Ease of Use vs. Increased System Complexity

April, 2004

A Look at the Influence of AVI RFID

Part II of a two-part series on the benefits versus the increased integration requirements of using AVI RFID (automatic vehicle identification radio frequency identification). Part I focused on the current technology changes and longer-term expected pervasiveness; Part II focuses on some of the integration issues associated with using a vehicle-mounted "tag" for multiple purposes, including parking revenue control.

Although there is a lack of technical interoperability on a national level, on a regional basis, we find widespread deployment of a given technology, e.g., "EZ-Pass" in the mid-Atlantic and New England states. Significant consumer acceptance of AVI on a regional basis opened the door for secondary application uses of the tag, such as security/access control in gated communities and corporate parking facilities, fast-food purchases and parking revenue control. The initial use of AVI in parking revenue control was for monthly parkers, but with the expanding practice of connecting to the clearinghouse used by the local toll agency, transient parkers are now also able to use their tags at locations other than their "home" garage. This business model is successfully operating today in two major metropolitan areas: Dallas/Ft. Worth, with both major airports participating; and the major airports operated by the New York/New Jersey Port Authority.
AVI tags today range in cost from $9 to $30 to the end-user, depending on the operational characteristics required by the end-user. (A truck company, for example, has different entry turning area requirements from a gated parking garage with concrete islands.) As the number of SCM RFID tags increases, the cost of manufacturing will drop, with subsequent price benefits to end-users, with $3 to $5 tag prices foreseeable in the not-too-distant future. Unless you have a vehicle population in the 10,000+ range, the real cost of today's small-to-midsize AVI system is in the readers, not the tags. Even in these early days of low-volume SCM RFID deployment, we have seen a corresponding reduction in tag prices from a previous low of $15 to $18 to today's $9 to $11.
As we have seen the first convergence of an AVI tag with a proximity card, we can easily see how this "single credential" can be further integrated with a smart card, providing personnel and vehicle access control, while offering the further opportunity to use the smart card to pay for transit, on-street parking or any other nonparking-related financial transaction. The idea of using a smart card to pay for tolls had been investigated in the early 1990s, but the cost of the tag in the vehicle (on-board unit or OBU) was in the range of $100 to $150 -- now that is an expensive tag. On the other hand, there have been ongoing discussions for several years now about the possibility of auto manufacturers applying a tag at the time of manufacture, so that every car delivered has a tag installed. One would simply need to "register" it so that it could be used in a given application. These tags would likely be of the SCM variety (e.g., non-battery, wafer-thin) due to the attractive price points.
And that brings us to the "ease of use vs. system complexity" issue. The traditional benefits of AVI are well-known: faster throughput, increased security (no need to roll down windows), reduced shrinkage, and a more pleasant experience for the motorist entering and exiting the facility. While more convenient for the motorist, it means the back-end system must do more. When a vehicle crosses the loops, do I activate the AVI reader first or the ticket-issuing device? I must make sure that I don't read the tag in the second vehicle in line and open the gate, letting the first vehicle through and likely creating passback issues. The proper sequence should be: Upon loop activation, enable the AVI reader/disable the ticket-issuing device. If no valid tag is read, disable the AVI reader/enable the ticket-issuing device. This additional logic must sometimes be added to the back-end system, if it doesn't already exist. This is a small technical issue compared to two other significant integration issues: clearinghouse connectivity and tag data format.
Until such time as there is national clearinghouse for AVI use -- this would include toll roads, parking, fast food, car washes -- parking revenue control manufacturers will need to write specific interfaces to allow them to process AVI payments. This is actually a two-step integration issue: The PARCS manufacturer must first create a separate device interface for the "regional" flavor of the AVI reader and then a separate back-end clearinghouse interface for the "regional" clearinghouse. This is admittedly unwieldy, and requires time and effort on the part of manufacturers. But it is analogous to credit cards: At some point, the functionality needed to be included, because customers were demanding it and it, in fact, provided two benefits -- reduced shrinkage and the idea of credit card in/credit card out. The railroads faced this similar situation and solved it elegantly: The American Association of Railroads independently operates a national clearinghouse for all railroads. A similar situation in our industry would allow the manufacturers to write one clearinghouse interface, significantly reducing development efforts and eliminating "regional" differences and the efforts required to support them. A few large national clearinghouses have looked at the idea, but nothing is imminent, unfortunately.
When these tags begin appearing in vehicles on the showroom floor, we should be prepared to read them in ASCII format. There are hardwired ASCII-to-Wiegand converters available today, but these simply add cost to an installation and induce another potential failure point in the overall system. Other systems provide for tag ID conversion, or "tag lists" directly at the reader level. However, these systems create security loopholes and permit potential financial fraud. By having a conversion list on the reader, it is easy to (re)program the reader, without having to notify or modify the host system.
An example: The reader reads Tag "1" and internally converts it to Tag "A," which is then sent to the host system. Let us further assume that the host recognizes Tag "A" as a monthly parker or "security." If passback control is not turned on, it becomes easy to program the reader to convert Tag "2" to Tag "A" -- and the host system will never know the difference, because all it sees is the reader sending Tag "A." Such systems indicate a lack of awareness of the requirements of the parking revenue control and security markets as well as a lack of development rigor. Accounting and security can now be compromised outside the host, even outside the booth, directly at the reader. To overcome this, the reader must pass directly through the ID of tag, with no modification -- and this is exactly how we should expect to operate in the future with the SCM tags.
If the vehicles come equipped with tags, the cost to a facility is reduced to the cost of the reader and installation, which in turn makes the cost proposition of AVI more attractive. As the Wal-Marts and Targets of the world start buying SCM tag readers, the price of the readers will start to fall also, further lowering the cost of AVI in a given facility.
National clearinghouses, Department of Defense announcements, an AVI RFID tag in every car, toll road frequency changes -- all are things that apparently may not have a direct connection or impact on parking today, but will surely be significant factors in how we operate in the future. They also are factors that are already driving subtle changes in how we operate today. It will behoove us as an industry to understand these upcoming changes and to be prepared, accept, integrate and accommodate them.

Michael Bigbee has been involved in the deployment of AVI RFID systems in transportation markets worldwide since 1995. He may be reached at mbigbee@hotmail.com.