A literature Review on Application of System Dynamics in software project Management Pijush Chandra Das #1, Dr. UtpalRanjan Dhar #2 #1 Research Scholar, Department of Business Administration, Gauhati University,Guwahati- 781014, Assam, India #2 Professor UtpalRanjanDhar, Currently Dean, Royal School of Business, Ex-Dean & Head of the Department of Business Administration, Gauhati University, Guwahati-781014, Assam, India Abstract:The success of software development project is gaining importance increasingly. The challenges of cost variance, schedule, effort, rework, cost of poor quality is a very common phenomenon is most software projects. Managers consistently track projects in controlling the various software engineering matrices so that effective project management can be done to maintain higher margins. The highly complex, interdependent and dynamic behavior of the various related components of project management makes traditional approach or tools ineffective except in case of operational issues or problems within the process.the application of system dynamics play a pivotal role in giving more insights and understanding about the different managerial policies and a systematic management of the strategic issues of project management. Key Words: system dynamics, software project management, system dynamics model, software development. Corresponding Author: Pijush Chandra Das Introduction:The need for a reliable technique for controlling and improving the software development process is immensely felt so that projects can be delivered timely and within budget. Most software projects face schedule and cost overrun.to ensure the quality aspect of RSPUBLICATION, rspublicationhouse@gmail.com Page 51
software projects a set well defined software development process and metrics need to be defined. An understanding of the software metrics and its process can be analyzedwell, using system dynamics model and its applications in monitoring and controlling the project.the use of system dynamics gives an insight in to the various interacting factors of the software project life cycle which impact the cost management and time management of projects [24]. The experience gained by Project managers from previous projects and their historical databasecannot help completely in balancing these interacting forces. Hence System dynamics is very useful in studying the causal and dynamic behavior of software projects and organizations which finally helps in building a set of new of tools that can contribute in taking policy decisions[23]. Literature Review:Software Project management has been a great challenge using traditional tools because they could not provide a strategic insight into the aspects of human resource component and the cause and effect part of various project interactions. The traditional decision making framework needs to incorporate system Dynamics so that the inadequacies of traditional Project management tool can be compensated[1].developing a software project right from scratch is quite a complex and risky task. The failure in completing software development projects are also large. This huge failure can be attributed to lack of improper project management and other non technical issues. The risk management paradigm has been extended by scenario based software project management by abstract system dynamics models which possibly highlights the issues and provide strategic resolutions to problemsthat a software project may face[2]. To study the dynamical behavior software project management variables describing nonlinear relationships including feedback loops several system dynamics models were developed[3] [4] [5] [6]. However these models were not thoroughly evaluated using experimental analysis, except a few to find out whether these models were really helpful or not to project managers in decision making during the course of the project[2].in 2000 Drappa and Ludewig[7] conducted an experiment using a case study to determine whether software Project models enhanced the capability of Project management of students. The role of 90% syndrome is very much applicable to software development project also. The 90% syndrome is that the duration of a RSPUBLICATION, rspublicationhouse@gmail.com Page 52
project finally become twice as planned when a project reaches almost 90% completion on schedule[8]. Traditional tools like WBS, PERT, CPM etc fail to establish the interrelationships and the complexity of the project s components. The traditional tools works on the assumption that if each component of Project can be understood separately that one can control the whole project easily. System dynamics provide an alternative view by concentrating on the whole project by aligning the project management objectives with organizational strategic objectives. Traditional tools also have a key role to play in the operational support to Project management while system dynamics can make distinctivestrategic contribution which is complementary to the traditional techniques.the traditional approach assumes the availability of all information at the beginning of the project.in traditional techniques the complete project work is decomposed in to activities and based on historical experience and data a reasonable estimation of schedule, cost, effort and duration is achieved for the whole project. The major concern is even-though greater accuracy in estimation is obtained at individual activity or element level but important intra-project forces are ignored. Non-linear Project dynamics are not successfully incorporated in traditional tools whereas in System dynamics considers a holistic approach of the whole project.the subjective factors like motivation, quality, productivity and scope changes in project which impacts the whole the project can be explicitly represented using causal loop diagram in System dynamics.system dynamics is not a complete alternative but it is complementary to the traditional analysis techniques.in case of a special issue in a project System dynamics which is till date relatively a rare technique, used along-with separate studies[9]. In 2009 LI Lei,ZHANG Chen and LI Hong-min [10]studied the alignment of the organizational strategic objective and the operational project management. The System dynamic model they developed had four sectors where in the intra-forces between the sectors and the underlying influences within the sectors are studied. F.P Brooks in 1978 in his book mentioned that adding more resources to a late Project makes it later,is popularly known as Brooks Law [11]. The use system dynamics model in managing software development projects also helps understanding the maturity level of the organizations, the different managerial policies and technological change[4]. This was an important advancement in the field of Project management. This work was used as a base for the development of several other system dynamic models to RSPUBLICATION, rspublicationhouse@gmail.com Page 53
study the complex dynamical behavior of similar kind of Projects [12].In the initial phases of a project when minimal information is available about the project, for estimating and understanding the complex behavior of the project, a reduced dynamic model(rdm) was developed by Ruiz,Isabel Ramos and Miguel toro[12]. They used the work of Eberlein[13] to simply the Abdel-Hamid and Madnick s model[4] to reach to the approach and development of RDM.The RDM technique can be used for management policy experimentation and for estimations when there is dearth of historical database in the organization, when little information is available about the software development project and when a simple and user friendly model is required.in scenarios where traditional tools have become obsolete to explain the magnitude and complexity of software Projects the multiple simulation environments like vensim, Ithink, etc have made new inroads[12]. Software developments projects are under tremendous pressure of cost overruns, untimely delivery, end-user dissatisfaction. The highly interrelated and interdependent variables of software project management process constitutes a complex mix whose underlying cause and effect is very well explained by system dynamics. The schedule pressure is such that effort sometimes increases to as much as 100 percent [14]. The impact of a delayed schedule puts time pressure on the developers which ultimately impacts the quality of the software giving rise to more error and rework which can have a negative effect on making progress on new tasks[15]. Thus the COQ will be higher which impacts the profit margin of the project. Another simulation model was developed by D. Pfahl and K.Lebsanftthat how frequently changing and unstable requirements impact the software project performance [16].Since in order to stabilize the highly volatile software requirement changes, consumes effort and investments in systems engineering, need to be analyzed so that cost of the project can be optimally managed [16]. System dynamics can very well cope with strategic issues of project management which traditional models or approaches cannot. The systematic management of strategic issues can be done through system dynamics. The operational problems within the process are supported by Traditional approaches while greater strategic issues and understanding about various managerial policies are taken care by system dynamics model. Alexandre Rodrigues in 1994 tried an approach to combine the lessons of system dynamics and traditional models within a single integrated project management methodology[17]. The strategic decisions RSPUBLICATION, rspublicationhouse@gmail.com Page 54
are effectively implemented using operational models as tools. Traditional operation control models need not be completely replaced or abandoned rather the existing operational models need to be improved by combining with strategic models. The differences and similarities of traditional and system dynamics model are also very clearly highlighted by Alexandre Rodrigues[17].The risk exposure of projects have considerably increased because of rapidly changing environment and the complexity level of projects. The risk management of projects can also be effectively done using a system dynamics based framework. System dynamics can be used as a useful tool for enhancing the existing Risk management process of PMBOK. Like Risk management process many other processes of the PMBOK framework can use system dynamics optimally.alexandre Rodrigues in 2001 proposed the use of SYDPIM-A system dynamics based Project-management integrated methodology for managing project Risk dynamics [18]. Agile software development is seen as an alternative approach by researchers and practitioners to tackle the traditional problem of cost, over budgeted effort, quality, schedule slippages and mismatch in estimation.a typical 20 working days Iterative development cycle is used in agile software development. In an agile set up emphasis is given on the delivery or release of work products at regular intervals.the length of the iterative cyclehave impact on the outcome of the project. The project outcome may suffer in case of sequential or low levels of agility or in high levels of agility. It is likely that the best outcome is obtained in case of moderate levels of agility. Investigations show that when the iteration cycle is for 50 working days inplace of 20 working days performance is better. The dynamics of agile software development has been analyzed [19] to gain both theoretical insights and practical suggestion for managing agile projects using system dynamics.the AgileGenome has been found to be very useful in the development of Agile Project dynamics model. The characteristics of Agile software development practices can be described by a unifying framework known as the Agile Genome. The Agile Project dynamics can be used for simulating both agile and waterfall approaches. All the seven agile genes are altogether not present in any Agile methodologies like scrum, XP(extreme programming), TDD, Feature driven development and Crystal. The APD model gives us insight in to the reasons why large-scale software projects have been unable to reap the full benefits of agile[20]. The impact of schedule pressure on the economic effectiveness of agile maintenance project is investigated using casual loop diagram and analytical theory of project investment. When the schedule RSPUBLICATION, rspublicationhouse@gmail.com Page 55
pressure is low the maintenance effectiveness of agile project is high and schedule pressure is High effectiveness is low. The study gave a new dimension of quantifying software development process applying analytical modeling [21]. CONCLUSION:Traditional approaches were not adequate enough to cope with the strategic issues of the highly dynamic software project management. In the recent years many project failures have brought to light that only the operative aspects of project management were taken care by traditional tools and approaches. The analysis of software project management was benefited immensely by the application of system dynamics. The insidious feedback-loops, delays, the non-linear relationships including the causal relationship give an insight in to the intra-project forces in describing the dynamic behavior of projects. But to effectively implement the strategic policies and decisions software project management will always require inputs from operational models.bothe traditional approaches and system dynamics models are providing complementary support to software project management. Hence further research need to integrate the best of both worlds and develop a single of methodology. REFERENCE: [1]Rodrigues, A.G., Williams, T.M., System dynamics in software project management: towards the development of a formal integrated approach. European Journal of Information Systems. 6, 51-56, 1997,. [2]Barros M, Werner C, and Travassos G, Evaluating the use of system dynamicsmodels in software project management, in Proceedings of 2002 International SystemDynamics Conference, Palermo Italy, Systems Dynamic Society, 2002 [3]Lin C and Levary R, Computer-aided software development process design, IEEE Transactions on Software Engineering, September 1989. [4] Abdel-Hamid T and Madnick S, Software Project Dynamics,Englewood Cliffs, NJ: Prentice- Hall, 1991. [5] Lin C Y, Abdel-Hamid T, and Sherif J S, Software engineering process simulationmodel (SEPS), Journal of Systems and Software, 38, 263 277, 1997. RSPUBLICATION, rspublicationhouse@gmail.com Page 56
[6] Pfahl D and Lebsanft K, Integration of system dynamics modelingwith descriptive process and goal oriented measurement, Journal of Systems and Software,46(2 3), 1999. (Initial version in Proceedings of ProSim Workshop 1998.) [7]Drappa A and Ludewig J, Simulation in Software Engineering Training, in Proceedings of the 22nd International Conference on Software Engineering, 2000. [8]Abdel-Hamid, T.K. "The Dynamics of Software Project Staffing: A System Dynamics Based Simulation Approach," IEEE Transactions on SoftwareEngineering, forthcoming 1988. [9]Rodrigues AG, Bowers J, System dynamics in project management: a comparative analysis with traditional methods, System Dynamics Review 12(2): 121 139, 1996. [10] Li Lei, Zhang Chen, Li Hong-min, Application of System dynamics to strategic Project Management, The Ist International Conference on information Science and Engineering (ICISE 2009). [11] Brooks F, The Mythical Man-Month, Reading, MA: Addison-Wesley, 1978. [12] Ruiz M, Ramos I, and Toro M, A simplified model of software project dynamics, Journal of Systems and Software, 59(3), 2001. [13]Eberlein, R.L., Simplification and understanding Review,1(5), 51-68, 1989. of models, System Dynamics [14]Boehm, Software Engineering Economics, Englewood Cliffs, NJ: Prentice-Hall, 1981. [15] DeMarco T, Controlling Software Projects, New York: Yourdon Press, 1982. [16] Pfahl D and Lebsanft K, Using simulation to analyse the impact ofsoftware requirement volatility on project performance, Information and Software Technology,42, 2000. [17]Rodrigues A.G, The role of system dynamics in Project management: A comparative analysis with traditional models, Proceeding 1994 International System Dynamics Society, Lincoln MA,USA (1994), 214-225. [18]Rodrigues AG. 2001, Managing and modelling project risk dynamics: a systemdynamicsbased framework,. Presented at the Fourth European Project ManagementConference: PMI Europe 2001, 6 7 June, London. [19]K.VanOorschot, K.Sengupta and Luk van Wassenhove (2009), Dynamics of Agile software development, Proceeding International Conference of the System Dynamics Society 2009. RSPUBLICATION, rspublicationhouse@gmail.com Page 57
[20]FirasGlaiel, Allen Moulton, Stuart Madnick, Agile Project Dynamics: A system dynamics investigation of Agile software development methods, March 2013. [21]Xiaoying Kong, Li Liu, Jing, Modeling Agile Software Maintenance Process using analytical theory of Project investment, Proceeding 2011 International conference on Advances in Engineering,138-142. [22]Ford D and Sterman J, Overcoming the 90% Syndrome: Iteration Management in concurrent development Projects, Concurrent Engineering Research and Application(CERA) Journal, September 2003. [23] Stefano Armenia, Diego Falsini et al., Improving management effectiveness and overall performance of software development projects through a system Dynamics Approach, 22 nd IPMA World Congress, 9-11 November 2008. [24] D. Rodriguez, M. Satpathy and D. Pfahl, Effective software Project Management Education through simulation models: An externally replicated Experiment, PROFES 2004, LNCS 3009, pp.287-301, 2004.. RSPUBLICATION, rspublicationhouse@gmail.com Page 58