Answers are grounded in this paper and cite their sources.
In real-world multi-attribute decision-making (MADM), understanding the inter-relationships and hierarchical structures among factors is paramount. Once established, the focus shifts to optimizing the goal attribute by addressing these factors effectively. Some practical decision domains, such as the healthcare system, education system, business sectors, etc. need continuous evaluation and modification of the optimal strategies to achieve the best results. This paper proposes a multistep self-regularized strategy for assigning effort to the factors in long-term continuous decision-making problems. In successive time steps, the strategy considers the higher-level factors for the effort allocation on a reducing-level basis. The strategy also uses the concept of feedback from upper-level factors to the lower-level factors in the underlying hierarchy. Two self-regularization parameters are used to trade-off between different heuristics associated with the factors for the effort assignment. The strategy is analyzed in a case study involving administrative factors in high schools that contribute to improving students' performance. A comparative analysis is performed between the proposed strategy and existing single-step strategies. The highest effort propagation achieved in the strategy is approximately 21.7143% when the feedback is used, whereas it is 14.7535% when the feedback is not considered. This underscores the importance of incorporating feedback in any continuous and long-term decision-making process.
No takes yet. Share an insight, caveat, or question.
Dhara et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: