A Multi-Objective Cooperative Scheduling Framework for Harvesters and Grain Trucks Considering Carbon Emissions and Weather-Dependent Operational Efficiency

Efficient coordination between harvesters and grain trucks is essential for reducing operating cost, machinery waiting time, unnecessary travel, and carbon emissions in large-scale harvesting operations. This study develops a multi-objective cooperative scheduling framework for harvesters and grain trucks, with the objectives of minimizing total operating cost and total carbon emissions. The model integrates field operation sequencing, harvester assignment, harvester–truck service matching, machinery capacity constraints, waiting cost, and weather-dependent operational efficiency. To solve the resulting discrete and nonlinear scheduling problem, an artificial lemming algorithm with adaptive behavior selection and neighborhood cooperation (ALA-AN) is proposed. The algorithm uses a three-layer encoding structure to represent field sequences, harvester assignments, and grain truck service relationships and incorporates adaptive search, neighborhood cooperation, fast non-dominated sorting, and crowding-distance-based selection to improve Pareto-front quality. Computational experiments were conducted using real farm data and scheduling instances of different scales. The results show that the ALA-AN achieves better Pareto-front performance than the benchmark algorithms in terms of convergence, diversity, operating cost, and carbon emissions, with more evident advantages in medium- and large-scale scenarios. In the retrospective case-based comparison using Jianbian Farm records, the ALA-AN-based scheduling scheme reduced the estimated total operating cost by 5.85% and estimated total carbon emissions by 3.07% compared with the experience-based scheduling scheme. These results indicate that the proposed framework can support cost-effective and low-carbon decision making for cooperative agricultural machinery scheduling.