The increased pressure to reduce aircraft emissions and the growing demand for air travel causes airports to consider alternatives for fuel-inefficient aircraft taxiing. External towing tugs can be deployed to effectively reduce on-ground fuel burn of outbound aircraft. These new operations are likely to aggravate ground delays as a result of flow-disturbing uncoupling processes on taxiways or holding platforms. The size of these delays has not been previously investigated. The goal of this research is to analyse the impact of various uncoupling scenarios on ground delays. Scenarios are created by varying the uncoupling duration and the available uncoupling points. A computational model is created with Amsterdam Airport Schiphol (AAS) and Taxibot as reference airport and towing system. Vehicle-specific operations are optimised by a hierarchical multi-agent path planning algorithm. The algorithm comprises augmented versions of a Priority Based Search (PBS) algorithm and a Safe Interval Path Planning (SIPP) technique. Simulated conventional operations are compared with Taxibot-enabled operations for traffic samples corresponding to high traffic demands. Minor delays are found for inbound traffic. For outbound traffic, most delays occur during towing and are mainly determined by the arrangement of uncoupling points, the uncoupling duration, and the traffic demand. Delays are furthermore, but to a lesser extent, determined by the changed kinematics of aircraft towing and the runway’s position. Delays resulting from uncoupling only on taxiways or holding platforms vary per runway as a result of different arrangements in uncoupling points. For every arrangement of uncoupling points, the number of points and the presence of alternative paths around the points are key in mitigating delays resulting from long uncoupling processes. The intensity of delays can be reduced by tuning search objective settings in the algorithm, in particular for uncoupling points arranged in series.