Any views expressed within media held on this service are those of the contributors, should not be taken as approved or endorsed by the University, and do not necessarily reflect the views of the University in respect of any particular issue.

School of Mathematics

School of Mathematics

Blog for the School of Mathematics community

MSc dissertation projects: 2023 case studies

Our MSc programmes offer students the opportunity to complete their dissertation project in partnership with industry. Read two case studies below from our most recent series of MSc projects; one consultancy style project, and one individual project.

Project Simply Business : Property insurance risk assessment: using public data to estimate building features that affect insurance value

MSc Statistics with Data Science

Consultancy style project

For estimating premiums, landlord and house insurance uses risk factors such as size of the property, number of rooms and bathrooms, age of constructions, wall and roof type, flood/fire assessments, type and safety level of the surrounded area (urban, rural) .

The ability to estimate some of these factors from publicly available images increases the accuracy of the insurance premiums we estimate. 

The goal of the project was to create a number of models (such as neural networks) that, as input, use known building characteristics combined with visual features extracted from the forefront image to predict the property attributes. This will help us to automate the process of claims risk evaluation. 

  1. Using the publicly available data to estimate the property attributes such as number of bedrooms, square footage, number of bathrooms and type if property (flat, terraced house, semi-detached, detached house) 
  2. Using the satellite and image data (e.g. google maps) estimate the the construction characteristics of the property (type of roof, type of windows) 
  3. Select models that give the best accuracy and explain why you think the approach you took gave the best performance. 


Gordon Baggott, Head of Machine Learning from Simply Business said:

University of Edinburgh’s Statistics with Data Science (SwDS) consultancy programme has helped us compare different deep learning approaches to predicting property types. As a small business insurance broker, this has helped us decide how to solve an active business question so we can improve our customer journeys. Working with the SwDS students is always really enjoyable and continues to provide us with useful inputs to Simply Business.


Flexitricity:  Optimal bidding strategy for grid-scale batteries participating in power markets and grid services

MSc Operational Research with Data Science

Individual Project

As the UK power network continues to decarbonise, decreasing reliance on coal and gas, and increasing usage of renewables, additional services are required to help stabilise the electricity system. In addition to participating in wholesale power markets, Battery Energy Storage Systems (BESS) can provide a range of services to National Grid ESO – supplying energy, absorbing energy (creating demand), and helping to smooth out imbalances in the energy network.

Flexitricity supports this by incentivising the use of these flexible systems. In making BESS worthwhile investments more units are brought online, thus increasing National Grid’s ability to utilise renewable energy.  

Optimising a battery for revenue is more complex than simply charging (buying) when power prices are low and discharging (selling) when prices are high. To maximise the potential revenue, a battery may operate in up to 10 different markets from day ahead trading to power delivery time in a 24-hour period. Bid prices must be entered in different auctions at different times of the day to maximise revenue, noting the various risks associated with potentially not securing contracts by being out-bid, and ensuring optimal schedules for batteries which respect operational constraints.  

Using optimisation algorithms and modelling techniques, our student worked with the data team at Flexitricity to develop a novel approach to identifying optimal bid prices to maximise revenue of our battery portfolio. 


Steve Sinclair, Head of Data Science at Flexitricity said:

Collaborating with a student from the University of Edinburgh was a wonderful opportunity for Flexitricity to bring a fresh viewpoint and mathematical rigor to a tricky problem. UK power markets are complex and take time to fully understand, but the OR with Data Science student made a genuine and valuable contribution to the team’s work. The project had a positive impact on Flexitricity’s portfolio, grid stability and, ultimately, on net zero goal.





Report this page

To report inappropriate content on this page, please use the form below. Upon receiving your report, we will be in touch as per the Take Down Policy of the service.

Please note that personal data collected through this form is used and stored for the purposes of processing this report and communication with you.

If you are unable to report a concern about content via this form please contact the Service Owner.

Please enter an email address you wish to be contacted on. Please describe the unacceptable content in sufficient detail to allow us to locate it, and why you consider it to be unacceptable.
By submitting this report, you accept that it is accurate and that fraudulent or nuisance complaints may result in action by the University.