The technological innovation efficiency of 13 listed oil field enterprises from 2005 to 2012 has been evaluated integrally and dynamically by using the DEA-Malmquist index analysis method. And an intelligent diagnosis has been conducted to analyze the results of the technological innovation efficiency by using the probabilistic neural network (PNN). The results show that the Malmquist productivity efficiency of the 13 listed oil field enterprises from 2005 to 2012 could not meet the efficient frontier boundary. And the recession performance of technological innovation efficiency is mainly caused by the low scale efficiency. Intelligent diagnosis could help decision makers to estimate technical invalid decision-making units and the efficiency model of the decision-making units quickly and efficiently, and put forward strategies for targeted improvement according to different efficiency models.