Confiabilidade humana e o funcionamento normal de uma refinaria de petróleo

Abstract

There is a distance between a design as planed and one developed effectively. It varies according to both the constraints involved in the design process and accuracy of design conception. Many projects are designed without taking into account the user‟s real needs. It makes difficult or impossible the developing of the tasks. To improve this scenario, the Ergonomics seeks to understand the activity to incorporate the workers perspective in the design and operation of sociotechnical systems. To reach this goal, it uses the Activity Analysis approach. It is on the situated activity that emerge the own skills and characteristics of human, in order to ensure the system stability. Thus, the human is the reliability factor of the system. It is called Human Reliability. It follows in this research, the assumption that the Activity Analysis is the most appropriate method to provide subsidies for a design focused on the needs of the task, increasing the supports that help ensure the operational reliability of the system. It were used analyzes carried out in a Petroleum Refinery for 3 years and a half of study. They enabled the development of designs that take into account the role of refinery workers as the most experts of his work. The analyzes used photographs and videos to understand more deeply the activity of the refinery workers. From these analyzes were extracted actions and strategies used to refinery workers to correct instabilities that are present at all times. It can generate incidents which lead to lose of control of the system as a whole, if not noticed at time. The workers do it using the representation of the system. The actions were classified into five categories, according to the elements mobilized for the recovery of the normal state of the system. It is proposed a Design Resilient Model that incorporates these actions and aims to ensure that new strategies, built from new constraints, are constantly evaluated and incorporated on the project, if possible. It emphasizes that even with the construction of the Model, it been possible realize that the conflicts that operators face in the daily routine of the Refinery are dynamic and cannot be translated for a Model. There will always be a remnant. There will always be hysteresis. Thus, the activity will continue to be the place where operators respond to these conflicts. Showing that, for the most appropriate to the needs of the task that is a model, the refinery workers will continue having to resolve conflicts in situated action.