arXiv:2606.07253v1 Announce Type: new Abstract: Traditional TOPSIS derives its reference points -- the Positive Ideal Solution ($PIS$) and Negative Ideal Solution ($NIS$) -- from the observed alternative set, making rankings susceptible to misalignment with decision-maker (DM) requirements, sensitivity to outlier performances, and rank reversal. This paper proposes TOPSIS-RAD, which addresses these issues by incorporating two arrays of DM-defined reference levels. Vetoed Performance Levels ($VPL$) exclude non-viable alternatives before normalisation, preventing them from distorting the ranking
This paper addresses known limitations in traditional multi-criteria decision-making (MCDM) methods, particularly as AI systems increasingly need robust and context-aware ranking mechanisms.
Improved decision-making algorithms, especially those that align with explicit human preferences and address ranking biases, are crucial for reliable AI agents and advanced analytical systems.
Decision-making processes in AI and other fields can become more robust and less susceptible to data outliers and misaligned reference points, leading to more human-centric outcomes.
- · AI agents developers
- · Decision support system providers
- · Organizations using MCDM for strategic planning
- · Systems heavily reliant on traditional TOPSIS for critical decisions
More accurate and stable rankings in various AI and analytical applications due to improved reference point definition.
Reduced 'rank reversal' issues and better alignment of automated decisions with human intent, increasing trust in AI systems.
Potential for broader adoption of sophisticated decision-making algorithms in AI, leading to more nuanced and context-aware autonomous systems.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.AI