Investigating the effects of conveyor speed and product orientation on the performance of wireless RFID system in production line using factorial design

The implementation of radio frequency identification (RFID) in the manufacturing industries specifically the production lines has enabled the creation of fully automated monitoring and tracking system. An investigation on the effects of various speeds and placements of product on a conveyor using statistical analysis is introduced in this paper. The experiment was conducted at the Auto-ID Laboratory (AIDL) at USM by setting up the RFID system on a conveyor set to determine whether the various speeds and placements of an RFID tag that was attached on the products would affect the performance of the proposed RFID system. The Design of Experiments (DOE) method known as two-factor factorial design was performed to find out the factors that influenced readability of an RFID tag detection. The Minitab statistical software was used to analyse the experimental results using interaction plots and two-way analysis of variance (ANOVA) procedure. The two factors investigated were conveyor speed at three different levels 0.1143 m/s, 0.1476 m/s and 0.179 m/s and orientation of the RFID-tagged product on the conveyor set for six different orientations. The statistical analysis showed that conveyor speed had no significant effect on the RFID system as its P-value was 0.369 which was greater than 0.05 level of significance. Product orientation had a significant effect with a P-value of close to 0.00. Therefore the orientation of an RFID-tagged product plays an important role in an RFID system. The best orientation to place the RFID-tagged product on a conveyor is by having its RFID tag directly facing an antenna. The latter was determined by looking at the mean tag detection percentage for the six orientations. The highest percentage of 95% was obtained when the tag directly faced the antenna.