Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Improving preference disaggregation in multicriteria decision making: incorporating time series analysis and a multi-objective approach

Betania S. C. Campello, Sarah BenAmor, Leonardo T. Duarte, João Marcos Travassos Romano

TL;DR

The paper addresses the limitation of static preference learning in multicriteria decision making by introducing UTASTAR-T, a tensor-based extension that incorporates time-series descriptive measures (e.g., average and trend) of criteria. It combines this tensorial PDA with a multi-objective and Monte Carlo framework to generate multiple compatible preference models and study convergence toward representative solutions. Computational experiments on real data demonstrate how descriptive measures influence marginal value functions and reveal the importance of multi-objective analysis for robust decision support. The approach advances PDA by modeling temporality, providing richer insights for long-term decision contexts and offering practical guidance on robustness and convergence in MCDA.

Abstract

Preference disaggregation analysis (PDA) is a widely used approach in multicriteria decision analysis that aims to extract preferential information from holistic judgments provided by decision makers. This paper presents an original methodological framework for PDA that addresses two significant challenges in this field. Firstly, it considers the multidimensional structure of data to capture decision makers' preferences based on descriptive measures of the criteria time series, such as trend and average. This novel approach enables an understanding of decision makers' preferences in decision-making scenarios involving time series analysis, which is common in medium- to long-term impact decisions. Secondly, the paper addresses the robustness issue commonly encountered in PDA methods by proposing a multi-objective and Monte Carlo simulation approach. This approach enables the consideration of multiple preference models and provides a mechanism to converge towards the most likely preference model. The proposed method is evaluated using real data, demonstrating its effectiveness in capturing preferences based on criteria and time series descriptive measures. The multi-objective analysis highlights the generation of multiple solutions, and, under specific conditions, reveals the possibility of achieving convergence towards a single solution that represents the decision maker's preferences.

Improving preference disaggregation in multicriteria decision making: incorporating time series analysis and a multi-objective approach

TL;DR

The paper addresses the limitation of static preference learning in multicriteria decision making by introducing UTASTAR-T, a tensor-based extension that incorporates time-series descriptive measures (e.g., average and trend) of criteria. It combines this tensorial PDA with a multi-objective and Monte Carlo framework to generate multiple compatible preference models and study convergence toward representative solutions. Computational experiments on real data demonstrate how descriptive measures influence marginal value functions and reveal the importance of multi-objective analysis for robust decision support. The approach advances PDA by modeling temporality, providing richer insights for long-term decision contexts and offering practical guidance on robustness and convergence in MCDA.

Abstract

Preference disaggregation analysis (PDA) is a widely used approach in multicriteria decision analysis that aims to extract preferential information from holistic judgments provided by decision makers. This paper presents an original methodological framework for PDA that addresses two significant challenges in this field. Firstly, it considers the multidimensional structure of data to capture decision makers' preferences based on descriptive measures of the criteria time series, such as trend and average. This novel approach enables an understanding of decision makers' preferences in decision-making scenarios involving time series analysis, which is common in medium- to long-term impact decisions. Secondly, the paper addresses the robustness issue commonly encountered in PDA methods by proposing a multi-objective and Monte Carlo simulation approach. This approach enables the consideration of multiple preference models and provides a mechanism to converge towards the most likely preference model. The proposed method is evaluated using real data, demonstrating its effectiveness in capturing preferences based on criteria and time series descriptive measures. The multi-objective analysis highlights the generation of multiple solutions, and, under specific conditions, reveals the possibility of achieving convergence towards a single solution that represents the decision maker's preferences.
Paper Structure (11 sections, 2 theorems, 23 equations, 5 figures, 4 tables)

This paper contains 11 sections, 2 theorems, 23 equations, 5 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Temporal analysis in multicriteria decision making: enhancing preferential learning
  3. Preference disaggregation analysis and robustness issues in MCDA: an overview
  4. Background on classical UTA and UTASTAR, and multi-objective problems
  5. Classical UTA and UTASTAR: notation and linear program
  6. Robustness issue in UTA and UTASTAR
  7. Multi-objective problems
  8. UTASTAR-T with a multi-objective approach
  9. Multi-objective UTASTAR-T linear program
  10. Computational experiments using real data
  11. Conclusions

Key Result

Theorem 1

The solution of the weighting problem (eq:somaponderada)-(eq:somaponderada2) is weakly Pareto optimal.

Figures (5)

  • Figure 1: Tensorial structure representing the time series and descriptive measures of the criteria over time.
  • Figure 2: Normalized value functions of descriptive measure 1: average.
  • Figure 3: Normalized value functions of descriptive measure 2: slope coefficient.
  • Figure 4: Weighted average of value function values for descriptive measure 1: average.
  • Figure 5: Weighted average of value function values for descriptive measure 2: slope coefficient.

Theorems & Definitions (4)

  • Definition 1
  • Definition 2
  • Theorem 1
  • Theorem 2