PhD “Machine learning assisted prediction of plastic recyclate quality Maastric

About the position:
Mechanical recycling of plastics is the most efficient way to treat plastic waste sustainably. However, the quality of recyclates is usually compromised, leading to products with inferior properties. The quality of reprocessed samples (recyclates) depends on many factors and is also application dependent. An effective recycling process benefits from products with low variability in properties such as rheological and mechanical characteristics.

To achieve this, several tests are required, and a sorting and recycling line needs to be optimized to predict the quality of the output. The predictions should be carried out in a scientifically and economically viable method to assure applicability in industry.
This project applies state-of-the-art machine learning models combined with a plethora of experiments to generate data for training the model. The significance of this project is governed by the dependency of the plastic industry on recyclates with consistent and high quality. The quality estimation is a topic which our quality model can address. However, predictions require a step backward, and to impute the data gaps to assure inline sorting and separation is feasible with high accuracy and high rates.
The result of this project can support the end users of the recyclates as well as the producers of the regrinds to position their products in the right market.
This PhD topic is meant to delve deep into polymer science and machine learning and therefore is highly interdisciplinary. Candidates with a background in polymer science and engineering and machine learning are encouraged to apply. This position provides the candidate with the necessary experience to flourish in the upcoming fields of circularity with a flavor of data science, engineering, and polymer science.

Requirements

• A polymer-oriented materials scientist/engineer with a background in data science. You should have a distinct track record in at least one of these polymer groups: packaging polymers like polyolefins and/or PET, technical polymers like ABS, PC, and (HI)
  PS. A background in textiles engineering or rubbers does not meet these requirements.
• Experience with polymer processing machinery such as injection molding and extrusion.
• Experience with polymer characterization methods such as tensile, impact, rheology, IR, Raman, and HDT are required.
• Fluency in Python or equivalent is a bare minimum. Therefore, candidates who aspire to learn the basic language models within the PhD are not encouraged to apply.

This position is available from January 1st, 2025, for a planned duration of 48 months. You will work in an international interdisciplinary team with cutting-edge equipment in melt-processing machinery, rheological characterization, and polymer analytics. You will be supervised for your interdisciplinary project, and therefore, world experts in data science, polymer science, and engineering are supporting you.

What we offer
As a PhD candidate at the Faculty of Science & Engineering
, you will be employed by the most international university in the Netherlands, located in the beautiful city of Maastricht. In addition, we offer you:

• Good employment conditions. The position is graded according to UFO profile PhD, with a corresponding salary from €2770.00 in the first year to €3539.00 gross per month in the fourth year (based on a full-time employment of 38 hours per week). In addition to the monthly salary, an 8.0% holiday allowance and an 8.3% year-end bonus apply.
• An employment contract for a period of 12 months with a scope of 1.0 FTE. Upon a positive evaluation, an extension of 3 years will follow.
• At Maastricht University, the well-being of our employees is of utmost importance. We offer flexible working hours and the possibility to work partly from home if the nature of your position allows it. You will receive a monthly commuting and internet allowance for this. If you work full-time, you will be entitled to 29 vacation days and 4 additional public holidays per year, namely carnival Monday, carnival Tuesday, Good Friday, and Liberation Day.
  
  If you choose to accumulate compensation hours, an additional 12 days will…