Hybrid optimization technique for RFID network planning to improve warehouse management efficiency

Elewe, Adel Muhsin (2017) Hybrid optimization technique for RFID network planning to improve warehouse management efficiency. PhD thesis, Universiti Tun Hussein Onn Malaysia.

[img]
Preview
PDF
10Mb

Abstract

RFID Network Planning (RNP) strategy based on the functional parameters is often deployed in large scale applications to efficiently track assets and can lead to significant revenue gain. The placement of RFID readers to support warehouse design is often done on a trial and error basis which is time consuming and results in less than optimal coverage. For this reason, this research has developed a RFID network planning model that can improve warehouse management. The mathematical model of the RFID network planning is concerned with two major issues. The first one was to specify the reader parameters. The parameters of the RFID network planning (RNP) problem specified in each optimization technique were adjusted to improve the quality of the solutions. The second issue was to specify the objective functions of RNP problems. The fundamental gap highlighted in this research was the effect of network topology, which represents the most important factor in hard optimization problems of network planning. The present methodology correlated the RFID Network Planning with topology network design. Optimizing the network topology design can be formulated as set of functions employed with a Monte Carlo Algorithm in order to optimize the warehouse management. The main methodology process was by integrating the RFID multi-objective network planning with the network topology design to improve the capability of reader distribution. The sequence of operation started with Monte Carlo simulation (MCS) which was used to generate tag placements based on network topology design modules as a method to evaluate the deterministic indicators in NP-hard problems. The generated data are utilized as an input representation to apply into firefly algorithm based on Density-Based Algorithm (DBSCAN) to find the optimal network solution. The current work has produced much superior results for large scale and multi-variance facility shapes. The results show the effectiveness of the method in L-Shape RNP and standard benchmark in addition to two types of standard warehouse design. The optimization results indicate reduction of power consumption by 31% due to the reduction the number of readers and propagation range. It also indicates the high reliability of this method to work with complex tags distribution in large scale area.

Item Type:Thesis (PhD)
Subjects:T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK6540-6571 Radio
ID Code:10004
Deposited By:Mr. Mohammad Shaifulrip Ithnin
Deposited On:28 May 2018 14:46
Last Modified:28 May 2018 14:46

Repository Staff Only: item control page