Study on the safe pre-warning model of construction project based on the rough sets with HGA and the ANN

The safe Pre-warning of construction project is a hot studying spot in current theoretical field. Whether the safety condition of a construction project can be accurately pre-warning or not, is key to the success of the project. This paper firstly overviews the present research condition of construction project safe pre-warning, sets up the pre-warning indicator system. Then accomplishes dynamic pre-warning for the construction project safety based on the great nonlinear function approaching ability of artificial nerve network (ANN). But it is known that the main shortages of ANN are the converging speed is often slow and the network is easily involved in local optimum, so this paper introduces the the rough sets theory (RS), to simplify the space dimension of input information and reduce the complexity of network structure. Also to escape the NP-hard problem of RS, the Hybrid Genetic Algorithm (HGA) is brought in. Thus establishing a safe pre-warning system to construction project based on the RS with HGA and the ANN, making the RS reduction, the HGA optimization, the dynamic study and induction of ANN, the pre-warning and evaluation of construction project safe organically combined. At last, an actual example is given to justify the validity and efficiency of this model. The study result can supplies a new mean for the dynamic pre-warning of construction project safety.