A software application for the selection of temperature measuring sensors using the analytic hierarchy process

This study presents an analytic hierarchy process (AHP) method to objectively select the best temperature sensor from among different alternative sensors in a certain industrial application. The underlying decision method based on AHP methodology, ranks temperature sensors with different features with a score resulting from the synthesis of relative preferences of each alternative with respect to the others at different levels considering independent evaluation criteria and sub-criteria. At each level, relative preferences of each candidate alternative with respect to the upper immediate level are calculated from pairwise comparisons among the candidate alternative sensors with respect to a selected application. Pair-wise comparison matrices are compiled based on views of experts in this field. Seven alternative sensors were considered: the thermocouple, the thermister, the resistance temperature detector (RTD), the bimetallic strip thermometer, the mercury-in-glass thermometer, the optical disappearing filament pyrometer, and the liquid crystal display semi conductor thermometer (LCD). Three industrial applications were also considered: Automotives, Chemical Processes, and Heating, Ventilating and Air Conditioning. A case study is conducted which involves selecting the best sensor for an automotive catalytic converter. The thermocouple is found to be the most preferred sensor for this application with the largest score of 0.37849, the second ranked sensor is the RTD with a score of 0.34589, and the least preferred sensor is the thermister with a score of 0.27560. To test the robustness of the proposed work, a sensitivity analysis was conducted in which variations in the relative preferences of the alternative sensors against sub-criteria and criteria were employed.