TY - CHAP
T1 - Efficiency decomposition in network data envelopment analysis
AU - Kao, Chiang
N1 - Publisher Copyright:
© Springer Science+Business Media New York 2014.
PY - 2014
Y1 - 2014
N2 - In order to measure the efficiency of systems composed of several processes more appropriately, various network data envelopment analysis (DEA) models have been developed. One type of the model, which is able to calculate the system and process efficiencies at the same time, is relational. This paper discusses the relationship between the system and process efficiencies measured from this model, and derives five properties. The first is general to all types of network structure, which states that the efficiency slack of the system is the sum of those of the component processes. This implies that a system is efficient if and only if all its component processes are. The second to fourth correspond to three types of structure, series, parallel, and dynamic. The last states that any unstructured system can be transformed into a series of parallel structures for efficiency decomposition. Numerical examples are used to help explain the idea of each type of decomposition. Efficiency decomposition enables decision makers to identify the processes that cause the inefficiency of a system, and thus to make effective changes to it.
AB - In order to measure the efficiency of systems composed of several processes more appropriately, various network data envelopment analysis (DEA) models have been developed. One type of the model, which is able to calculate the system and process efficiencies at the same time, is relational. This paper discusses the relationship between the system and process efficiencies measured from this model, and derives five properties. The first is general to all types of network structure, which states that the efficiency slack of the system is the sum of those of the component processes. This implies that a system is efficient if and only if all its component processes are. The second to fourth correspond to three types of structure, series, parallel, and dynamic. The last states that any unstructured system can be transformed into a series of parallel structures for efficiency decomposition. Numerical examples are used to help explain the idea of each type of decomposition. Efficiency decomposition enables decision makers to identify the processes that cause the inefficiency of a system, and thus to make effective changes to it.
UR - http://www.scopus.com/inward/record.url?scp=84955083101&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84955083101&partnerID=8YFLogxK
U2 - 10.1007/978-1-4899-8068-7_3
DO - 10.1007/978-1-4899-8068-7_3
M3 - Chapter
AN - SCOPUS:84955083101
T3 - International Series in Operations Research and Management Science
SP - 55
EP - 77
BT - International Series in Operations Research and Management Science
PB - Springer New York LLC
ER -