The overall objective of this project is to develop and implement practical prediction models for protected cultivation systems in soft fruit production (strawberries and Rubus crops such as raspberries and blackberries). The project follows a strongly applied, results-oriented approach, combining data acquisition, model development, and real-world validation.
On the data and modelling side, the research focuses on:
- Analyzing sensor-based climate data (e.g. temperature/relative humidity) to simulate inside microclimate conditions based on outdoor weather and different cropping systems.
- Developing and refining prediction models for pest and fungal disease risks.
A key component involves analyzing sensor data to establish relationships between outdoor weather conditions and microclimate conditions inside different protected cropping systems (e.g. tunnels, rain shelters, greenhouses). Two modelling approaches will be explored:
- Mechanistic (white-box) models, using regression techniques to describe the relationship between outdoor and indoor climate variables and their residual patterns.
- Data-driven (black-box) models, including neural networks, to capture complex, non-linear interactions and improve predictive accuracy. Model interpretability will be enhanced using explainable AI techniques such as SHAP, LIME, and ICE plots.
In addition, the project aims to integrate phenological models with AI approaches to predict pest occurrence and fungal infection risks. These models will be trained using field observations, sensor data, and crop development information, and validated through real-world case studies in protected soft fruit systems. The outcome will be robust, hybrid prediction systems that combine biological understanding with advanced AI methods, enabling improved decision support for growers.
Although the focus is definitely on research, the candidate can be asked to give guidance to Bachelor and Master students, to be involved in teaching and to supervise and correct exams (maximum load 20%).
The successful candidate will be able to enrol in a fulltime PhD position at the Operations Management research group (Faculty of Economics and Business) of KU Leuven. Successful completion of the PhD project will lead to obtaining a doctoral degree (PhD). He/she will find a dynamic and pleasant working environment in Brussels and Leuven, in groups that are actively involved in scientific research at the highest international level. Candidates will be expected to participate in seminars and international conferences.
