Agility as a tool for the Management of Systems Engineering Projects

Project performance is considered as an important factor to ensure the success of a project. Companies are interested in the use of efficient practices through efficient methods and tools to design and deliver innovative products and services and decrease the time to market. Project duration, costs, and performance are aspects that normally face changes during the project development. These changes should be treated by using adapted and optimized processes in order to better control, coordinate, manage, and improve projects. Agile methods seem to be efficient for the management of successful projects, however they are mainly use in companies where the business domain is software. Agile methods recently received a growing interest from industry and now are well accepted and deployed in software engineering. This thesis thus tackles the point of transferring the agile methods from software to systems engineering, and issues that are induced.The report first introduces the notion of agility and the birth of the agile movement as well as the principles and values of agile software development. It also presents the main agile methods, as well as other philosophies that share a number of similarities with Agile. Project attributes can be defined, from the literature, to help contextualizing agile projects; we describe and use these project attributes to compare different agile methods and identify the differences between them. Agile and Lean are compared to determine why Lean is used in software development, and how it differs from other agile methods. Finally, we identify several issues to transfer agile methods in the context of systems engineering.We then focus on the understanding of agility in systems engineering. Two meanings of "agile" are found in literature. Considering that agility is focused in the rapid change of convincing, designing, and implementing processes of products and systems in an easy way, we explore the question of introducing agility in systems engineering. A first analysis is led to identify any notion of agility in systems engineering standards. The results of this analysis help us to highlight the issues and challenges of transferring agility into systems engineering. Focusing on the issues, we then present a four steps research methodology. The first step aims to define a contextual model for systems engineering development. The contextual model contains the organizational factors and the project attributes for engineering projects. This contextualization lead us to identify if and which agile method could be used for the management of engineering projects (step two). The step three justifies our selection of the Scrum Framework, between agile methods, for the management of engineering projects. Scrum Practices are defined and evaluated in the project attributes for engineering projects. However, several difficulties are identified and listed while using the Scrum Practices in engineering projects. Finally, the step four proposes some solutions to solve a set of difficulties.This work finally proposes the use of Scrum Practices in two engineering projects. An educational project is analyzed first. This project aims to develop a connected robot. By starting from the contextual model for systems engineering development (cf. section III.4.2.d), we characterize the project to identify what type of project is, then we propose the use of the graphical view of the Scrum Framework to plan the development of the robot. Following the same schema, a second industrial project is analyzed. The second project aims to develop an automotive application for engine management.