Project staff and support

Stephen Glavin (Intern)
Abhijit Sengupta (Company supervisor)
Frank Smith (Academic mentor)
Vera Hazelwood (Technology Translator)

This 6-month Internship project was carried out at Unilever Corporate Research in conjunction with UCL. It is part of the KTN’s Industrial Mathematics Internships programme, co-funded by EPSRC.

"I recently completed a six month internship at Unilever with the Industrial Mathematics Internship Programme. The internship took place half way through my PhD and gave me the opportunity to work in an unrelated subject area over the summer term. I found this very helpful for a number of reasons. Firstly it allowed me to broaden my mathematical knowledge by exposing me to an area that I was not so familiar with, yet still allowing me to apply skills I had developed during my time at university. It also gave me the chance to experience working in an industrial research setting, which is a great help with choosing the kind of career I would like to pursue after completing my PhD. The internship allowed me to strengthen team working skills through working collaboratively with others in industry and also gave me the opportunity to present work during team meetings. There is also, now, a possibility for future collaborations and the work may even be useful in my final thesis write up. Overall I really enjoyed my six months with Unilever and was offered a lot of support and help by my industrial supervisor, academic supervisor and also the technology translator supplied by the KTN," said intern and PhD student Stephen Glavin, University College London.

Project summary

150 million times a day, someone somewhere chooses a Unilever product. Unilever is one of the largest consumer goods companies in the world. Its mission is to add vitality to life, meeting everyday needs for nutrition, hygiene and personal care. It manages around 400 brands spanning 14 product categories.

This internship project contributed to Unilever’s aim of developing a holistic model of shoppers’ behaviour, taking into account heterogeneity in their demographic profile, as well as past responses to pricing, promotion and marketing strategies. Using large disaggregated panel data sets of supermarket transactions, Stephen’s work developed, implemented and validated an agent-based model (ABM) of consumer choice.

A key challenge for Unilever’s researchers has been to come up with a validation methodology using real-life data for the choice models that they have been developing. Stephen’s contribution as an Intern goes a long way in taking those very important first steps in the validation exercise. These are important steps, not only for this project, but also as a contribution to the growing literature on the use of ABMs to study consumer markets. Unilever are not only more confident now of the use of the ABM methodology for business purposes, but will also benefit from the increased visibility of their research group within the ABM community.

“The project has been a great success. It has played a key role in taking a step towards one of the challenges we were facing – i.e. validation of agent based consumer choice models. As Stephen’s industrial supervisor, I found the experience extremely rewarding and I look forward to future collaborations and publications”, said company supervisor Dr. Abhijit Sengupta, Agent Based Modelling Group, Mathematical and Psychological Sciences, Unilever Corporate Research.

“The project matched well with the original plans but was also allowed to veer a little in a constructive way. It benefited all and involved some new departure for all sides concerned. The venture has been very refreshing from my (academic) side”, said academic mentor Professor Frank Smith, FRS, Goldsmid Professor of Applied Mathematics from University College London.