By definition, a system is comprised of hardware, software and "liveware". It also interacts with other systems composed themselves of those elements. However, the "human" element tends to be neglected in many projects, leading to unsafe or inefficient systems. Although some studies have shown that sociotechnical approaches to project management can generate economic gains of 20%, not to mention social gains, in practice, few projects integrate human factors correctly.

Many reasons can explain this lack of integration. Humans alone are much more difficult to model and understand than technology. When considering groups or organizations, the problem increases exponentially. Hence, traditional engineering and risk management methods cannot be used to address the human side of a system. There exist approaches and methods to use our current understanding of human behavior, however these tend to be understood and used only by a small number of specialists. Most project managers, designers and engineers have insufficient knowledge of their existence or do not understand how to make good use of them.

There are two major challenges in the integration of human factors. The first one is to justify an interest in such an approach. Given the educational background and experience of many engineers, this is no easy task. The SNCF (French Railways) has chosen to face this challenge and achieved quite good results. However, this does not solve the problem, as project managers and engineers then request tools and methods. Fulfilling this need represents the second challenge. This is the subject of this study: to make a shift from technology-centered approaches to design and risk management to a more sociotechnical approach thanks to a macroergonomics project framework.

Human factors engineering and ergonomics is a multi-disciplinary domain. It goes from human resources management to physical ergonomics and integrates such subjects as psychology, sociology and human reliability. To improve the reliability or efficiency of systems, one approach is to develop a single tool addressing one aspect of human factors or integrating it with one kind of activities. However, many of those tools already exist, even if they have remained at the state of research results yet or been applied only in some very specific sectors.

Hence, for this research, it was decided to develop a method that covers the whole process of a project and contains the different considerations related to human factors as well as the activities required to ensure the safety of the system.

Recent research led by the US Army and adapted by the UK and Canadian Armies as well as Eurocontrol have lead to the emergence of a new discipline called Human Factors Integration (HFI). This discipline proposes a project management process that covers different domains of human factors: manpower, personnel, training, ergonomics, safety, health and hazards, survivability. HFI is a good starting point but it responds only partly to our expressed need. Indeed, the SNCF requires a more general approach, easily accessible, with a greater emphasis on organization and risk management.

During this study, the HFI method was extended based on recent research results, especially in human and organizational reliability. The main improvements made are the addition of the "organization" domain and the development of safety-related activities. Many other principles were also integrated including barriers, prescribed vs. real tasks, redundancy, recovery, degraded situation, system dynamics and measurement. Some interests of this method are its inheritance from systems engineering, its capacity to be utilized by users from different cultures and experience, and its independence from specific models of human behavior or task processing. The main output of the study is a documentation of this method defining the activities and tasks for each phase of the project as well as the composition of the team.

The method was evaluated based on its application on the "Sécurité des Travaux Organisation Réalisation Préparation" (STORP) project. This project aims at redesigning the infrastructure maintenance system of the SNCF, modifying the concepts, principles, guidelines and documentations, in order to improve its efficiency and safety. This application enabled to test the coherence and usability of the method, as well as highlight its main advantages, while underlining and improving the human factors integration in STORP. Through this evaluation, this study constitutes one of the first attempts to apply HFI to a non-military domain and to non-specific projects.