Hybrid Technique for Arrival Rate Extraction and Size Reduction using Singular Spectrum Analysis and Fourier Series

摘要

Service providers aim at having low pressure on their traffic systems and this positively influences the provided services; this has made the Arrival Process as a very important factor of the queuing structure. Examples of arrivals comprise, arrival of the costumers to a supermarket and passengers to a train station to mention a few. This paper suggests extracting the arrival rate using singular spectrum analysis (SSA), SSA is a popular method in time series analysis. However, it was not used before in extracting the arrival rate. The novel contribution of this paper is introducing a combination of SSA and Fourier Series Reduction for an effective size reduction for arrival rate. The hybrid technique was applied to a real life example and compared with other algorithms. The experimental results show the superiority of SSA in extracting the arrival rate accompanied with a dramatic yet efficient size reduction when followed by a Fourier Series Reduction.