Viewpoint: Timing is everything in transportation technology and innovation

This essay appeared originally in the IEEE Spectrum, January, 1999, pp 102-3, Technology 1999: Analysis and Forecast Issue. It is posted here with permission.

One of the most famous stories in electrical engineering lore goes as follows: while one of Thomas Edison's greatest inventions was the electric light bulb, his real genius lay in innovation, in persuading politicians, corporation heads, and financiers to create the electric power utility that would bring the light bulb into widespread use in the home and office.

Despite the growing conviction that technological innovation occurs rapidly (as is often the case with computer software and, to a lesser extent with computer hardware), this is not the case in many other areas. In the field of transportation, for example the existence of a new technology often precedes its application to transportation systems by as much as a decade or more. This delay is not due to an anti-technology bias in the transportation community but to the complexity of modern transportation systems.

At its most basic level, the transportation s is perceived largely as vehicles and infrastructure: automobiles and highways, airplanes and airports, railroad trains and track, ships and ports. At a higher level, the operation of vehicles and infrastructures becomes part of the equation, thereby adding human factors - still the single largest cause of transportation safety failures. Yet another layer involves the interaction of people, vehicles and infrastructure, an interaction leading to the formation of a transportation network.

This network, in turn, creates the need for communications, navigation, surveillance and decision-support systems to help people manage the flow of transportation in a safe, efficient manner. For engineers, the level of the transportation system that is most enigmatic is the institutional level. It is also the most frustrating. Yet due to the tremendous importance of transportation to national economies, as well as the obvious public safety considerations, institutions (and their concomitant bureaucracies) exist at all levels of government - national, state, and local. And, Of course, there are the commercial elements as well: airlines, railroads, shipping companies, vehicle manufacturers, electronics manufacturers, and so on.

The Global Positioning System (GPS) leads semantically to a limited concept of transportation as a technological system. From the standpoint of the effective introduction of new technologies to transportation, a more precise characterization comes from the emerging science of complexity, that is, viewing transportation as a complex, adaptive system.

In this context, a transportation system consists of a large number of intelligent agents, using technology to satisfy personal, economic, operational or institutional goals. It is, in effect, more like a microcosm of a national economy than a collection of integrated technologies. Therefore, the real applicability of a new technology is not always easy to discern: and worse, the negative side effects of a new technology may well outweigh local positive effects. In other words, a new technology must be accepted at all levels of transportation.

As technology advocates, our concept of transportation as a system such as the GPS model, which is being developed not only for navigational use in aviation, but also for sea and land transportation as well, raises some interesting questions.

For example, what are the implications of GPS for existing navigation system manufacturers or for the many transportation users who have bought in-vehicle navigation electronics? How fast can a nationwide navigation system be replaced and what is the cost of operating parallel navigation systems during the transition? How reliable, in the presence of interference (intentional and otherwise), is GPS for aircraft and ships where pinpoint accuracy is a necessity?

The complexity of transportation systems is not the only roadblock to the timely introduction of new technology. While as engineers we have a justifiable confidence in technology, as advocates we often forget the importance (and difficulty) of good engineering to ensure the effective, economic introduction of new technologies. Given people's natural reluctance to change, new technology needs advocates-champions, really--to overcome the added institutional inertia. But too often we oversell the benefits or oversimplify the engineering needed for implementation

This means that schedule and budget estimates tend to be too optimistic. When the anticipated benefits fail to materialize, a familiar sequence is set in motion: institutional and financial support begins to flag, and the initiative slows perceptively or stops altogether.

To avoid this pitfall, we should first illuminate the engineering risks so that false expectations are not raised. Then we should evolve the technology to reduce the engineering (and operational) risks. Successfully implementing a limited application of a new technology builds far more confidence than an unsuccessful demonstration, especially one that is visibly unsuccessful.

For example, the introduction of GPS as a navigation system for aviation needs to evolve from a supplementary system, to a primary system, and then, if and only if operationally and economically practical, to a sole means system. This takes time, as a friend of mine from Boeing Co. once said, "You can beta-test a new software program, but you can't beta-test a 747. Likewise, we must introduce in-vehicle intelligent transportation technologies before trying automatic traffic control technologies and eventually fully automated highways.

Most technology forecasts are too optimistic for the same reasons that transportation technology forecasts are too optimistic: the inherent complexity of the systems to which new technology is applied and the underestimation of the hard work of good engineering. All too often, this optimism is itself a deterrent to timely implementation.

To deal more effectively with the engineering issues we must think as engineers, not merely as technologists This means that we must do value engineering more than we do, and we must also design with the right audience in mind.

The GPS equipment in your new car will look considerably different than the GPS avionics in the cockpit of a DC-9. As engineers, we need to remember that what is second nature for a trained pilot may simply confound the average driver.

Unfortunately, academia often confuses invention and innovation (another semantic problem!) or, worse, value invention over innovation (despite Edison's example). Innovation is a tag-team race; every element must be successful to win.

The issue of system complexity suggests another semantic problem. Too often in engineering, the word "research" implies technological research. If innovation is to be stressed, we must put much more emphasis on system research, including the institutional and economic aspects of the system. Only by better understanding the dynamics of complex, adaptive systems will we be better equipped to predict the real benefits of new technologies and the implications they invariably pose for all concerned.

John J. Fearnsides (F) is senior vice president and general manager of the MITRE Corp, and director of its Center for Advanced Aviation System Development in McLean, Va., a Federally-funded R&D center sponsored by the U.S. Federal Aviation Administration (FAA). He is a former deputy under-secretary and chief scientist at the U.S. Department of Transportation.

HOME

Last modified: January 19, 1999