0-l背包问题是子集选取问题。一般情况下，0-1背包问题是NP难题。0-1背包 问题的解空间可用子集树表示。解0-1背包问题的回溯法与装载问题的回溯法十分类 似。在搜索解空间树时，只要其左儿子结点是一个可行结点，搜索就进入其左子树。当 右子树有可能包含最优解时才进入右子树搜索。否则将右子树剪去。设r是当前剩余 物品价值总和；cp是当前价值；bestp是当前最优价值。当cp+r≤bestp时，可剪去右 子树。计算右子树中解的上界的更好方法是将剩余物品依其单位重量价值排序，然后 依次装入物品，直至装不下时，再装入该物品的一部分而装满背包。由此得到的价值是 右子树中解的上界。-0-l knapsack problem is selected subset of the problem. Under normal circumstances, 0-1 knapsack problem is NP-hard. 0-1 knapsack problem the solution space available subset of the tree said. Xie 0-1 knapsack problem with the law retroactively loading the retroactive law very similar. The search solution space trees, as long as their son left node is a viable nodes, the search entered its left subtree. When the right subtree is the optimal solution may contain only son into the right tree search. It will cut right subtrees. Suppose that r is the total value of the remaining items; Cp is the current value; Bestp is currently the best value. When cp r bestp, they can cut right subtrees. Calculation right-tree solutions to the sector is a better way to the remaining units according to the