Courses
Scientific methods for better decisions
Decision sciences integrate quantitative methods, analytical models, data, artificial intelligence, behavioural insight, and professional judgement. These approaches help decision-makers structure complex problems, assess alternatives, understand uncertainty, and translate evidence into responsible action.
Data Analytics and Artificial Intelligence
Data analytics transforms large and complex datasets into useful information for decision-making. Descriptive, predictive, and prescriptive analytics combine statistics, machine learning, optimisation, and domain knowledge to explain past performance, anticipate future outcomes, and recommend appropriate actions.
Applications include demand forecasting, customer analytics, healthcare prediction, fraud detection, supply-chain planning, intelligent manufacturing, and public-sector decision support. Responsible use requires attention to privacy, security, data quality, transparency, and algorithmic bias.
Performance and Efficiency Analysis
Data Envelopment Analysis is a non-parametric approach for evaluating the relative efficiency of comparable decision-making units that use multiple inputs to produce multiple outputs. DEA constructs an efficiency frontier from observed best performers and identifies opportunities for improvement.
The method is widely used for benchmarking hospitals, universities, public services, banks, manufacturing systems, supply chains, and environmental programmes. CCR and BCC models support different assumptions regarding scale and production conditions.
Decision Analysis
Decision analysis provides a structured framework for comparing alternatives under uncertainty. It combines objectives, possible actions, uncertain events, consequences, preferences, and probabilities to clarify trade-offs and support transparent choices.
Common tools include decision trees, expected-value analysis, utility models, sensitivity analysis, scenario analysis, Bayesian reasoning, and game-theoretic approaches. Applications span business, healthcare, engineering, investment, cybersecurity, and public policy.
Forecasting and Predictive Analytics
Forecasting uses historical information, explanatory variables, and analytical models to anticipate future events, demand, risks, and performance. It supports planning, resource allocation, budgeting, capacity management, and strategic decision-making.
Techniques include time-series models, exponential smoothing, regression, simulation, machine learning, and probabilistic forecasting. Applications include sales, energy demand, financial markets, epidemiology, transportation, operations, and supply-chain management.
Multicriteria Decision-Making
Multicriteria decision-making addresses problems in which alternatives must be evaluated against several quantitative and qualitative criteria. The approach makes conflicting objectives, stakeholder preferences, and trade-offs explicit.
Methods such as AHP, TOPSIS, ELECTRE, PROMETHEE, and multi-attribute utility analysis support ranking, selection, and portfolio decisions. Applications include supplier selection, investment, healthcare, sustainability, infrastructure, urban planning, and technology assessment.
Optimisation
Optimisation identifies the best feasible solution according to one or more objectives and a defined set of constraints. It supports efficient resource allocation, scheduling, routing, production, network design, planning, and system configuration.
The field includes linear, integer, nonlinear, dynamic, stochastic, and multi-objective optimisation, as well as metaheuristic methods such as genetic algorithms and particle swarm optimisation. Modern machine-learning systems also rely extensively on optimisation.
Risk and Uncertainty Management
Risk management identifies, analyses, evaluates, and responds to uncertainties that may influence organisational objectives, projects, operations, investments, public policy, and strategic decisions. It supports resilience by clarifying both potential threats and emerging opportunities.
Common methods include scenario analysis, probability models, sensitivity analysis, decision trees, risk matrices, simulation, stress testing, and portfolio analysis. Applications extend across finance, engineering, cybersecurity, healthcare, supply chains, sustainability, and public administration.
Where decision sciences create value
Decision-science methods support planning, analysis, evaluation, and responsible action across public and private sectors.
Develop knowledge through cooperation
CSDS welcomes enquiries regarding lectures, workshops, methodological training, educational initiatives, and institutional cooperation in decision sciences.