To ensure the fishing boat security in the coastal area of Fujian, it is imperative to develop efficient forecasting control system to manage the boats operation in a safe and orderly process. The core issue of establishing the forecasting control system is to predict the relationship of the real time traffic flow and the boat security condition. However, literature review shows that limited reports have addressed on this problem. Hence, a new intelligent forecasting control method base on the Chaos-Particle Swarm Optimization (PSO) and Fuzzy Neural Network (FNN) is proposed for the short time traffic flow prediction in this paper. The Empirical Mode Decomposition (EMD) was first used to denoise the original ship traffic flow observation and then the Chaos-PSO-FNN was applied to the forecasting of the ship traffic flow and hence established the relationship of the real time traffic flow and the boat security condition. The advantage of the proposed approach is that the Chaos-PSO is employed to optimize the FNN parameters to overcome the premature problem of the FNN. As a result, the forecasting control performance is enhanced greatly. In the experiment analysis, the fishing boat traffic information provided by the Fujian marine bureau has been used to evaluate the newly proposed method. The analysis results demonstrate that the proposed Chaos-PSO-FNN method can extract distinct features of the traffic data and the prediction rate is beyond 93.7%. In addition, it found that the fishing boats is supposed to be safe when their travel time avoids the rush time of the traffic flow. This result agrees well with the real data. Thus, the new intelligent forecasting control method for fishing boat security can be used in practice.