The optimal design of the foundation pit row-pile algorithm could effectively solve the existing probability of foundation pit accident. The traditional foundation pit row-pile algorithm has strong subjectivity and lacks constraint conditions. Thus, an optimal excavation row-pile algorithm based on grey fuzzy model was designed in this paper. The row-pile structure of two pile was taken by the model as an example. Firstly, the disadvantage of the traditional equivalent beam method was described. The grey fuzzy theory was used to construct the rowpile data evaluation assemblage to be optimized, and then the environmental information and architectural constraint condition were added to the assemblage. Through querying change data between external factors and event ontology which is difficult to describe, the formula was used to describe the modeling membership of grey fuzzy theory in the assemblage to be evaluated and build membership function of model. Moreover, the least square method was used to give the objective function of foundation pit row-pile algorithm, so as to realize the design of foundation pit row-pile algorithm. Through the simulation results, the designed algorithm can effectively solve the shortcomings of traditional methods. The optimization effect is obvious, and the practicability is relatively high.