Parameter Estimation of COCOMO II using Tabu Search

-Estimating the cost and effort of software product is one of the biggest challenges. With the growth of technology and changing frequent user requirements it becomes very difficult to make accurate software cost estimation. But the accurate results are required for proper project planning because any error results in huge losses. The objective of software cost estimation method is to estimate the cost and effort required for software production. The cost estimation is usually dependent upon the size estimate of the project. There are several different techniques for performing software cost estimation among which COCOMO II is commonly used because of its transparency and simplicity. There are several soft computing techniques to optimize the effort estimated. Various optimization techniques such as Genetic algorithm, neural network, simulated annealing and many more were used earlier but in proposed model the author used the best technique called Tabu Search. The proposed model will tend to reduce the uncertainty of COCOMO II post architecture model coefficients i.e. a, b, c and d. Tabu search provides the optimized results in comparison with all other methods. Keywords--COCOMO II, Tabu Search, Scale factors, Effort multipliers, Size