This research discussed the Multi-Vehicle Capacitated Vehicle Routing Problem (MCVRP) in the rice commodity supply chain. This study considered the impact of weather conditions and carbon emissions on route decisions. These factors influenced travel time and rice quality, which can lead to delays, route changes, and increased supply chain costs. To account for weather conditions, the proposed model integrated historical weather data into route decisions. Additionally, the model incorporated carbon emissions as a significant factor in route decisions, aiming to reduce the environmental impact of transportation. This was achieved by considering vehicle fuel consumption and corresponding carbon emissions, optimizing route decisions to minimize the overall carbon footprint. The objective of this research was to develop a routing model that minimizes total costs while adhering to vehicle capacity constraints and customer delivery demands. Adaptive Large Neighborhood Search (ALNS) was proposed as an optimization method to solve the problem. Particularly, novel destroy and repair operators of ALNS were developed to specifically reduce the transportation cost, emission cost, and lost sales cost due to weather-induced damages. The results indicated that the proposed ALNS significantly decreased delivery expenses compared to the initial solution, achieving a 32% reduction in costs. The ALNS algorithm yielded superior outcomes compared to the standard LNS with lower objective and faster computing time. This research contributed to the development of sustainable supply chain practices in the rice commodity industry. The proposed approach provided a solution for MCVRP that considered weather conditions and carbon emissions while ensuring efficient commodity transportation.