Structured Pareto Front Representation

21498 8 113638 2 342910 document 113638 spfrZenodo.pdf application/pdf c2428484af2915b86b5fe41511aa952d MD5 2363221 2026-04-23 10:48:54 https://pure.iiasa.ac.at/id/eprint/21498/1/spfrZenodo.pdf 21498 1 1 application/pdf text en public cc_by_4

spfrZenodo.pdf

published 113639 1 342917 document 113639 indexcodes.txt text/plain f753b3e2c18ec05e8a36a868dffd5591 MD5 14200 2026-04-23 10:58:51 https://pure.iiasa.ac.at/id/eprint/21498/2/indexcodes.txt 21498 2 2 text/plain other Generate index codes conversion from text to indexcodes en public

indexcodes.txt

lightbox.jpg

preview.jpg

medium.jpg

small.jpg

http://eprints.org/relation/isVersionOf https://pure.iiasa.ac.at/id/document/113638 http://eprints.org/relation/isVolatileVersionOf https://pure.iiasa.ac.at/id/document/113638 http://eprints.org/relation/issmallThumbnailVersionOf https://pure.iiasa.ac.at/id/document/113638 archive 353 disk0/00/02/14/98 2026-04-23 10:50:42 2026-04-23 10:50:42 2026-04-23 10:50:42 article show 1 Makowski M. 1254 0000-0002-6107-0972 Granat J. Nahorski Z. Shekhovtsov A. Ren H. Zhao J. Structured Pareto Front Representation inpress ECE IACC S3 The paper presents a novel methodology for autonomous generation of the Pareto-Front Representation (PFR) of Linear Programming (LP) models. Following the Structured Modeling principles, the developed approach supports multiple-criteria analysis of independently developed diverse LP models. The analysis is done by seamless linkage with the dedicated implementation of reusable Multiple Objective Programming (MOP) model, which represents the developed method of PFR's generation. The MOP is a small LP model, therefore the linked models are optimized by an LP solver. The methodology enables autonomous (i.e., parametrization-free) generation of a sequence of single-objective LP optimization tasks, each providing a Pareto-efficient solution that improves the PFR distribution in terms of the distances between neighbor Pareto solutions. Furthermore, the method's recent enhancement by the structured PFR generation has two key advantages: it dramatically decreases the computation time, and it substantially improves the distribution of the PFR's elements. Moreover, the method properly and efficiently computes the extreme points of the~PF also when optimization of a~criterion has non-unique solution. The approach supports objective Multiple Criteria Model Analysis (MCMA), i.e., equitable treatment of all criteria in the whole space of Pareto solutions. The paper provides examples of research projects in various fields of science, which required effective support for generation of preference-free PFR. Such analysis is also helpful for preference-guided MCMA because the provided PFR clusters are a good starting point for exploration of diverse Regions of Interest (ROI) based on diverse simultaneously reachable goals for conflicting criteria. 2026-04-23 published 10.5281/zenodo.19704566 193 public Zenodo FALSE info:eu-repo/semantics/article