Power generation expansion planning in an autonomous island system using multi-objective programming: The case of Milos island

Kourempele, M; Mavrotas, G; Geronikolou, L; Rozakis, S

dc.contributor.author	Kourempele, M	en
dc.contributor.author	Mavrotas, G	en
dc.contributor.author	Geronikolou, L	en
dc.contributor.author	Rozakis, S	en
dc.date.accessioned	2014-03-01T01:34:20Z
dc.date.available	2014-03-01T01:34:20Z
dc.date.issued	2010	en
dc.identifier.issn	11092858	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/20698
dc.subject	Augmented ε-constraint method	en
dc.subject	Energy regional planning	en
dc.subject	Fuzzy demand	en
dc.subject	Multi-objective mixed integer linear programming	en
dc.subject	Reference point	en
dc.title	Power generation expansion planning in an autonomous island system using multi-objective programming: The case of Milos island	en
heal.type	journalArticle	en
heal.identifier.primary	10.1007/s12351-009-0063-5	en
heal.identifier.secondary	http://dx.doi.org/10.1007/s12351-009-0063-5	en
heal.publicationDate	2010	en
heal.abstract	This paper presents the application of multiple objective linear programming for power generation expansion planning on Milos island. The model considers those economic and environmental objectives which typically conflict (cost minimization vs. CO2 emission reduction maximization) subject to a number of constraints. Due to uncertainties in future power demand, the latter is handled as a fuzzy parameter. Fuzziness is dealt with by the addition of a third objective function, the maximization of the degree of demand satisfaction. The MOLP model developed is solved in two ways. First, the use of the augmented ε-constraint method which produces the trade offs between cost and CO2 for different values of the degree of demand satisfaction, and second the reference point framework, a generation and an interactive method, respectively. In the second approach decision makers set their aspiration levels concerning the different criteria, converging after a number of iterations in a compromise solution, whereas in the augmented ε-constraint generation method the decision makers have to choose their preferred solution between all the efficient points depicted in the trade-offs. A comparative analysis of the above methods concludes the paper, highlighting advantages and shortcomings. © Springer-Verlag 2009.	en
heal.journalName	Operational Research	en
dc.identifier.doi	10.1007/s12351-009-0063-5	en
dc.identifier.volume	10	en
dc.identifier.issue	1	en
dc.identifier.spage	109	en
dc.identifier.epage	132	en