Forest fires represent a constant threat to ecological systems, infrastructure and human lives. Great affords are therefore made to achieve early forest fire detection, which is traditionally based on human surveillance. A rather more advanced approach to human forest fire surveillance is installation of remotely controlled video cameras on monitoring spots. The next, even more advanced step is automatic surveillance and automatic early forest fire detection. In almost every country which encounters high risk of forest fires at least one such systems was developed, proposed and implemented. In Croatia that is iForesFire developed at University of Split. Croatia belongs to countries with enhanced summer forest fire risk particularly the Dalmatian coast and islands. Because of that a lot of efforts have been made in forest fire prevention. Split and Dalmatia County is the leader in those efforts. Since 2003 University of Split, Split and Dalmatia County authorities and Fire Brigade Administration for the Coast work together on development and establishment of holistic forest fire prevention approach. Its technical part is based on automatic forest fire monitoring network and advanced information system. One of quite important tasks in development of the forest fire monitoring network is the monitoring stations location determination. This paper describes the approach applied in Split and Dalmatia County. It was based on Analytic Hierarchy Process (AHP) as multi- criteria decision making procedure and intensive use of GIS data. The final result was the network of 56 monitoring location optimally located according to set of various fire- monitoring, technical and economic criteria.

Computer science