Warning Information System for Humanitarian EmergencieS (WISHES)

The implementation of an information system, aimed at forecasting the risk of humanitarian crises with special reference to women and children needs, is an ambitious project. This activity is rather complex and requires the cooperation of experts from various fields. Moreover, the system has to be designed according to the existing constraints (such as the financial resources, the project completion time, the possibility of establishing cooperative relationships between components of Save the Children and external partners in order to improve the data collection process).

The system should help to assess the probability that natural or man-made disasters (conflicts, refugees’ displacement, …) could affect a certain area putting the life of a large number of people at risk. Moreover, it should provide a number of statistical indicators which are able to measure countries’ vulnerabilities and to anticipate possible crises. Finally, caution must be exercised when choosing information sources. Any system must combine multiple sources of information (in order to guard against the possibility that a source may cease to operate, to reduce the prediction error and to avoid systematic distortions). The danger that contradictory signals enter the information system without an appropriate evaluation of their reliability should be avoided. The uncertainty of the results produced by any statistical analysis is, in fact, significantly influenced by the quality of the data on which it is based.

This exploratory study commanded by Save the Children Italy concluded that the accomplishment of such an objective heavily depended on the initial system design which involved the specification of the data to be collected, the organisation of the information flows, the identification of the statistical techniques for the data processing and the construction of a user-oriented communication interface. IARAN designed a forecasting exercise, based on a statistical model, to be conducted exclusively in a probabilistic framework and assuming specific scenarios. Ultimately, the implementation of the system was constrained by a lack of resources to continue to run the project. However, the learning from the IARAN work is detailed in a Feasibility Study published by the University of Naples Federico II.