中文摘要：

在整数解的搜索过程中,对传统的分枝定界算法进行了改进,给出了如何沿最快下降的方向寻找下一分枝的算法,从而减少了分枝定界算法的分枝数,得到了一个可以用于解算混合整数平差模型的最小二乘估计方法。数值试验结果表明,所给算法大大改进了传统的分枝定界算法,并有广泛的适用性。

英文摘要：

This paper carried out a series of improvements on the traditional branch and bound algorithm.During the searching process of integer solutions,paper researched how to seek the next branch along the fastest descent direction.So,the branch numbers are decreased,and a branch and bound algorithm for solving integer least squares estimation of adjustment model is gave.The numerical results shows that the new algorithm is not only a great improvement on the traditional branch and bound algorithm,but suitable for different complex problems.