Select Projects

A selection of projects where we've helped organisations transform data into actionable insights and measurable outcomes.

Windstorm risk assessment for offshore wind farms in the North Sea
Finance

Windstorm risk assessment for offshore wind farms in the North Sea

Insurance/Reinsurers Research

In case of a 1-in-200-year windstorm event, helped insurers and reinsurers quantify the Value at Risk (VaR) for off-shore wind farm assets located in the North Sea. This full study can be found here: https://onlinelibrary.wiley.com/doi/full/10.1002/we.2351 

Key Outcomes

  • The solvency capital requirement (SCR) in case of yawing turbines was estimated at €0.049 billion (2019)
  • With the assumption that wind speeds grow by 5% and the frequency of storms increases by 40%, the SCR is seen to rise substantially to €0.264 billion (2019)
  • Climate change could increase the relative SCR to levels as high as 0.75%.

Technologies & Methods

MATLABMonte Carlo Simulation
Simultaneous Forgery Identification and Localization in Paintings using Advanced Correlation Filters
ForensicsResearch Services

Simultaneous Forgery Identification and Localization in Paintings using Advanced Correlation Filters

Advances in high-resolution digital imaging have driven growing interest in statistical and image processing techniques for authenticating works of art, with applications extending to biometrics and counterfeit currency detection. In this work, we demonstrated a closed-form solution for distinguishing forged paintings from originals, including detecting whether an original had been modified. This full study can be found here: https://ieeexplore.ieee.org/abstract/document/7532336

Key Outcomes

  • Achieved detection accuracies of up to 100%
  • Ability to distinguish between a low-quality digitized representation of a painting and its forgery
  • Ability to specifically indicate where the differences occur and where a given replica is particularly faithful to the original

Technologies & Methods

PythonMATLABOptimal Trade-off Synthetic Discriminant Function (OTSDF) Filter
FinanceData Analysis, ML & AI

Credit Scoring Engine for MVend Group

Mvend Group

Leveraged data pertaining to repayment patterns, member savings/balances, and demographic attributes (age, gender) to develop individual and group-level credit scores for MVend’s savings groups.

Key Outcomes

  • Achieved individual-level credit scoring relating to lending and re-payment patterns
  • Developed group-level credit scoring incorporating gender considerations
  • Strengthened Mvend Group's loan allocation criteria

Technologies & Methods

RPostgreSQL
Manufacturingsoftware-development

End-to-end Inventory Management

Roba Industries

Built a comprehensive inventory management system for Roba Industries, enabling goods requests and transfers from production warehouses down to the smallest mobile warehouse. The system also supports sales operations and integrated data analytics.

Key Outcomes

  • Improved visibility into movement of goods from production to final sale
  • Reduction in revenue leakages
  • Regional insights into what goods are fast moving and otherwise

Technologies & Methods

Next.jsPostgreSQL
Education Policy Dashboards
EducationData Analysis, ML & AI

Education Policy Dashboards

Ministry of Education of Rwanda

Built policy dashboards for the Ministry of Education using data spanning all education levels, from pre-primary to upper secondary. The dashboards integrated ministry data with external sources, notably the National Institute of Statistics of Rwanda, to compute additional education-specific indicators for evidence-based policymaking.

Key Outcomes

  • Ability to track and monitor progress of the Education Sector Strategic Plan indicators
  • Comparison of key indicators across schools and regions
  • Drill down capability into regional and school performance

Technologies & Methods

R ShinnyPostgreSQL

Have a Project in Mind?

Let's discuss how our expertise can help solve your data challenges.

Get In Touch