Designing a human-centred, mobile interface to support real-time flood forecasting and warning system

Abstract

There is a demand for human-centred technology which improves the management of flood events. This thesis describes the development, design and evaluation of a mobile GIS-based hydrological model. The application provides hydrological forecasts and issues flood warnings. The thesis reports on the usability and practicality of the application. The application, a mobile-based hydrological modelling system, permits the integrated handling of real-time rainfall data from a wireless monitoring network. A spatially distributed GIS-based model integrates this incoming data, approximating real-time, to generate data on catchment hydrology and runoff. The data can be accessed from any android-based mobile computer or mobile phone. It may be further analysed online using several GIS and numerical functions. A human-centred approach to design was taken. Design guidelines for a user-centric application were developed and deployed in the first prototype. There was intensive consultation with potential users. Particular attention was paid to the ease of use of the mobile interface. Users’ needs and attitudes were relevant in the achievement of a highly functional but intuitive interface. The first prototype underwent intensive testing with users. After the initial testing of the first prototype an interactive approach was taken to development. This generated a high-fidelity prototype which was matched to the taxonomy from a user’s mental model. Users were interrogated under controlled laboratory conditions as they performed predefined tasks which were selected to generate data across all aspects of the system and to identify weaknesses. Subsequent to this work there was a major prototype re-design. User test data, identified issues and an improved mental taxonomy closer were used to further refine the application. Of particular note was new functionality which aligned with user expectations and enhanced the applications credibility. The final evaluation of the system was undertaken with diverse subjects. Overall, the subjects considered the system efficient and effective. Users said the system was easy to learn and integrate into their work. Task completion rates were satisfactory. The final interviews with users confirmed that the application was ready to proceed to the implementation phase.