Feven Desta is a post-doctoral researcher in the section of Resource Engineering. Her research areas of interest include machine learning, data fusion, sensor technologies, material characterization, automation, remote sensing, geoinformation, geostatistics, environmental systems modelling, geology, environmental studies and mining. She conducted her doctoral research on “Sensing and data fusion opportunities for raw material characterization in mining”. Her main interest is in the development of methodological approaches using data fusion and chemometrics concepts to derive study-specific insight from sensor data. These approaches can be applicable in geological, mining, and environmental studies. The approaches can also be upgraded for insitu and automated material characterization.
- Utilize sensor-derived data for raw material characterization. The sensor technologies include visible near-infrared (VNIR), short-wave infrared (SWIR), mid-wave infrared (MWIR) and long-wave infrared (LWIR), hyperspectral imaging, RGB imaging, Raman spectroscopy
- Machine learning
- Data mining
- Data acquisition, pre-processing, modelling, information extraction and validation
- Data fusion
- Remote sensing
- PhD. Resource Engineering, Delft University of Technology, The Netherlands
- MSc. Geo-information science and earth observation (Major – Geoinformatics), University of Twente, The Netherlands
- MSc. Geo-Environmental Systems Analysis, Addis Ababa University, Ethiopia
- BSc. Geology, Addis Ababa University, Ethiopia
- Prior to TUD, Feven worked for eight years in the area of geosciences as a geotechnical engineer, geospatial analyst, GIS/RS and environment specialist, and geophysicist.
Participation in research projects
Current research projects
INSITE: International consortium to develop in situ ore grading system using LIBS in harsh environments
Past research project
Real-Time Mining (RTM): International consortium to develop a real-time process-feedback control loop linking online data acquired during extraction at the mining face rapidly with a sequentially up-datable resource model associated with real-time optimization of long-term planning, short-term sequencing and production control decisions. The project included research and demonstration activities integrating automated sensor-based material characterization, online machine performance measurements, underground navigation and positioning, underground mining system simulation and optimization of planning decisions, state-of-the-art updating techniques for resource/reserve models.