For many years, governments, banks, hospitals and schools have depended on data-driven tools to determine which organisations operate efficiently and which fall behind. Yet new research from the University of Surrey suggests that these long-trusted methods may be offering an unreliable picture. The study argues that traditional efficiency rankings consistently overlook how performance shifts over time and the impact of unexpected shocks such as recessions, pandemics, or major supply chain disruptions. By treating efficiency as a fixed snapshot rather than a changing process, these older models risk producing assessments that are both misleading and unfair.
The research, published in Expert Systems With Applications, introduces an alternative approach, Time Envelopment Analysis (TEA). Unlike conventional models, TEA is designed to track how an organisation’s efficiency changes across multiple periods, offering a more fluid and realistic portrayal of performance. Using extensive economic data from 63 countries, the researchers found that widely used methods routinely misrepresent performance. They often fail to capture either the abrupt effects of crises or the gradual improvements that accumulate through steady investment or policy change. TEA, by contrast, highlights fluctuations that static tools cannot accommodate.
A distinctive feature of TEA is its combination of three analytical components. One examines how shocks unfold and influence organisations over time; another ranks organisations in relation to one another; and a third tests how external pressures shape their performance. Together, these tools produce a more comprehensive and resilient efficiency assessment. After running 400,000 tests, the research team showed that TEA offers far greater reliability than the methods currently relied upon by policymakers and industry leaders.
Dr Mehdi Toloo, co-author of the study and Reader in Business Analytics at the University of Surrey, underscored the consequences of relying on outdated evaluation models. He noted that critical decisions—whether involving government budgets, investment choices, or even hospital safety ratings—are often based on tools that freeze organisations at a single point in time. Such an approach fails to recognise that no institution operates in isolation from real-world shocks. By incorporating the effects of crises, whether financial, social or health-related, TEA provides an assessment that mirrors the complexities that organisations actually face. Misjudging efficiency, Dr Toloo warned, leads to flawed policy, misallocated resources and unjust comparisons.
The study also highlights the usefulness of TEA in situations involving minor, recurring disturbances. For example, gradual declines in technical performance or shifts in operational conditions can accumulate in ways that conventional models fail to register. Yet TEA handles these subtleties while remaining robust in the face of major disruptions. This versatility has significant implications. Governments could use TEA to evaluate how health systems recover from crises such as Covid-19, while businesses might rely on it to measure the long-term returns of adopting new technologies.
Dr Toloo added that TEA encourages more accountable decision-making by replacing one-off efficiency scores with dynamic assessments that change alongside real conditions. For those responsible for shaping policy or overseeing investment, TEA offers fairer comparisons, stronger risk evaluation and more unmistakable evidence to guide reform. In a world where shocks are increasingly common, such an approach promises to bring much-needed nuance and accuracy to organisational assessment.
In all, the research calls for modernising efficiency measurement. As institutions confront economic volatility, technological shifts, and unpredictable global events, tools that account for how performance evolves are becoming indispensable. TEA represents a significant step in that direction, offering a methodology better suited to today’s complex and fast-changing environment.
More information: Madjid Tavana et al, Time envelopment analysis: A new method for effectively incorporating time series in data envelopment analysis, Expert Systems with Applications. DOI: 10.1016/j.eswa.2025.127791
Journal information: Expert Systems with Applications Provided by University of Surrey