TY - GEN
T1 - Show Me the Money
T2 - 2020 IEEE International Conference on RFID
AU - Wolbitsch, Matthias
AU - Hasler, Thomas
AU - Helic, Denis
AU - Walk, Simon
PY - 2020/9/28
Y1 - 2020/9/28
N2 - Over the course of recent years, Radio Frequency Identification (RFID) technology has been applied in several different business domains to improve a diverse array of practical applications. For example, handheld RFID readers combined with passive RFID tags are used to perform fast and accurate stocktakes for fashion retailers. However, while this approach enables efficient inventory management, automatic localization of RFID-tagged goods in stores is still an open problem. To tackle this problem, we equip fixtures (e.g., shelves, tables,..) with reference RFID tags and use data we collect during typical RFID-based stocktakes to map articles to fixtures. Knowing the location of goods within a store enables the implementation of several practical applications, such as automated Money Mapping (e.g., creating a heat map of sales across fixtures) or visual merchandising evaluations (e.g., monitoring sales of fixtures before, during, and after the implementation visual merchandising strategies). Specifically, we conduct (i) controlled lab experiments and (ii) a case-study in two fashion retail stores to evaluate our presented approaches for article-to-fixture predictions. The approaches are based on calculating distances between read event time series of article and reference tags using dynamic time warping, and clustering of read events using DBSCAN. We find that we can use read events collected during RFID-based stocktakes to assign articles to fixtures with an accuracy of more than 90% in several of our experiments. Hence, in this paper we present an exploratory venture into novel and practical RFID-based applications, beyond the scope of stock management.
AB - Over the course of recent years, Radio Frequency Identification (RFID) technology has been applied in several different business domains to improve a diverse array of practical applications. For example, handheld RFID readers combined with passive RFID tags are used to perform fast and accurate stocktakes for fashion retailers. However, while this approach enables efficient inventory management, automatic localization of RFID-tagged goods in stores is still an open problem. To tackle this problem, we equip fixtures (e.g., shelves, tables,..) with reference RFID tags and use data we collect during typical RFID-based stocktakes to map articles to fixtures. Knowing the location of goods within a store enables the implementation of several practical applications, such as automated Money Mapping (e.g., creating a heat map of sales across fixtures) or visual merchandising evaluations (e.g., monitoring sales of fixtures before, during, and after the implementation visual merchandising strategies). Specifically, we conduct (i) controlled lab experiments and (ii) a case-study in two fashion retail stores to evaluate our presented approaches for article-to-fixture predictions. The approaches are based on calculating distances between read event time series of article and reference tags using dynamic time warping, and clustering of read events using DBSCAN. We find that we can use read events collected during RFID-based stocktakes to assign articles to fixtures with an accuracy of more than 90% in several of our experiments. Hence, in this paper we present an exploratory venture into novel and practical RFID-based applications, beyond the scope of stock management.
KW - DBSCAN
KW - DTW
KW - Money Mapping
KW - RFID
UR - http://www.scopus.com/inward/record.url?scp=85097272485&partnerID=8YFLogxK
U2 - 10.1109/RFID49298.2020.9244903
DO - 10.1109/RFID49298.2020.9244903
M3 - Conference paper
AN - SCOPUS:85097272485
T3 - 2020 IEEE International Conference on RFID, RFID 2020
BT - 2020 IEEE International Conference on RFID, RFID 2020
PB - Institute of Electrical and Electronics Engineers
Y2 - 28 September 2020 through 16 October 2020
ER -