A temporal case-based reasoning approach for performance improvement in intelligent environmental decision support systems

One of the major problems when designing control and supervision systems for environmental systems is the need to be adapted to the particularities of each system. The use of Artificial Intelligence (AI) techniques instead of classical control approaches have been used in recent years in the design of Intelligent Decision Support Systems (IDSS). A static Case-Based Reasoning (CBR) approach for the control and supervision of environmental systems has been proposed in previous works, providing a general efficient methodology to allow scalability to further types of systems. However, the dynamic nature of environmental processes suggests temporary dependencies between cases, hence the use of a temporal CBR (TCBR) approach could provide better performance. The main contribution of this research is to propose a new TCBR method providing improved retrieval process, leading to improved overall performance by considering the dependency of consecutive cases. Thus, the retrieval process is addressed considering not only a single case but a set of consecutive cases named episodes. Our proposal is based on using fixed-length episodes. The approach presented has been tested in a real facility within the ambit of a local water administration in the area of Barcelona. The results indicated that the TCBR approach improved the accuracy of the obtained solutions and case diagnosis with respect to the single-case CBR approach. A tuning process with different episode lengths is performed in order to find a good trade-off between performance and computing time.