In the era of digital transformation, organisations must adopt advanced analytical tools to remain competitive, resilient, and sustainable. Data Envelopment Analysis (DEA), long recognised as a robust framework for assessing efficiency and benchmarking performance, is now increasingly integrated with Artificial Intelligence (AI) and Machine Learning (ML). Together, these methods form a powerful toolkit for uncovering hidden patterns, improving predictive accuracy, and translating complex data into actionable strategies—particularly for measuring performance and enhancing decision-making.
This workshop will explore how combining DEA with neural networks, decision trees, classification techniques, regression models, and support vector machines creates new opportunities for optimising decision-making and performance evaluation. Neural networks enhance efficiency prediction, decision trees and classification methods provide meaningful data segmentation, and support vector machines strengthen the classification of decision-making units based on their performance levels.
Through case studies and empirical illustrations, attendees will gain practical insights into how DEA, when augmented by AI and ML, empowers decision-makers with transparent, interpretable, and accountable approaches—ultimately shaping more adaptive and future-ready organisations in an increasingly data-driven world.
Workshop Agenda
Focus: Foundations of performance measurement and the role of analytical models.
-
Overview of Performance Measurement in the Digital Era
- Why performance measurement matters for competitiveness, resilience, and sustainability.
- Key challenges organisations face in evaluating efficiency and benchmarking.
-
Introduction to Data Envelopment Analysis (DEA)
- DEA as a robust framework for measuring efficiency.
- Applications in banking, healthcare, energy, supply chain, and sustainability.
- Strengths and limitations of DEA in real-world contexts.
-
Hands-On Demo
- Performance Improvement Management (PIM) software — live walkthrough of input/output modelling, efficiency scoring, and benchmarking.
- Quick example with a small dataset to show how performance evaluation works in practice.
Learning Outcomes
- Understand the principles of DEA and its role in performance benchmarking.
- Recognise how analytical tools support evidence-based management.
Focus: Introducing AI and ML tools relevant to performance measurement.
-
Overview of Machine Learning in Business Analytics
- Supervised vs. unsupervised learning.
- Linking predictive analytics with performance management.
-
Key Algorithms and Their Roles
- Neural Networks: Enhancing predictive accuracy of efficiency scores.
- Decision Trees & Classification: Segmenting data and revealing decision rules.
- Support Vector Machines: Strengthening classification of decision-making units (DMUs).
- Regression Models: Forecasting performance indicators.
-
Practical Illustration
- Example of how ML algorithms can improve interpretability and predictions in performance analysis.
Learning Outcomes
- Gain a clear understanding of core ML algorithms.
- See how ML complements traditional performance evaluation models.
Focus: Combining DEA and AI/ML for predictive performance management and real-world applications.
-
Integration Framework
- Using DEA outputs (efficiency scores, slacks, targets) as input features for AI models.
- Creating hybrid models: DEA for benchmarking + AI for prediction.
-
AI for Predicting Efficiency
- Neural Networks trained on DEA efficiency results to forecast performance of new or unseen units.
- Support Vector Machines and Decision Trees for classifying units as efficient/inefficient.
- Regression models for predicting efficiency trends over time.
-
Case Studies & Applications
- Predicting hospital efficiency with DEA–NN models to guide resource planning.
- Forecasting bank efficiency with DEA–SVM for risk-adjusted decision-making.
-
Interactive Demo (Optional, time permitting)
- Use DEA scores from Session 1 as training data for a simple ML model (e.g., regression or classification).
-
Managerial Insights
- Predictive models help anticipate efficiency outcomes before decisions are implemented.
- Enables proactive planning instead of reactive benchmarking.
-
Future Directions
- Building AI-enabled decision support systems using DEA-based predictions.
- Ensuring transparency, interpretability, and accountability of hybrid DEA-AI models.
Learning Outcomes
- Learn how DEA results can serve as training data for AI.
- Understand how AI extends DEA from descriptive benchmarking to predictive decision-making.
- Gain insights into practical applications of efficiency prediction in real industries.
Workshop Facilitator
Prof. Ali Emrouznejad
Professor and Chair in Business Analytics; Director, Centre for Business Analytics in Practice
Surrey Business School, University of Surrey, UK
Professor Ali Emrouznejad is Professor and Chair in Business Analytics at Surrey Business School, UK, and Director of the Centre for Business Analytics in Practice. A leading expert in performance measurement, efficiency and productivity analysis, and the integration of Artificial Intelligence (AI) with Data Envelopment Analysis (DEA), his research applies advanced analytics and AI-driven tools to optimise decision-making and organisational performance. He has been recognised by Stanford University as one of the world’s top 2% most influential scientists, and with an h-index exceeding 80, he was named among the World Top 100 Business & Management / Business Administration Scientists 2025. With over 250 publications and extensive experience in national and international projects, his work provides actionable insights for organisations aiming to leverage DEA, AI, and Machine Learning for smarter, data-driven strategies—making him uniquely positioned to guide this workshop on AI and business analytics for performance measurement and decision-making.