Optimising logistics in intelligent transportation systems through CNN-IT: a convolutional neural network approach for enhanced efficiency and precision

Abstract

This study addresses the critical challenge of optimising logistics within intelligent transportation systems (ITS) by introducing a novel approach leveraging convolutional neural network-IT (CNN-IT). The overarching objective is to enhance the efficiency and precision of logistics operations through advanced machine learning techniques. The methodology involves the design and implementation of a CNN-IT model trained on a diverse dataset comprising sensory data from various transportation modes, including cars, buses, and trains. The model’s architecture is carefully crafted to extract intricate patterns from the sensor data, enabling accurate classification of different vehicles. The process begins with the collection and preprocessing of a diverse dataset, encompassing accelerometer, gyroscope, magnetometer, and audio sensor data from various transportation modes. Results demonstrate the CNN-IT model’s remarkable accuracy, precision, recall, and F1 score, showcasing its capability to make well-informed predictions. The model achieved an outstanding accuracy of 92%, with precision, recall, and F1 score at 91%, 93%, and 92%, respectively. These metrics collectively reflect the model’s exceptional performance in accurately classifying transportation modes while maintaining a balanced trade-off between precision and recall. The strategic engineering of the model’s architecture to effectively handle sensor data and accurately categorise vehicles plays a significant role in optimising logistics operations.