Generative AI and Data Envelopment Analysis for Performance Measurement and Public Value

Call for Papers – Generative AI and Data Envelopment Analysis for Performance Measurement and Public Value

Journal: Socio‑Economic Planning Sciences
Publisher: Elsevier
Journal metrics: Impact Factor 5.4, CiteScore 10.3

Submission deadline: 1 September 2026

This special issue explores the two‑way relationship between generative AI (GenAI) and Data Envelopment Analysis (DEA) to advance rigorous performance measurement and public‑value evaluation in the age of AI. It invites methodological and empirical research showing how DEA can assess efficiency and productivity in GenAI‑enabled systems, and conversely how GenAI can enhance DEA design, data generation, and performance‑measurement frameworks in socio‑economic planning and policy contexts.

Why this issue matters

• Generative AI (large language models, image generators, and related tools) is rapidly diffusing in public administration, education, health, finance, and supply chains, yet its efficiency, reliability, and alignment with public‑value objectives remain uneven and poorly structured analytically.

• Data Envelopment Analysis (DEA) offers a nonparametric, multi‑input/multi‑output framework to benchmark decision‑making units (DMUs), incorporate quality, equity, and risk dimensions, and track productivity change over time, making it ideally suited to evaluate AI‑augmented services.

• This SI seeks to bridge operations research, AI, and socio‑economic policy, providing rigorous tools for planners and policymakers to assess when, where, and how GenAI creates genuine efficiency gains and public value.

Core focus and relationship between GenAI and DEA

The special issue explicitly follows a two‑way logic:

1. DEA applied to GenAI‑enabled systems

   • Measuring efficiency and productivity of GenAI‑augmented workflows (e.g., public‑service delivery, healthcare, education, customs, tax administration, municipal operations).

   • Integrating accuracy, safety, fairness, and user satisfaction as part of efficiency and eco‑efficiency benchmarks.

2. GenAI applied to extend or improve DEA

   • Using GenAI‑generated synthetic data, variable‑selection aids, or model‑suggestion tools to support DEA design, frontier estimation, and interpretation.

   • Addressing data‑quality issues (hallucinations, leakage, missing data) via GenAI‑assisted bias correction and robustness checks.

Indicative topics and research questions

Submissions must be methodologically grounded in DEA or rigorously apply DEA to GenAI‑augmented processes. Conceptual or narrative AI‑only papers are excluded.

• DEA models comparing GenAI‑enabled vs conventional workflows

  • Education, health, public‑sector administration, finance, and supply chains.

  • Benchmarking DMUs before and after GenAI adoption via Malmquist or panel‑DEA productivity analyses.

• Innovative integration of GenAI into DEA

  • Using GenAI for feature selection, model specification, or frontier‑estimation support in DEA.

  • GenAI‑assisted synthetic‑cohort generation with bias‑adjustment to validate DEA frameworks.

• Multi‑dimensional and equity‑sensitive DEA

  • Incorporating accuracy, reliability, safety, fairness, and user satisfaction as outputs in DEA.

  • Distributional‑constrained DEA to assess accessibility, inclusion, and equity (e.g., for diverse learners, vulnerable populations).

• Eco‑efficiency, footprint, and latency

  • By‑production DEA models that capture energy use, carbon footprint, compute costs, and latency alongside service quality.

  • Assessing the marginal contribution of GenAI to eco‑efficiency and operational‑cost savings.

• Decomposition and bottleneck analysis in AI‑augmented pipelines

  • Breaking down data curation → model use → service delivery to identify bottlenecks and spillovers.

  • Evaluating human–GenAI collaboration (fully human‑driven vs assisted teams) within DEA frameworks.

• Policy, resource allocation, and governance

  • Using DEA results to inform funding, capacity planning, and procurement decisions for GenAI use in universities, municipalities, and public agencies.

  • Integrating compliance, explainability, and auditability into DEA models for model‑lifecycle stewardship.

Guest editors

• Prof. Vincent Charles, Queen’s Business School, Queen’s University Belfast, UK

• Prof. Ali Emrouznejad, Surrey Business School, University of Surrey, UK

• Dr. Marios Kremantzis, University of Bristol Business School, UK

• Prof. Tatiana Gherman, Northampton Business School, University of Northampton, UK

Submission details

• Submission portal: Elsevier Editorial Manager for Socio‑Economic Planning Sciences:
  https://www.editorialmanager.com/seps

• When submitting, select article type: “VSI: GenAI and DEA”.

• Publication timeline:

  • Full‑paper deadline: 1 September 2026

  • Review notifications: 1 December 2026

  • Revision deadline: 1 March 2027

• Submissions must comply with the journal’s Guide for Authors and demonstrate clear DEA modelling and empirical or methodological depth.

• Manuscripts lacking an explicit DEA component or sufficient analytical grounding will not be considered.

Papers should explicitly show how the integration of generative AI and DEA yields new insights into efficiency, productivity, or public‑value measurement in socio‑economic planning domains.

ServiceSetu Academics — Premier Platform for Academic Opportunities & Research Collaboration

Visit official website of the publisher