This feasibility effort will demonstrate neural network-based (NN) automated route planning for aircraft, adaptable to the user's flight preferences and mission needs. This system, termed pilot adaptable optimal planning (POP), will be designed to perate in two modes. In the first mode it will provide the means for route planning and estimating the success of the user's route plan by flying and training the simulated missions. The user will pretrain missions for general requirements with flight patterns corresponding to the user's individual preferences. These missions will be collected in mission specific libraries. In the second mode, the user will select an appropriate mission form the library and optimize it to the desired mission using the lates intelligence, maps, weather, etc. This will achieve two goals: (1) introduce pilot flight preferences into reroute planning and (2) provide optimization of route plans. The plan can be evaluated in monte carlo simulation, with scoring indicating the probability of a successful mission. In our vision, the proposed pop system would eventually be implemented on an a mission planning workstation like MASSE II and MASSE III where it may enhance current route optimization process.