Seismic data processing and interpretation is the core business of geophysical research, relating to project management, processing, interpretation, and computer departments. The project quality control system of seismic data processing and interpretation needs to establish a unified information management platform, with the rapid development of geophysical study. The idea of BPR(Business Process Reengineering) guides re-carding of seismic data processing and interpretation business chain, designing of deployable two-way information flow control model of management. And process quality control management achieves real-time, visual, traceable, fine and continuous optimization scientific management mode, then improves the quality and efficiency of geophysical project management. It lays a good foundation for better and faster to complete more oil & gas projects.

The genetic algorithm is a stochastic search method of global optimization, but also it has the disadvantages of low local search ability and precision. In order to solving the disadvantages this paper introduces pattern search, using the strong local search ability and high precision of pattern search to offset it. First of all, using generitc algorithm to compute a number of generations, and take all searched points as the first points of pattern search, then quicken convergent velocity. The analysis of theorized model and example computing validate the method of mixed optimum algorithm.