Data and Results

This repository contains the data of the Yangtze River and experimental results of the paper:

A branch-and-price approach for optimizing all-electric ships scheduling in waterborne transport

Data

Instance

• Benchmark_instances.xlsx: Instance settings used in Section 7. Numerical experiments.

Data for Computational Results (Section 7.2.)

Port26 contains the data for R5–R25, and Port30 contains the data for R30.

• port_distances.csv: Distance matrix of all ports (arc length of the inland waterway shipping network).

• port_segment_dis.csv: Sailing distances between all adjacent ports (arc length of the realistic inland waterway topology).

• port_seq.csv: Ports are categorized by type, where Type 1 indicates a port on the main waterway and Type 0 indicates a port on a tributary.

• port_traveltime.csv: Flow-dependent sailing time of each arc in the inland waterway shipping network.

Data for Sensitivity Analysis (Section 7.3.)

Network data used in Section 7.3.1. Impact of the network density.

Results

Results of Section 7. Numerical experiments are provided.

• Result_BS_mode.xlsx: results of the three battery swapping strategies baseline, heuristic, and BP used in Section 7.2.1.

• Result_Gurobi&BP.xlsx: benchmarking results comparing Gurobi and BP across instance scales R5–R30 as reported in Section 7.2.2.

• Result_BP_with_Gurobi_Incum.xlsx: results comparing BP initialized with Gurobi’s first incumbent versus BP initialized with the construction heuristic in Section 7.2.3.

• Result_Schedules.xlsx: scheduling solutions for representative instances across scales R5–R30, including ship routes, battery swapping locations, and route costs.

• Result_sens_density.xlsx: results of the network density sensitivity analysis presented in Section 7.3.1.

• Result_sens_horizon.xlsx: results of the planning horizon sensitivity analysis presented in Section 7.3.2.