Transportation logistics monitoring for transportation systems using the machine learning

Abstract

To decrease the number of accidents, Transportation Systems (TS) work to increase traffic efficiency and vehicular flow in urban areas. The production of datasets to carry out an in-depth analysis of the data using machine learning techniques is made possible by the generation of huge volumes of data generated by all the digital devices connected to the transportation network. This paper proposed a machine learning technique called Gradient Descent K-Nearest Neighbors (GD-KNN) for transportation logistics monitoring to improve route optimization, demand forecasting, vehicle maintenance, real-time monitoring, freight optimization, risk assessment, and continuous improvement. By harnessing data from various sources such as GPS devices, sensors, telemetric, and historical transportation data, machine learning algorithms can analyze and process this data to make accurate predictions and recommendations. The collected dataset was pre-processed using z-score normalization, and then Independent Component Analysis (ICA) was applied for the feature extraction process. Real-time monitoring enables the detection of anomalies and delays, providing alerts for timely actions. Freight optimization is achieved by analyzing parameters like weight, size, and delivery locations, resulting in cost reduction and improved load balancing. GD-KNN assesses risks and security threats using data from security systems, ensuring the safety of goods and personnel. Continuous learning allows the system to adapt to changing conditions and improve predictions over time. Overall, GD-KNN empowers transportation logistics monitoring to optimize operations, enhance customer service, and reduce costs in transportation systems.