Abstract: A fast and efficient heuristic packing algorithm is presented for large-scale orthogonal packing problem.For the present packing position（horizontal line），this paper uses greedy method to choose the combination of unpacked rectangles that satisfy packing condition.According to the matching degree of packing position and its corresponding combination of rectangles，it chooses optimal packing position（optimal horizontal line） for packing.For the convenient of later packing，it ranks the rectangles combination of packing position with height before packing.Three largest-scale test instances containing 196 or 197 pieces supplied by E.Hopper are computed by the algorithm of this paper，the utilization ratios of all are beyond 99%，the average utilization ratio and consumed time are 99.38% and 1.12 seconds respectively.Compared with the best results that have been published in related papers，it is showed the efficiency of algorithm of this paper for solving large scale rectangle packing problem.