An integrated approach for estimating software cost estimation using Adaptive Neuro-Fuzzy Inference System and the Grey Wolf Optimization algorithm

Estimating the cost of software is a crucial aspect of project planning and implementation in software development. More accurate cost estimation can significantly impact the success of projects by ensuring efficient resource allocation and informed decision making. This study presents a novel approach that integrates the Adaptive Neuro-Fuzzy Inference System (ANFIS) with the Grey Wolf optimization (GWO) algorithm for estimating software effort. ANFIS excels at capturing complex and nonlinear relationships in the data, while GWO offers optimization capabilities that make it a promising combination for accurate cost estimation. This study presents the two-stage ANFIS-GWO model, which is evaluated using predefined criteria, including mean magnitude of relative error (MMRE) and prediction accuracy (PRED). The results obtained with other algorithms such as PSO, DE and GA show that ANFIS-GWO performs better than other models in terms of cost estimation accuracy.