Data-Driven Decision-Making Lab (D3)

Data-driven decision-making can cross many application areas, from finance to healthcare and even soft OR. D3 offers a common space where members from different groups and expertise can collaborate. It also seeks to foster cross-disciplinary collaborations both within and outwith the University. The main research areas which form the core of D3 are:

Contextual optimization
Data-rich applications of OR

Contextual optimization

Diagram illustrating the data-driven decision-making process, highlighting key steps and flow of information.

The recent explosion of data availability and the development of sophisticated and accurate predictive modelling tools have led to an important question: what does one do with good predictions? The idea of data-driven decision-making is to integrate data models within decision models so as to optimise decision making in the face of uncertainty. Traditionally, this optimisation would involve only the main information necessary for decision making. Optimising a humanitarian supply chain would access demand information to make relief items allocation decisions. Optimising power systems expansion with green generators would take into account the intermittence of wind and solar power. Over time, it has become evident that side information (also known as covariates) is as crucial in decision-making. The age, vulnerability and affluence of affected people are essential in making equitable humanitarian relief allocation. Other meteorological information matter when deciding the types of power generators to install in order to avoid blackouts.

The field of data-driven decision-making, also known as contextual optimization, is rapidly expanding and recent research have unlocked the huge potential of the area in not only analytic and algorithmic terms, but also in actual practical impact.

Data-rich applications of OR

Diagram illustrating the various stages in the data-driven decision-making process.

Organizations now have access to unprecedented volumes of information and must make critical decisions in a data-rich world. The critical question for modern organizations is no longer if they should use this data, but how to use it to optimize their operations in an increasingly uncertain world. By integrating mathematical programming with large-scale data analytics, we can create decision frameworks that are not just smart but truly resilient. Real-world applications include:

Humanitarian Logistics

The most critical test of resilience. When a disaster strikes, decisions must be made in an instant. This requires pre-processing diverse data streams—satellite images of damaged infrastructure, pre-disaster population statistics, and logistical supply levels—to construct scenario trees that inform stochastic models. These models produce optimal, mitigation, preparedness, response and recovery decisions.

Power Generation Planning

Optimizing the energy mix (renewable vs. thermal, for instance) over days and seasons by processing weather forecasts, historical demand data, and market price volatility to ensure a stable grid.