and to indicate those areas that need improvement and those that function well. The success of this process does, however, require a symbiotic relationship between the human factors investigator and the designers and evaluators of systems. The problems inherent in this approach have been pointed out elsewhere; for example, Baker and Marshall ( 1988) made the point that "However desirable a co-operation between designers and human factors experts might be, human factors specialists are still not sufficiently involved in the design phase with the result that, often, the anticipated benefits from the system in question are not spelled out in any clearly testable way. Typically, psychologists are simply requested to validate or demonstrate the advantages of a new system" (p. 83). In the United Kingdom at least, very few ATC-related incidents can be traced directly to problems related to inadequate or less than optimal equipment. Rather, poor equipment and facilities tend to be implicated as contributory, not causative factors. Nevertheless, investigations do reveal areas in which equipment development is needed. A good deal of attention has been focused, for example, on alerting systems that inform the pilot or controller of an impending collision. This is well and good and very necessary if airborne collisions are to be avoided. However, relatively less attention has been focused on the development of systems that aid the planning and decision making aspects of the ATC tasks, that is, to prevent the development of situations in which conflicts arise in the first place. The investigation of the human factors aspects of incidents and accidents can be helpful here in highlighting those aspects of the planning and decision making process most in need of support.
Feedback is not, however, restricted to ergonomic and equipment-related issues. The adoption of the system approach discussed earlier facilitates the gathering of information on all aspects of ATC functioning. Human factors recommendations that ensue from investigations can range from fairly basic "quick fixes" to more far reaching issues involving, for example, such aspects as training or the role of management. In Reason's terms ( Reason, 1989), both the "active" and "latent" failures in the system need to be addressed, and careful sifting and analysis of the information gathered from investigations can reveal, not only those areas in which failures have already occurred and errors been made, but also those aspects of the system that if left unaddressed could well lead to problems in the future. The existence of these generic problems, which may not have manifested themselves directly in an incident or whose connection to an occurrence may seem somewhat tenuous, is often difficult to demonstrate. This is one area where the advantages of incident as well as accident investigation are most evident. It may be difficult to demonstrate, say, on the basis of one accident, that a particular problem exists. However, if it can be shown that similar conclusions have been reached as a result of the more numerous incident investigations, the case for a closer examination of the problem and perhaps the initiation of research will be greatly strengthened.
The role of human factors in incident and accident investigation has received increased attention in recent years. Even so, the extent to which human factors considerations are taken into account during the investigation process varies from state to state. This chapter has focused on the investigation of civil ATC-related incidents and accidents in the United Kingdom, where a human factors specialist is routinely included as part of a multidisciplinary investigation team.