Parking patterns of last-mile delivery vehicles: a comparison between light-duty trucks and electric cargo tricycles in downtown Toronto

• GPS–shipment data fusion enables robust identification and validation of delivery-related parking events. • E-cargo tricycles exhibit significantly shorter parking durations than trucks despite higher stopping frequencies. • Parking duration increases with delivery intensity, including parcel volume, number of receivers, and walking distance. • E-cargo tricycles park closer to delivery destinations than trucks, reducing pedestrian travel effort. Electric cargo (e-cargo) tricycles provide an alternative mode for freight deliveries in dense urban areas. This study develops a data fusion framework to analyze and compare the parking patterns of delivery trucks and e-cargo tricycles conducting courier deliveries in downtown Toronto. Global Positioning System (GPS) data are processed to identify trip ends and stop locations. Shipment records are fused with the GPS trajectories to distinguish stops associated with delivery activities. The resulting dataset is used to develop parking duration models for trucks and e-cargo tricycles and quantify the influence of trip, shipment, and parking location attributes on the parking duration. Results show that e-cargo tricycles park for shorter periods than delivery trucks yet stop more frequently per tour. Parking durations increase with the number of receivers served, the number and weight of parcels handles, morning deliveries, high stop-frequency tours, and greater walking distances between parking locations and delivery destinations.