The reliability of a software depends on the quality. So, the software growth models require efficient quality assessment procedure. It can be estimated by various parameters. The current paper proposes a novel approach for assessment of quality based on the Generalized Weighted Laplacian (GWL) method. The proposed method evaluates various parameters for detection and removal time. The Mean Value Function (MVF) is then calculated and the quality of the software is estimated, based on the detection of failures. The proposed method is evaluated on process CMMI level 5 project data and the experimental results shows the efficiency of the proposed method.