A comprehensive understanding of shippers’ preferences can help transport freight forwarders create targeted transport services and enhance long-term business relationships. This research proposes an integrated approach to learn shippers’ preferences in synchromodal transport operations and optimize transport services accordingly. A preference learning method was developed to capture shippers’ preferences through pairwise comparisons of transport plans. To model the underlying complex nonlinear relationships and detect heterogeneity in preferences, artificial neural networks (NNs) were employed to approximate shippers’ utility for a specific plan. Leveraging the learned preferences, a synchromodal transport planning model with shippers’ preferences (STPM-SP) was proposed, with the objectives of minimizing the total transportation cost and maximizing shippers’ satisfaction. A case study based on the European Rhine-Alpine corridor was conducted to demonstrate the feasibility and effectiveness of the proposed approach. The results demonstrated that artificial NNs have the capacity to identify complex (i.e., nonlinear and heterogeneous) relationships in shippers’ preferences. The planning results showed that the STPM-SP effectively found solutions with a significant satisfaction improvement of 37%. This research contributes to learning shippers’ preferences in the transport operation process and highlights the importance of incorporating these preferences into the decision-making process of synchromodal transport planning.

With the electrification in freight transportation, the availability of fast-charging facilities becomes essential to facilitate en-route charging for freight electric vehicles. Most studies focus on planning charging facilities based on mathematical modeling and hypothetical scenarios. This study aims to develop a data-driven integrated framework for fast-charging facility planning. By leveraging the highway traffic data, we extracted, analyzed, and compared spatial and temporal flow patterns of general traffic and freight traffic. Furthermore, graph theory-based network evaluation methods are employed to identify traffic nodes within the highway network that play a significant role in accommodating charging infrastructure. A candidate selection method is proposed to obtain potential deployment locations for charging stations and to-go chargers. Based on this, we present a multi-period bi-objective optimization model to provide optimal solutions for the placement of charging facilities, with the objectives of minimizing investment cost and maximizing demand coverage. The case study on the Amsterdam highway network shows how existing traffic data can be used to generate more realistic charging demand scenarios and how it can be integrated and evaluated within the optimization framework for facility planning. The study also shows that the proposed model can leverage the potential of early investment in improving the charging demand coverage.

Bike-sharing systems have rapidly expanded around the world. Previous studies found that docked and dockless bike-sharing systems are different in terms of user demand and travel characteristics. However, their usage regularity and its determinants have not been fully understood. This research aims to fill this gap by exploring smart card data of a docked bike-sharing scheme and GPS trajectory data of a dockless bike-sharing scheme in Nanjing, China, over the same period. Both docked and dockless bike-sharing users can be classified into regular users and occasional users according to their usage frequency. Two systems are cross-compared regarding their travel characteristics. Then, binary logistic models are applied to reveal the impacts of travel characteristics and built environment factors on the regularity of bike-sharing usage. Results show that for both bike-sharing systems, regular users and occasional users share similar riding time and distance, while significant differences in the spatio-temporal distribution between docked and dockless bike-sharing systems are observed. The regression model results show that the “Trips during morning and afternoon peak hours” are positively associated with the regularity of both docked and dockless bike-sharing usage. However, the “Riding distance” variable is negatively associated with the usage regularity of both systems. Built environment factors including working point of interest (POI), residential POI, and transit POI promote the usage regularity of both bike-sharing systems. Finally, policy implications are proposed, such as increasing the density of docking stations in suburban areas and developing high-quality parking area for dockless bike-sharing around public transport stations. This study can help operators or governments to launch or improve the service of bike-sharing systems.