The definition of lightning attachment zone in the standard regulatory documents have been inferred from empirical approach. In the framework of the European FULMEN program, the actual physical processes involved in a lightning strike to aircraft or helicopter are studied to improve the determination of the initial lightning attachment.In flight experiments performed in the 80’s have shown that there are two main processes which lead to a lightning strike to aircraft. The first process occurs when an aircraft flies into a region of the thunderstorm where the electric field is intense. The result is the triggering from the aircraft of preliminary discharges which lead to lightning development. The second process is the interception by the aircraft of a natural lightning channel developing in the vicinity of the aircraft.The experiment, performed at the test center of CEAT in November 1997, has been designed to simulate the initial phases of both processes. The « triggered » configuration is simulated by a plane-plane gap into which the mock-up is placed. The location of the discharge initiation on the mockup is investigated as a function of the angular position of the mockup in the gap. For the « intercepted » configuration, the natural discharge is produced in the vicinity of the mockup by a sharp rod. The location of the attachment point on the mockup is studied as a function of the distance between the mockup and the HV electrode.The chronological events occurring during the development of the discharge are investigated through the current and electric field measurements on the mock-up.The numerical simulations consist in computing the characteristics of the positive and negative corona developing from the mock-up. The comparison of the corona charge produced on different zones of the aircraft enables to determine the points where the probability to be struck by the stroke is the highest.