Topic ID No.: 2026-G08
Title of research topic:

Fast and low-cost near-infrared (NIR) spectroscopy for plant oil quality assessment at processor level

 

Description:

This project investigates the use of low-cost near-infrared (NIR) spectroscopy for rapid, non-destructive assessment of plant oil quality at the processor level. It focuses on developing and validating calibration models to predict key quality parameters in locally produced oils and evaluating portable NIR devices under real processing conditions.

Target region or country (if applicable):

South Africa

Topic background information / scientific relevance:

Ensuring consistent quality of edible oils is a persistent challenge for processors in sub-Saharan Africa, where conventional chemical analyses remain expensive, slow, and laboratory dependent. Near-infrared (NIR) spectroscopy offers a transformative alternative: a rapid, non-destructive, and reagent-free analytical approach capable of predicting multiple quality indicators such as oil content, fatty acid composition, peroxide value, free fatty acids, and moisture. Studies (Armenta  et al, 2007; Zhang et al. 2017) demonstrate NIR’s ability to provide reliable results when coupled with robust chemometric models. However, limited local calibration datasets, instrument standardisation, and validation under African processing conditions hinder adoption. This project will develop and validate low-cost NIR models adapted to small- and medium-scale oil processors in South Africa. Using regionally sourced oilseeds (soybean, sunflower), the candidate will calibrate and test predictive models for quality parameters, benchmarked against reference chemical analyses. The research will support value-chain transparency, reduce quality-control costs, and promote digital transformation within African agri-food processing.

Research objectives:
  • Develop locally adapted NIR calibration models for key oil quality parameters.
  • Validate models against standard chemical assays under processor-level conditions.
  • Evaluate low-cost portable NIR devices for on-site oil quality control.
  • Integrate data analytics to enable real-time decision support for oil processors.
Required skills and qualifications of the applicant:
  • Background in food science, agricultural engineering, chemistry, or data science.
  • Experience in laboratory or industrial quality-control settings advantageous.
  • Familiarity with spectroscopy and machine learning preferred.
  • Programming skills (Python, R, MATLAB) and solid statistical literacy desirable.
  • Strong motivation for interdisciplinary research at the interface of analytical technology and sustainable food systems.
Contact person and institute in charge:

Prof. Dr. Joachim Müller. University of Hohenheim. Agricultural Engineering in the Tropics and Subtropics.

joachim.mueller@uni-hohenheim.de