need report on the previous assignment
Greening Software Requirements Change Management Strategy Based on Nash Equilibrium Greening Software RequirementsChange Management Strategy Based on Nash Equilibrium Research Article Discussion Journal: Wireless Communications and Mobile Computing Introduction Proposes requirement change management approach Problem modelled as a game Players: Developer and Stakeholders Payoff matrix developed Requirement change proposed to be managed through Nash equilibrium-based algorithm to achieve greening them. Game theoretical methods have been successfully applied in various fields such as optimizing the spectrum allocation among different mobile users [19], and modeling the socially aware utility maximization problem, and deriving the distributed decision [20] 2 Other change management Models NBM Model: 4Ws ( Who, Where, Why, What) Process Model: Activities, artifacts, roles Requirement change classification: fixed, less likely, most likely Phased approach UML- based change management Agent-oriented approach for RCM process Game Theory Model for RCM Model Components Participators: Stakeholder, Developer Strategy: A1= stakeholder’s strategy space A2= Deveoepr’s strategy space Payoff Matrix The model works with two types of players in the game: stakeholders who propose or not propose change request and the other are developer who make local or overall adjustment. The options can be seen as one, when stakeholders choose to propose the change request and the other, stakeholders choose not to propose change request. 4 Game Theory Model Analysis Situation A:I + F < c="" ="" developer’s="" strategy-="" local="" adjustment="" situation="" b:="" k=""> f Stakeholder’s strategy- Propose Requirement change Situation C:I + F < c="" and="" k=""> f No pure Nash equilibrium In first situation, the overall adjustment cannot change the original plan for stakeholders. Thus RC is proposed and Nash Equilibrium point includes local adjustment and propose. In second situation, the satisfaction achieved is much more, thus local adjustment is chosen for I+F are less than C and overall adjustment otherwise. In situation C, stakeholders get maximum returns when developers make overall adjustments. When no more requests are made, local adjustments are made leading the problem to mixed strategy equilibrium. 5 Mixed Equilibrium Strategy: Situation C- I + F < c="" and="" k=""> f Profit expectations of Stakeholders: Profit expectations of developer: Mixed equilibrium achieved No more increase in profits possible The first two equations show the profit expectations of each player. However, no profit increment is possible once equilibrium is achieved as shown in third equation. 6 Strategy-Making for Overall Adjustment Conflict: development cost control & improvement in stakeholder’s experience Method: Stakeholders experience is quantified as satisfaction Mathematical model created for overall adjustment strategy Analysis of game mechanism Determining overall adjustment scheme using Nash equilibrium Mathematical model Assumptions: M and n are provided Satisfaction c : (0, 100, ] Higher t higher development cost D = {d ( c , t) | 0 < c="">< 100="" ,="" t=""> 0 } S = Evaluation Result Number of stakeholders (m) is provided. Number of distinct Requirements (n) for each stakeholder are provided. Satisfaction score c will achieve 100 once all requirements are met. Development cost is calculated through t. Higher t implies higher development cost. 8 Game Mechanism: Theoretical Algorithm Algorithm Design Flowchart of game theoretical algorithm Result Analysis: Superiority Parameter Selection Principles Avoidance of optimal solution violating objective facts. Guarantee is assured for accommodating stakeholder’s satisfaction variance in system evaluation. Superiority Analysis by Game theoretical Algorithm A = 1 B = -5 C = -1 The principles considered for selecting parameters are listed. The values taken in the algorithm are 1,-5 amd -1 respectively for A, B, C. The table depicts a comparison of two algorithms. 10 Result Analysis: Computational Computability Analysis Analysis of proposed model for large scale cases is shown in the figure. The run time increases linearly with the increase in the number of requirements. 11 Result Analysis: Case Study 2 Stakeholders: Emp1, Emp2 5 requirements 2 parameters Evaluation parameters A = 1, B = -5, C = -1 Two stakeholders with 5 requirements each are taken. Two parameters are taken A, B, C assuming values 1, -5 and -1 respectively. Satisfaction gain and development cost are specified in the brackets for each requirement in the table. 12 Conclusion More efficient method proposed Effective RCM methods are possible through efficient use of computing resources and recycling of software. In small cases, overall adjustment through proposed solution is similar to results of exhaustive algorithm Results show that the game theoretical algorithm is more efficient than exhaustive algorithm. 13