CMMI (Capability Maturity Model Integration) consists of best practices for the development and maintenance activities of software products and services. One of the most important areas covered by CMMI is Project Management (PM) and the processes associated to the conduction of this activity according to the recommended practices proposed in the framework. Although CMMI describes what must be carried out to achieve a particular maturity level and effectiveness in the processes implemented in the organization, it does not provide the methods, guidelines and/or procedures to implement those processes. Consequently, process design depends completely on the organization, using for that purpose their previous knowledge and their experience with all the implications that this situation entails. As a result, organizations lack of a solid foundation to face situations such as: (1) how to improve their current processes or, (2) how to implement their planned processes in an optimal way, taking advantage of the assets that they already have available.
After studying and analyzing the application of simulation techniques for process improvement in the realm of CMMI and, in order to cover the existing gap detected between CMMI and PM, we designed and implemented a hybrid simulation model, integrating Discrete-event and Agent-based simulation paradigms, named SIM4CMMI. The proposed model aims to help organizations in the design, assessment and optimization of CMMI organizational and PM processes with the purpose of achieving higher CMMI maturity levels.
SIM4CMMI supports the Specific Practices (SPs) covered by CMMI Level 4 Process Areas, namely Organizational Process Performance and Quantitative Project Management. Additionally, the model has been designed to cover the SPs of Project Planning, Measurement and Analysis, and Project Monitoring and Control Process Areas belonging to CMMI Level 2, Risk Management, Organizational Process Definition, Organizational Process Focus, Decision Analysis and Resolution and Integrated Project Management Process Areas belonging to CMMI Level 3. Moreover, it also covers Organizational Performance Management and Causal Analysis Resolution Process Areas belonging to CMMI Level 5. The scope of the model covers the whole life cycle of projects conducted in a real-life organization.
After model validation and verification, we performed a series of running-simulation experiments with the aim of evaluating the effect of different values of input parameters on key process indicators. For this purpose, highly probable real-project scenarios were designed, simulated and analyzed, from risk impact analysis to piloting process improvements before implementing them in the real world.
In addition, simulation-based optimization experiments were also performed. In this case, our goal was to find the values of the input parameters that either maximize or minimize a set of objective functions. To this end, single- and multi-objective optimization experiments using heuristic and metaheuristic methods were conducted to find the values of the input parameters that result in optimal values for the most relevant outputs of a project, namely time, cost and quality.
The findings of this thesis allow us to consider SIM4CMMI as a useful tool to assist organizations in decision-making in the realm of CMMI. SIM4CMMI enables to mature the capability of project management processes, especially in the high maturity levels that frequently organizations struggle to reach, and supports the improvement of the key capabilities that elevate the organization’s performance, quality and profitability.
© 2001-2024 Fundación Dialnet · Todos los derechos reservados