Abstract

Innovations in last-mile delivery, such as shared parcel lockers (PLs), promise more sustainable urban logistics, but realised benefits depend not only on consumer preferences but also on infrastructure and operating rules. This study develops an agent-based model that integrates the socio-cognitive HUMAT framework with the MASS-GT freight simulator to test how social influence interacts with parcel-locker access, capacity, and emptying frequency in shaping realised locker use. Using empirically informed synthetic demand and infrastructure data for Thessaloniki, Greece, we simulate scenarios that vary customer preferences, access regime, capacity, and collection frequency. Results show that infrastructure and operational design are the primary determinants of realised locker uptake: collaborative (public) lockers achieve up to 90% utilisation in most serviced zones when emptied daily, whereas exclusive (private) lockers remain under-utilised; doubling locker capacity nearly eliminates unmet demand; and socially mediated preference change produces only modest variation in realised use once operational constraints are taken into account. These findings suggest that social influence may affect adoption intentions, but interoperability, capacity, and turnover are more decisive for sustained parcel-locker performance.

Önder Gürcan

Senior Researcher

My research interests include multi-agent systems, collective intelligence, self-organization and self-adaptation, simulation of biological systems, distributed clock synchronization and behavioural economics.