A soft sensors is a computer program for the inference process variables, which are very hard used to measure them. The objective of this paper is to design a modified soft sensor based on a Takagi-Sugeno-Kang (TSK) fuzzy model. In this paper a soft sensor is proposed based on a novel modification of the nearest neighbor distance-based clustering optimized by the Particle Swarm Optimization (PSO) algorithm. To evaluate the performance of proposed soft sensor, it is applied in an industrial process. The results affirm that the designed soft sensor based on the proposed method is more precise even for contaminated dataset by white Gaussian noise.