delay from other sources, or they have a strict arrival time.

TravInfo:
At the end of the TravInfo field test, surveys showed that the vast majority of Bay Area households were not aware of the TravInfo Traveler Advisory Telephone Service or traffic Web sites. Of the 9% of Bay Area households that were aware of TravInfo, very few had actually tried it because respondents did not remember the telephone number.

Surveys showed that the vast majority of Bay Area households were not aware of the TravInfo Traveler Advisory Telephone Service or traffic Web sites. Of the 9% of Bay Area households that were aware of TravInfo, very few had actually tried it because respondents did not remember the telephone number.

Monthly call volumes ranged between 50,000 to 65,000. Volumes remained consistent during the field test except on two occasions: the Bay Area Rapid Transit (BART) District strike in September 1997 and the floods in February 1998, when volumes rose significantly, but temporarily.

Among those who used the TravInfo Service, satisfaction was consistently high. They rated quality of information to be far superior to radio or television reports and perceived it to be useful in trip planning. Because of this, over 80% were repeat users. Many TravInfo users were also cell phone customers: 31% of calls came from cellular phones, and this number increased to 41.8% by the program's end.

Los Angeles SmartTraveler:
SmartTraveler could be used via public kiosks or PC-modem links, using software distributed to approximately 500 people (with permission to make copies). Tracking modem usage is limited to system access counters. On average, the system logged over 400 calls on weekdays and about 150 calls on weekends. Calls were more frequent during peak commute times and more likely to be made for trips home from work. Personal interviews were conducted with some individuals who were mailed the software. Out of 21 people contacted, only three used the traffic information software; the rest had not yet installed it or had installation problems, including hardware/software incompatibility. Those who did not install it said they got the information from other sources or didn't have the time. Some complained of poor technical support from Caltrans.

Kiosk usage was tracked using the kiosks' internal log and user intercept questionnaire. Each kiosk was available an average of 131 days (out of a maximum of 224). Each time the main menu was selected counted as a user transaction. Kiosks averaged 25.3 transactions per day, and each transaction lasted for approximately five minutes on average. Usage declined as the test progressed, at a rate of 0.21 transactions per day, per week. Overall, each kiosk was actively used about two hours per day (the rest of the time it was idle or broken).

Usage patterns by location suggest that kiosk access was a leisure activity and was positively correlated with pedestrian traffic. Indeed,use was low in office locations and high near shopping centers and discount stores. Usage patterns suggest that most were novices, using the kiosk more out of curiosity rather than in need of travel information. Furthermore, the help feature was frequently requested, and about 16% of all transactions ended at the main menu (that is, users did not delve deeper into the system). About half of all requests for transit route and schedule information ended in a trip itinerary printout. In contrast, about one in ten ridesharing requests ended in a ride match printout.

An intercept survey of kiosk users found that they were twice as likely to use transit to travel to work as the general population, and their commutes were longer, both in time and distance. Not surprisingly, transit route and schedule information was the most frequently requested, although in the questionnaire, respondents reported requesting freeway condition maps more than transit information.

The Los Angeles SmartTraveler evaluation did not explore information impacts on traveler decision making. Due to the scale of the test and usage levels, the system had minimal impacts on traffic congestion and transit usage. Similar results are true for Boston SmarTraveler and TravInfo. Moreover, a marketing study of the Boston SmarTraveler estimated, even with a high service awareness among the population, use would likely remain too low to have a significant impact on traffic congestion or air quality.

• 50% of users call to verify that their planned route is feasible;
• about 30% of callers use the information to choose between two or more alternative routes;
• 15% of users have changed their usual route, and 14% have changed their departure times, in response to information about congestion;
• upon learning of a possible trip delay, about 6% of users have called to let others know they will arrive later than anticipated;
• very few have changed modes or cancelled their trips;
• about 2/3 of users list anxiety-reduction as a benefit from using the system;
• about 50% of users indicate the information lets them avoid traffic problems, save travel time, or arrive on time at their destination; and
• only 7% of users believe they don't receive any benefit from calling SmarTraveler.

TravInfo was able to influence travel behavior far more effectively than radio or television traffic broadcasts. Twenty-five percent of those who obtained relevant information from radio or television changed their travel behavior, while nearly twice as many of the TravInfo callers (45%) reported to have altered their trips after obtaining information specifically on their routes.

COSTS

Cost information is available for the Los Angeles operational test only. Average unit costs per user interface are detailed below. For comparison purposes, consider that direct costs to the LA County Metropolitan Transportation Authority for a transit service call are estimated at 32 cents.

Kiosks:
Direct costs per kiosk varied depending on the features available. There were four types of kiosks, with direct purchase costs ranging from $15,361 to $28,128. The basic kiosk was priced at $18,150. There was a one-time site preparation, negotiation, freight, and installation fee of $1,543 per kiosk and a maintenance fee between $214 and $303 per month per kiosk. Cost per use was estimated between $2.0 and $4.6 dollars, depending on the lifetime of each kiosk and assuming a demand level similar to the test (see Figure 1).

Source: Giuliano et al., 1995

PC Software:
PC software development costs totaled $63,400. Because 500 copies were distributed and only four copies were in use, cost per copy was approximately $500.

OVERALL ASSESSMENT

It appears that pre-trip travel information, while adding value to each individual users' decision-making process, is not likely to have a significant impact on a system level (e.g., impact congestion or emissions levels). While market penetration has been relatively low thus far, Boston SmarTraveler studies suggest that this is not likely to change when information is more widespread.

The absence of system-wide effects, coupled with high costs, suggests that public agencies should be careful in spending tax dollars to finance these projects. If there is a market for pre-trip travel information, services should be provided by the private sector. A commercial model has been adopted in a few cases (reviewed here), notably TravInfo and Boston SmarTraveler.

For instance, the TravInfo system was not as efficient as originally envisioned due to its heavier-than-expected dependence on manual entry of traffic information into the Traveler Information Center. TravInfo needs an automated system that is flexible enough to keep up with rapidly advancing technologies, which will likely require its system components to be enhanced and upgraded. Despite its challenges, the TravInfo system is now in a transitional phase leading to full deployment of a regional traveler information system for the Bay Area.

The U.S. federal government continues to support ATIS services in several metropolitan areas, particularly through its Model Deployment Initiative. These applications are larger in scope and expected to reach more users (e.g., TravInfo deployment). Thus, it is still too soon to draw conclusions about the effectiveness of pre-trip travel information.

REFERENCES

Guiliano G, R.W. Hall, and J.M Golob. Los Angeles Smart Traveler Field Operational Test Evaluation. PATH Draft Research Report No. D95-35. Berkeley, California: University of California, Institute of Transportation Studies, 1995.

Multisystems, Inc. Evaluation of Phase III of the SmarTraveler Advanced Traveler Information System Operational Test. Final Report. Boston, Massachusetts: Multisystems, Inc, 1995.

Yim, Y.B. and M. A. Miller. Evaluation of TravInfo (Trademark Symbol) Field Operational Test: Final Report. PATH: UCB-ITS-PRR-2000-7. April, 2000.

Author: Rosella Picado. Last update: 05/30/00