Abstract—Accurate effort estimation is the state of art of the Software Engineering activities and of course it is a complex process. On the other hand, it is also widely accepted that due to the inherent uncertainty in software development requirements and activities, it is unrealistic to expect very accurate effort estimates over the software development processes. Among the diversified effort estimation models, empirical estimation models are found to be possibly accurate compared to other estimation schemes. The work reported in this paper aims at improving the accuracy of one of the popular effort estimation models, COCOMO II. Since the accuracy of the COCOMO II stands over the cost derives and scale factor, this work investigates the influence of the cost drives and scale factor to improve the accuracy of the effort estimation. It is proved that, with a set of possible modifications in scale factor and cost drives, the overall accuracy of the COCOMO II can be improved. The improvement has been proved in terms of the performance validation factors such as Magnitude of Relative Error (MRE), Mean Magnitude of Relative Error (MMRE), Root Mean Square (RMS) and Relative Root Mean Square (RMS & RRMS). Index Terms—Effort Estimation, COCOMO II, graphical estimation, SLIM, SEER. I. INTRODUCTION The most crucial part in the software development activity is the effort estimation. For the past few decades many researches has been carried out to predict the actual effort to develop a software project. But, still it is night mare to achieve the closer result, because it involves both measurable and non measurable factors in other words we can say that it involves functional and non functional aspects of the software development process. To allocate the recourses in terms of man and machine the effort estimation plays a vital role and also in scheduling the task. Moreover, many researchers are dedicating their precious research time and money to work on various software effort estimation models to perk up the accuracy of those software effort estimation models. Although a great amount of research time, and money have been devoted to improving accuracy of the various estimation [1]. Though there is no proof on software cost estimation models to perform consistently accurate within 25% of the actual cost and 75% of the time [2], still the available cost estimation models extending their support for intended activities to the possible extents. The accuracy of all the models is depends on the software Manuscript received December 16, 2013; revised April 8, 2014. Ziyad T. Abdulmehdi, M. S. Saleem Basha, Mohamed Jameel, and P. Dhavachelvan are with the Department of Computer Science and Information Technology, Mazoon University College, Muscat, Oman (e-mail: Ziyad@mazooncollege.edu.om, saleem@mazooncollege.edu.om, mohammed.jameel@mazooncollege.edu.om). data of that particular project and the way how they calibrate those factors / values and of course the accuracy is an important factor that decides the applicability of the individual models in the appropriate environments. For the corporate, it is vital to maintain the precision and reliability of the effort estimation to grab the attention of the customers and also among the competitive companies. There are many estimation models have been proposed and can be categorized based on their basic formulation schemes; estimation by expert [3], analogy based estimation schemes [4], algorithmic methods including empirical methods [5], rule induction methods [6], artificial neural network based approaches [7]-[9], Bayesian network approaches [10], decision tree based methods [11] and fuzzy logic based estimation schemes [12], [13]. There are many diversified estimation models are there. But in particular the empirical models are believed to be an accurate estimation models when compare to such other diversified models. To name some of the popular empirical estimation models from the literature are COCOMO, SLIM, SEER-SEM and FP analysis schemes are popular in practice [14], [15]. During the past decades of empirical estimation models, the estimation factors are gathered from pragmatic values obtained from several similar projects and derived an obsolete values for the parameters to find near values of the estimation. But, now a days by the use of enormous techniques and tools namely neural method, bio inspired methods, Genetic algorithmic methods, etc., the parameter's values are finetuned and comes with different names of the estimation models. Accurate effort and cost estimation of software applications continues to be a critical issue for software project managers [16]. Due to above said scenario of using various tools and techniques, it is noticed that there are more changes, updates and versions of the same model. A common modification among most of the models is to increase the number of input parameters and to assign appropriate values to them. Despite the fact that few estimation schemes are flooded with more and more input parameters to achieve additional features of that scheme to create the credibility among the customers and competitors, in this way unknowingly they are injecting complexities into their estimation models. But fails to achieve the accuracy of their estimation schemes. Although they are diversified, they are not generalized well for all types of environments [17]. Hence there is no silver bullet estimation scheme for different environments and the available models are environment specific. Since the research focus of this paper is to refine the COCOMO II estimation scheme and to provide an improved estimation scheme, our discussion is limited with COCOMO II estimation models only. COCOMO II A Variant of COCOMO II for Improved Software Effort Estimation Ziyad T. Abdulmehdi, M. S. Saleem Basha, Mohamed Jameel, and P. Dhavachelvan 346 International Journal of Computer and Electrical Engineering, Vol. 6, No. 4, August 2014 DOI: 10.7763/IJCEE.2014.V6.851