Abstract: China is a large country of grain production and population so that the food problem has a significant impact on China's economic and social development. However, in the background of supply-side reform, the existing grain supply chain "wholesale layer-by-layer" circulation system can not meet consumers individual requirements. So, in this paper, based on the needs of individual customers, the application polling system theory is used to studying the mechanism of automatic control of order picking in flour, rice and other food materials.
Based on the sorting characteristics of the continuity material and the actual business needs, common order is distinguished from priority order. Besides, establishing two level priority polling order sorting system model of exhaustive parallel limited-1. The proposed model takes exhaustive control strategy to sort the priority order and takes limited-1 control strategy to sort the common order, which can ensure priority order customers receive better service and the fairness of order sorting systems. Using the embedded Markov chain, multidimensional probability generating function and Laplace transform to solve theoretical model. Then through numerical calculation and analysis to get system's first order characteristics and second order characteristics and other key performance parameters for intensive study of the control mechanism of order sorting.
The continuous material automated order picking polling control mechanism is studied and the priority order and general order which ensures priority orders get better customer service, and fair ordering system are distinguished. Finally, a numerical example verifies the polling order picking system stability and high efficiency. The results of this study expand the depth of the application polling theory and the breadth of the order sorting theory research, which is of great significance for the realization of the supply side of the grain industry structural reforms.

Key words: continuity material, order sorting, polling system, on-demand customization