Glossary · approach
Data Envelopment Analysis
Non-parametric LP-based frontier-efficiency method that measures the relative technical efficiency of similar units (DMUs) operating with multiple inputs and multiple outputs.
DEACCR ModelBCC Model
Data Envelopment Analysis (DEA) is the non-parametric LP-based frontier-efficiency method, founded by Charnes-Cooper-Rhodes (1978), that measures the relative technical efficiency of a set of Decision Making Units (DMUs) — typically organisational units such as bank branches, hospitals, schools, farms — that consume multiple inputs to produce multiple outputs. Output: a 0-1 efficiency score per DMU plus a set of efficient peer-benchmarks for the inefficient ones. Two core variants: **CCR** (Charnes-Cooper-Rhodes 1978) constant returns to scale; **BCC** (Banker-Charnes-Cooper 1984) variable returns to scale, decomposing technical + scale efficiency. Modern extensions: SBM (Tone 2001 — slack-based, non-radial), super-efficiency (Andersen-Petersen 1993 — ranking among efficient ones), Malmquist productivity index (time-dependent dynamics), Tobit regression / Simar-Wilson (2007) bootstrap (second-stage determinant analysis). Canonical sources: Charnes-Cooper-Rhodes (1978) foundational + Cooper-Seiford-Tone (2007) textbook + Cook-Seiford (2009) 30-year summary.
Örnek
A bank runs DEA-CCR + BCC in parallel on its 200 branches with 5 inputs (headcount, space, computers, rent, IT) and 4 outputs (deposits, loans, customers, revenue); 60 branches come out fully efficient (score 1), 140 are inefficient; a peer-benchmark report is produced for the inefficient branches (e.g. branch X at 0.74 efficiency, peers Y and Z); slack analysis gives improvement targets.