Limited data-driven router bandwidth configuration for cyber physical internet

Abstract

In the Cyber Physical Internet (CPI), router bandwidth indicates the router’s capacity to handle and transmit data. Proper configuration of router bandwidth is essential to satisfy throughput demands, making the evaluation of router throughput a vital aspect of bandwidth configuration. By introducing a novel bandwidth configuration method, we address the supply and demand balance in the logistics field from a fresh perspective. This paper explores the configuration of CPI router bandwidth in scenarios where data accessibility is limited. The study utilizes the Cyber Physical Internet-Autoregressive Integrated Moving Average-Kalman (CPI-AK) hybrid evaluation method to assess fluctuations in router throughput. Additionally, the CPI-Router Bandwidth Configuration (CPI-RBC) model is employed to implement the bandwidth configuration. Finally, freight data from Huyue station, spanning from January 2020 to December 2021, were analyzed and processed. The proposed bandwidth configuration methodology was validated and evaluated through a real case study of this rail-water intermodal station. The results demonstrated that the proposed model was able to represent throughput fluctuations more accurately and achieve short-term bandwidth configuration within a fault-tolerant and with a limited observation dataset.