Fuel rationing strategy in farmland based on intelligent management platform for agricultural machinery

Agricultural machinery has been widely used in large-scale operation and wide distribution in recent years. However, the returning duration of refueling and queuing has seriously limited the operational efficiency of agricultural machinery, particularly in tens of acres or even hundreds of acres of land. Apparently, the return refueling cannot fully meet the requirements for the continuous operation of agricultural machinery in the context of large-scale production. It is very necessary to guarantee the fuel rationing and supply of agricultural machinery with less time and high yield in the field. Fuel rationing in the field for agricultural machinery refers to the fuel supply point distributing the required amount of fuel to the farmland, where the agricultural machinery is located within the required time and completing fuel refueling. Fuel rationing in actual agricultural production includes fuel delivery into the field via filling buckets, and mobile refueling service by mobile refueling trucks. The rationing relies mainly on unsafe, untimely, and manual decision-making with a low level of digitization. It is still lacking in the systematic decision-making on rationing, in order to improve the efficiency of fuel rationing and distribution cost-saving. In this study, a fuel rationing in-field strategy was proposed using an intelligent management platform. The fuel demand of agricultural machinery was combined to utilize the agricultural Internet of Things (IoT), taking the synergistic advancement of "demand analysis-order allocation-route planning" as the core. Firstly, a two-stage mathematical model of fuel rationing in agricultural machinery was established using real-time data and the theory of trustworthiness in the intelligent management platform. In the first stage of pre-optimization, a multi-objective vehicle route planning model was established with the fuzzy demand and fuzzy time window using the platform data. In the second stage, the distribution route planning was re-routed to update the relevant information for undelivered orders using the dynamic events in agricultural production. Secondly, the solution was designed after the two stages. Taking the lowest total distribution cost and the highest total satisfaction as the objectives, a multi-objective cooperative co-evolutionary genetic method was applied to realize the distribution route planning using an order analysis. The dynamic optimization was realized to obtain the dynamically optimal scheme of fuel rationing. Finally, an empirical study was conducted with the Nenjiang Farm. An intelligent management platform was selected to implement the scheduling of agricultural machinery. According to the Beidou navigation satellite system terminal carried by the agricultural machinery, the operation data was acquired in the field, such as the location, the amount of operation, and the status information. The production data of the farm was also adopted to conduct simulation experiments. A comparison was then made between the distribution cost and current indicators to verify the feasibility and practicality. The function of fuel rationing was realized in field planning. The results show that high satisfaction and distribution cost-saving were achieved in the fuel rationing in the field for the agricultural machinery, according to the real-time data from the intelligent platform. The final scheme of dynamic optimization was reduced to 14 distribution routes, compared with the current one. The total distribution cost was reduced by about 15 times. The satisfaction rate of 72.85% was slightly lower than the actual one. The relevant functions were promoted in the intelligent platform for agricultural machinery. The finding can provide a scientific basis for the intelligent management of fuel rationing in the field of agricultural machinery.