算法设计与分析中的任务分配问题。

用回溯法解该问题时，用完全n叉树表示解空间。可行性条件select 剪去不满足行、列约束的子树。递归方法

backtrack(1)实现对整个解空间的回溯搜索；backtrack(i)搜索解空间中的第i层子树，sum记录当前已找到的可行方案

数。在算法backtrack中，当i>n时，算法搜索至叶子结点，当前已找到的可行方案数sum加1；当i<=n时，当前扩展结点

Z是解空间中的内部结点。该结点有x[i]=1,2,...,n共n个儿子结点。对当前扩展结点Z的每一个儿子结点，由select方法

检查其可行性，并以深度优先的方式递归地对可行子树搜索，或剪去不可行子树。-Algorithm design and analysis of task allocation problem. When the solution of the problem using backtracking, with a completely n-tree represents the solution space. Feasibility cut does not meet the conditions select rows, columns, constraints subtree. Recursive method backtrack (1) to achieve the entire solution space backtracking search　backtrack (i) the search of the solution space of the i-layer sub-tree, sum the number of records feasible solutions currently found. In the algorithm backtrack, when i>　n, the algorithm searches to leaf node, the current program has been found feasible to add a number of sum　When i < = n, the current expansion node Z is an internal node in the solution space. The node has x [i] = 1,2, ..., n of n son nodes. For each son node Z' s current expansion node, check its feasibility by the select method, and depth-first manner subtree recursively search for feasible or infeasible cut subtree.