【摘要】 在分析现有采购拍卖研究多为较少数量物品拍卖和投标规则过于简单的基础上,针对价格折扣问题,提出了更符合企业采购实际的模型:允许投标人分段报价,每一段中边际产品价格不变,但产品平均价格随着数量的增加而下降。由于模型的计算复杂性(NP-Hard),无法用经典方法求解最优解。通过在采购拍卖中引入智能算法思想,运用遗传算法进行求解,并同时利用启发式算法引导其搜索过程,提高了搜索效率。计算实例表明,混合遗传算法结果优于启发式算法。更多还原

【Abstract】 AAnalyzed that most of the researches on procurement auctions are focused on a small quantity of products with simple bidding rules. A more practical model containing a single buyer who wishes to procure large quantity of homogeneous products and many suppliers is presented. The bidding language of the suppliers is decreasing piecewise linear supply curves which indicate the price charged as a function of the purchased quantity. It is usually impossible to achieve the optimal solution with classic methods because of its high computational complexity (NP-Hard) .By adopting intelligent optimization algorithms, a Genetic Algorithm was presented on the base of the ideas of heuristics algorithms which can greatly increase the searching efficiency of GA. An example was calculated by the proposed computer program and it demonstrated that the results obtained by Genetic Algorithm was better than those achieved by heuristics algorithm.更多还原