Utilization of comfortability data collected by HEMS for product recommendation

Home Energy Management System (HEMS) comprises many types of sensors that collect a variety of data. In this study, a product recommendation system based on the user's indoor comfort preference, which is a function of HEMS, is proposed as a secondary use of HEMS data. The proposed system utilizes the user's indoor comfort preference based on data collected by HEMS sensors, and recommends items that enhance their comfort in summer. The system has two main functions to facilitate the recommendation items: (i) collection of environmental and physical data to generate the predicted mean value (PMV), which is an indicator of indoor comfort, and (ii) identification of the appropriate preference for each user and provision of the collected information via e-mail and the webpage. The information of the product recommendation and instructions pertaining the effectiveness of the products for maintaining the indoor comfort are provided to the user. To test the system's applicability, we installed the system in three households, and then conducted the ABAB test, which shows recommends random items during period A and the selected items during period B. Results show that the willingness of buying the recommended products increased by approximately 34.2% when the user's indoor comfort preference was considered as compared to when indoor comfort preference was not considered.