Optimizing design, maintenance and test tasks in Complex Industrial Systems

J. NÚÑEZ MC LEOD; S. RIVERA; J. BARÓN

Santiago, Chile

Congreso; International Conference on Production Research Americas 2004; 2004

The International Foundation of Production Research

A new approach to optimize design and maintenance and test tasks, based on global risk reduction for complex industrial systems (e.g. fossil and nuclear power plants), is described in this paper.  Regading design, the optimization is performed on components quality as well as redundancy levels.  While in maintenance and test the optimization is focused on scheduling tasks and human reliability.  The models based on probabilistic safety analysis (PSA) are used to evaluate several design schemes and schedules proposed by a hybrid genetic algorithm. The best alternative is chosen to minimize the economical risk of downtime for production or the risk to have an accident for all events considered in the PSA. This approach has resulted in a new methodology to assure the risk for complex industrial systems in a global way. So, it is possible to simultaneously consider several aspects such as component qualities, redundancy levels, task schedules for maintenance or tests tasks, and reliability human as a whole.