Is Paper Safer?

Air traffic control is a complex, safety-critical activity, with well-established and successful work practices. Yet many attempts to automate the existing system have failed because controllers remain attached to a key work artifact: the paper flight strip. This article describes a four-month intensive study of a team of Paris en route controllers in order to understand their use of paper flight strips. The article also describes a comparison study of eight different control rooms in France and the Netherlands. Our observations have convinced us that we do not know enough to simply get rid of paper strips, nor can we easily replace the physical interaction between controllers and paper strips. These observations highlight the benefits of strips, including qualities difficult to quantify and replicate in new computer systems. Current thinking offers two basic alternatives: maintaining the existing strips without computer support and bearing the financial cost of limiting the air traffic, or replacing the strips with automated versions, which offer potential benefits in terms of increased efficiency through automation, but unknown risks through radical change of work practices. We conclude with a suggestion for a third alternative: to maintain the physical strips, but turn them into the interface to the computer. This would allow controllers to build directly upon their existing, safe work practices with paper strips, while offering them a gradual path for incorporating new computer-based functions. Augmented paper flight strips allow us to take advantage of uniquely human skills in the physical world, and leave the user interface and its subsequent evolution in the hands of the people most responsible: the air traffic controllers themselves.