Naslov
Dual multipliers in DEA

Autori
Valter Boljunčić Luka Neralić

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, znanstveni

Izvornik
APMOd2006 International workshop and conference / Laureano F. Escudero, Antonio Alonso-Ayuso, Andres Ramos - Madrid : Universidad pontificia Camillas, 2006

Skup
APMOD 2006

Mjesto i datum
Madrid, Španjolska, 18.06.2006. - 21.06.2006

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
DEA; efficiency; sensitivity analysis; VRS

Sažetak
Dual Multipliers in DEA Data Envelopment Analysis (DEA) is a set of methods and models used to evaluate efficiency of a chosen Decision Making Unit (DMU) compared to set of DMUs. This is done using LP models where usually primal LP program is called envelopment model, while dual is called multiplier model. The goal is to obtain such multipliers (i.e. weights for inputs and outputs) so that weighted sum of outputs over weighted sum of inputs for the DMU under observation is greater or equal then the same sum for each of remaining DMUs.. Of interest are such multipliers that, for example among other things, even after certain decrease in outputs and increase in inputs of DMU under observation (what leads to decrease of its efficiency) have property that DMU still proves efficient using these multipliers. Such dual multipliers can be obtained using primal-dual relationship of applied LP model, exploiting data from the obtained Simplex tableau, and then, performing parametric programming applied to RHS of applied LP model, we will obtain different set of multipliers with the above characteristics.

Izvorni jezik
Engleski

Znanstvena područja
Matematika

POVEZANOST RADA