To reduce the environmental impact of aviation, new aircraft configurations are investigated. One of these researches is the PARSIFAL project, that investigates an efficient box-wing aircraft for passenger transport. In previous work, an engine sizing tool is developed to design the engines of the PARSIFAL aircraft. The tool is developed as part of the Multi-Model Generator (MMG), a knowledge based engineering application developed in the Flight Performance section. This work also included an engine integration study, which showed that a fuselage mounting is preferred over a wing mounting. However, the integration study lacked the desired design sensitivity for small design changes, e.g. engine diameter and length. Because of this lack of design sensitivity, the question whether the engine installation following from this work is ideal remains open. Therefore, in this thesis a best practice for engine integration studies during the preliminary design phase is investigated. The found best practice is then used to re-asses the engine installation by aligning the engines with the local flow. The best practice is also employed to asses under wing engine installations, to establish the potential use of the practice for this type of installations. Before the trade-off study for a best practice could be performed, changes to the MMG had to be made. Three primitives, the building blocks that create an aircraft geometry, had to be adapted. A new wing primitive, through flow nacelle (TFN) geometry, and a pylon primitive are introduced. In search of a best practice, textbook analyses, 3d panel solvers, and combinations of the two are compared. To fully assess an engine integration, information on lift, drag, and pitching moment is required. Since textbook methods only provide information on drag, these are not satisfactory on their own. Nonetheless, these methods can be used to predict the drag addition of a pylon, when only airframe and nacelle are simulated. In two panel solvers, FlightStream and VSAERO, two analysis practices are tested. The first being a simulation including airframe, nacelle, and pylon and the second being a simulation including airframe and nacelle complemented with a textbook correction for the pylon drag. Following a trade-off, using speed, accuracy, ease-of-use, and design sensitivity as criteria, a simulation with FlightStream of airframe and nacelle with a textbook addition of pylon drag proved to be the best analysis practice. This best practice is employed to re-asses the engine installation of the PARSIFAL aircraft. While the location is fixed, since it aligns with the structure inside the airframe, the orientation is changed to align with the local flow. This alignment results in a 3% decrease of aerodynamic efficiency loss and a 17% reduction of pitching moment increase, with respect to the original engine orientation. In a second study, the potential of the best practice to analyse under wing engine installations is tested by changing the location of an engine under the rear wing of the PARSIFAL box-wing. These results show that with proper placement, the same aerodynamic efficiency loss can be obtained as for fuselage installations.