Skip to contentScholarGate
LibraryBookshelfDeskReview StudioAssistant
Sign in
On this page
IntuitionHow it worksWhen to use itStrengths & limitationsCommon pitfallsApplicationsFrequently asked🔒 Read the full methodSourcesRelated methods
Cite this pageSpotted an issue on this page? Report or suggest a fix →
Home›Efficiency Analysis›Network Data Envelopment Analysis (Network DEA)
Regression modelEfficiency analysis

Network Data Envelopment Analysis (Network DEA)

Network Data Envelopment Analysis · Also known as: Network Data Envelopment Analysis, Network Efficiency Analysis, Multi-Stage DEA, Ağ Veri Zarflama Analizi

Network Data Envelopment Analysis (Network DEA) is a nonparametric efficiency measurement framework introduced by Färe and Grosskopf (2000) that extends classical DEA to multi-stage or multi-division production processes. Rather than treating a decision-making unit as a black box, it explicitly models the internal structure — the divisions and the intermediate products that flow between them — enabling stage-level and overall efficiency scores to be estimated simultaneously within a single coherent model.

ScholarGate
  1. Regression model
  2. v1
  3. 1 Sources
  4. PUBLISHED
Cite this page →
Tools & resources
Download slides
Learn & explore

Read the full method

Members only

Sign in with a free account to read this section.

Sign in

Method map

The neighbourhood of related methods — select a node to explore.

Network DEA
Bootstrap DEAMalmquist Productivity I…By-Production Technology…Super-Efficiency DEAWindow DEA

When to use it

Network DEA is appropriate when the production process has identifiable internal stages connected by measurable intermediate products, such as banking (deposit-gathering then lending), healthcare (diagnosis then treatment), or supply chains (procurement, production, distribution). Key assumptions include: intermediate flows are observable and quantifiable, the overall technology satisfies convexity or constant returns to scale, and a sufficient number of DMUs exist relative to the total number of inputs, outputs, and intermediate variables. It is less suitable when internal structure is unobservable or when intermediate products cannot be measured. Alternatives include dynamic DEA for temporal links or standard two-stage regression-based DEA for post-hoc analysis.

Strengths & limitations

Strengths
  • Simultaneously estimates stage-level and overall efficiency, revealing where within the process inefficiency originates.
  • Maintains full nonparametric flexibility — no functional form for the production technology is required.
  • Prevents inconsistent results that arise when stages are evaluated independently with mismatched intermediate flow assumptions.
  • Directly applicable to a wide range of multi-division settings including banking, healthcare, energy, and supply chain management.
Limitations
  • Requires detailed data on intermediate products, which may be unavailable or difficult to measure in practice.
  • Model complexity and the number of constraints grow rapidly with the number of stages and intermediate variables, potentially straining feasibility.
  • The choice of weights for aggregating divisional scores into a system score is somewhat arbitrary and can influence conclusions.
  • Like all DEA variants, scores are sensitive to outliers and to the selection of inputs, outputs, and intermediate variables.

Frequently asked

How does Network DEA differ from running separate DEA models for each stage?

Running separate DEA models for each stage treats intermediate products independently in each program and can produce inconsistent results — for example, one stage's optimal plan may require more intermediate output than the next stage's optimal plan allows. Network DEA enforces linking constraints that couple the stages into a single coherent system, ensuring that intermediate flows are consistent across all divisional programs.

Can Network DEA handle more than two stages?

Yes. The framework generalises directly to D sequential divisions and also accommodates parallel or mixed network structures in which some divisions operate simultaneously rather than sequentially. Each additional stage introduces its own set of divisional constraints and intermediate linking constraints, increasing model size but not changing the fundamental approach.

What minimum sample size is recommended for Network DEA?

A widely cited rule of thumb in standard DEA is that the number of DMUs should be at least three times the total number of inputs plus outputs. For Network DEA this requirement extends to include intermediate variables in the count, so the effective minimum sample size rises with the complexity of the network structure. Violating this guideline risks discrimination failure and inflated efficiency scores.

Sources

  1. Färe, R., & Grosskopf, S. (2000). Network DEA. Socio-Economic Planning Sciences, 34(1), 35–49. DOI: 10.1016/S0038-0121(99)00012-9 ↗

How to cite this page

ScholarGate. (2026, June 2). Network Data Envelopment Analysis. ScholarGate. https://scholargate.app/en/efficiency-analysis/network-dea

Related methods

Bootstrap DEAMalmquist Productivity Index

Which method?

Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.

  • Bootstrap DEAEfficiency Analysis↔ compare
  • Malmquist Productivity IndexEfficiency Analysis↔ compare
Compare side by side →

Referenced by

Bootstrap DEABy-Production Technology DEASuper-Efficiency DEAWindow DEA

Similar methods

DEA-NETWORKData Envelopment Analysis (Productivity)Super-Efficiency DEADEA-NETWORK-SBMData Envelopment Analysis of Firm Strategic EfficiencyDEA Hospital EfficiencyDEA-DYNAMIC-NETWORKBy-Production Technology DEA

Related reference concepts

Input–Output Tables and AnalysisNetwork AnalysisHospital Operations and ManagementNetwork Formation and Analysis: TheoryInput–Output ModelsStructure, Process, and Outcome Measures

Spotted an issue on this page? Report or suggest a fix →

ScholarGate — Network DEA (Network Data Envelopment Analysis). Retrieved 2026-07-21 from https://scholargate.app/en/efficiency-analysis/network-dea · Dataset: https://doi.org/10.5281/zenodo.20539026
Quick facts
Originator
Färe & Grosskopf
Year
2000
Type
Multi-stage nonparametric efficiency model
Subfamily
Efficiency analysis
Orientation
Input/output/both
Returns To Scale
CRS or VRS
Related methods
Bootstrap DEAMalmquist Productivity Index
ScholarGate

A content-first reference library for research methods — what each one is, how it works, and where it comes from.

Open data (CC-BY)

Explore

  • Library
  • Search the library…
  • Browse by field
  • Fields
  • Journey
  • Compare
  • Which method?

Reference

  • Subjects
  • Atlas
  • Glossary
  • Methodology
  • Philosophy

Your tools

  • Bookshelf
  • Desk
  • Chat

Company

  • About
  • Pricing
  • Contact
  • Suggest a method

Entries are compiled from published sources for reference. Verifying the accuracy and suitability of any information for your own use remains your responsibility.

© 2026 ScholarGate · A research-method reference library
  • Privacy
  • Cookies
  • Terms
  • Delete account