Monday, January 25, 2010

COMPLEXITY 8

This is the eighth and last post in a series on complexity in scientific research. The theme of this collection is that scientific progress, particularly in the realm of healthcare, has declined as a consequence of the high complexity in software and other technologies.

-- POST BEGINS HERE --

The U.S. military enjoys working on huge, complex projects, and the scientists involved in these projects will go to extremes to keep failed efforts alive. A fine example of a long-running military research effort is the V-22 Osprey, affectionately renamed "The Grand Ole Osprey." In the history of engineering, there have been many attempts at dual-purposed devices: automobiles that can sprout wings and fly, boats that come ashore and covert to automobiles, washing machines that also dry clothes, houses on wheels that can be towed, behind a car. All of these devices exist, but they have not replaced their single-purposed components. It's very difficult to engineer a reliable and inexpensive composite device when each component is complex.

Circa 1980, the Pentagon decided it needed a hybrid aircraft that could take-off and land like a helicopter, but fly like a plane. Thus began the long, expensive and disappointing sago of the V-22 Osprey. After more than a quarter century in the making, and $16 billion dollars spent, the U.S. government has not created a safe and dependable vertical take-off airplane. Through the years, multiple crashes of the Osprey have resulted in 30 deaths.

In January, 2001, the New York times reported that a Marine Lieutenant-Colonel had been fired for falsifying Osprey records and for ordering the members of his squadron to do the same (1). "We need to lie or manipulate the data, or however you wanna call it," he said (1). The lies were intended to win new funding, but a squadron member caught the orders on tape, and the plan backfired.

Despite these problems, funding continued (2). The Osprey became operational in 2006 and is currently used in a wide variety of operations for the military.

If you try hard enough and long enough, it is possible to create functional complex products (hospital information systems, manned expeditions to mars, supersonic transports). The purpose of this series of posts is to show that complexity has very high costs. Aside from the money, the highest cost of a complex item comes from our inability to fully understand what we have created. We cannot always predict how complex objects will operate, how they will fail, or how they can be fixed. In many cases, it is better to acknowledge our limitations, by building very simple systems, and by developing ways of simplifying complex systems.

[1] Ricks TE. Data Faking Could Lower Osprey's Prospects Further. Washington Post Jan 21, 2001.

[2] Berler R. Saving the Pentagon's Killer Chopper-Plane. 22 years. $16 billion. 30 deaths. The V-22 Osprey has been an R&D nightmare. But now the dream of a tilt-rotor troop transport could finally come true. Wired 13.07, July 2005.

© 2010 Jules Berman

key words:informatics, complexity, jules j berman, medical history
My book, Principles of Big Data: Preparing, Sharing, and Analyzing Complex Information was published in 2013 by Morgan Kaufmann.



I urge you to explore my book. Google books has prepared a generous preview of the book contents.