An interactive method to multiresponse surface optimization based on pairwise comparisons
Dong‐Hee Lee, Kwang-Jae Kim, Murat Köksalan
Abstract
Dong‐Hee Lee, Kwang-Jae Kim, Murat Köksalan
Abstract
In Multi-Response Surface Optimization (MRSO), responses are often in conflict. To obtain a satisfactory compromise, the preference information of a Decision Maker (DM) on the trade-offs among the responses should be incorporated into the problem. Most existing methods employ preference parameters to incorporate the DM’s subjective judgment on the responses. The preference parameter values are specified in advance or adjusted in an interactive manner. However, it is often difficult to specify or adjust the preference parameter values that are representative of the DM’s preference structure without use of a systematic method. An interactive method for MRSO is developed in this article in which the DM provides preference information in the form of pairwise comparisons. The results of these comparisons are used to estimate the preference parameter values in an interactive manner. The required preference information is relevant and therefore easy for the DM to provide. The method is effective in that a highly satisfactory solution for the DM can be obtained through a few pairwise comparisons, regardless of the type of the DM’s utility function, in the problems solved in this work.
OpenAlex reports 31 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
In Multi-Response Surface Optimization (MRSO), responses are often in conflict. To obtain a satisfactory compromise, the preference information of a Decision Maker (DM) on the trade-offs among the responses should be incorporated into the problem. Most existing methods employ preference parameters to incorporate the DM’s subjective judgment on the responses. The preference parameter values are specified in advance or adjusted in an interactive manner. However, it is often difficult to specify or adjust the preference parameter values that are representative of the DM’s preference structure without use of a systematic method. An interactive method for MRSO is developed in this article in which the DM provides preference information in the form of pairwise comparisons. The results of these comparisons are used to estimate the preference parameter values in an interactive manner. The required preference information is relevant and therefore easy for the DM to provide. The method is effective in that a highly satisfactory solution for the DM can be obtained through a few pairwise comparisons, regardless of the type of the DM’s utility function, in the problems solved in this work.
Key concepts: Pairwise comparison, Preference, Decision maker, Revealed preference, Computer science, Mathematical optimization, Quality (philosophy), Function (biology)