Improving short-term bike sharing demand forecast through an irregular convolutional neural network

Abstract

<p>As an important task for the management of bike sharing systems, accurate forecast of travel demand could facilitate dispatch and relocation of bicycles to improve user satisfaction. In recent years, many <a href="https://www.sciencedirect.com/topics/engineering/deep-learning" title="Learn more about deep learning from ScienceDirect's AI-generated Topic Pages">deep learning</a> algorithms have been introduced to improve bicycle usage forecast. A typical practice is to integrate convolutional (CNN) and recurrent <a href="https://www.sciencedirect.com/topics/social-sciences/neural-network" title="Learn more about neural network from ScienceDirect's AI-generated Topic Pages">neural network</a> (RNN) to capture spatial–temporal dependency in historical travel demand. For typical CNN, the <a href="https://www.sciencedirect.com/topics/computer-science/convolution-operation" title="Learn more about convolution operation from ScienceDirect's AI-generated Topic Pages">convolution operation</a> is conducted through a kernel that moves across a “matrix-format” city to extract features over spatially adjacent urban areas. This practice assumes that areas close to each other could provide useful information that improves prediction accuracy. However, bicycle usage in neighboring areas might not always be similar, given spatial variations in built environment characteristics and travel behavior that affect cycling activities. Yet, areas that are far apart can be relatively more similar in temporal usage patterns. To utilize the hidden linkage among these distant urban areas, the study proposes an irregular convolutional Long-Short Term Memory model (IrConv+LSTM) to improve short-term bike sharing <a href="https://www.sciencedirect.com/topics/engineering/demand-forecast" title="Learn more about demand forecast from ScienceDirect's AI-generated Topic Pages">demand forecast</a>. The model modifies traditional CNN with irregular convolutional architecture to leverage the hidden linkage among “semantic neighbors”. The proposed model is evaluated with a set of <a href="https://www.sciencedirect.com/topics/computer-science/model-benchmark" title="Learn more about benchmark models from ScienceDirect's AI-generated Topic Pages">benchmark models</a> in five study sites, which include one dockless bike sharing system in Singapore, and four station-based systems in Chicago, Washington, D.C., New York, and London. We find that IrConv+LSTM outperforms other <a href="https://www.sciencedirect.com/topics/computer-science/model-benchmark" title="Learn more about benchmark models from ScienceDirect's AI-generated Topic Pages">benchmark models</a> in the five cities. The model also achieves superior performance in areas with varying levels of bicycle usage and during peak periods. The findings suggest that “thinking beyond spatial neighbors” can further improve short-term travel demand prediction of urban bike sharing systems.</p>