University of Birmingham > Talks@bham > Optimisation and Numerical Analysis Seminars > Uncertain data envelopment analysis with box uncertainty

## Uncertain data envelopment analysis with box uncertaintyAdd to your list(s) Download to your calendar using vCal - Emma Stubington (Lancaster University)
- Wednesday 13 November 2019, 12:00-13:00
- Strathcona, SR5.
If you have a question about this talk, please contact Sergey Sergeev. Data envelopment analysis (DEA) is a nonparametric, data driven technique to perform relative performance analysis among a group of comparable “decision-making units” (DMUs). Eﬃciency is assessed by comparing input and output data for each DMU via linear programming. Traditionally in DEA the data are considered to be exact. However, it is likely that the values for the variables used in the analysis are imprecise. To account for the inherent uncertainty present in many real-world data instances we assume that the variables are in fact realisations from a range, called an uncertainty set and consider the uncertain DEA problem (uDEA). We wish to determine the minimum amount of uncertainty required for an ineﬃcient DMU to be evaluated as eﬃcient. That is, we leverage the uncertainty present in the data such that ineﬃcient DMUs may improve. We develop the uDEA problem for the case of box uncertainty and show that for small problems this can be solved exactly. We explore the relationship between the amount of uncertainty required for an ineﬃcient DMU to be deemed eﬃcient and how this eﬃciency is achieved. We determine the minimum amount of uncertainty required for a DMU to be deemed eﬃcient. For small problems, this value can be easily calculated but for larger instances, this becomes computationally intensive as it requires the facets of the Production Possibility Set to be known. This study of uncertainty is motivated by the inherently uncertain nature of the radiotherapy treatment planning process. This talk is part of the Optimisation and Numerical Analysis Seminars series. ## This talk is included in these lists:Note that ex-directory lists are not shown. |
## Other listsCombinatorics and Probability Seminar Dinner Table Terrorism - Achieving Food Security Mathematics Colloquium## Other talksSchool Seminar From 2nd to 3rd generation GW detectors Machine Learning and Dynamical Systems meet in Reproducing Kernel Hilbert Spaces The imprint of their explosions: Using supernova remnants to understand stellar death |