Recurrent Neural-Based Vehicle Demand Forecasting and Relocation Optimization for Car-Sharing System: A Real Use Case in Thailand

Abstract

A car-sharing system has been playing an important role as an alternative transport mode in order to avoid traffic congestion and pollution due to a quick growth of usage of private cars. In this paper, we propose a novel vehicle relocation system with a major improvement in threefolds: (i) data preprocessing, (ii) demand forecasting, and (iii) relocation optimization. The data preprocessing is presented in order to automatically remove fake demands caused by search failures and application errors. Then, the real demand is forecasted using a deep learning approach, Bidirectional Gated Recurrent Unit. Finally, the Minimum Cost Maximum Flow algorithm is deployed to maximize forecasted demands, while minimizing the amount of relocations. Furthermore, the system is deployed in the real use case, entitled “CU Toyota Ha:mo,” which is a car-sharing system in Chulalongkorn University. It is based on a web application along with rule-based notification via Line. The experiment was conducted based on the real vehicle usage data in 2019. By comparing in real environment in November of 2019, the results show that our model even outperforms the manual relocation by experienced staff. It achieved a 3% opportunity loss reduction and 3% less relocation trips, reducing human effort by 17 man-hours/week.

A cooperative strategy for optimizing vehicle relocations and staff movements in cities where several carsharing companies operate simultaneously

“…However, the problems of vanishing gradient and exploding gradient exist with the long-sequence input. Gated recurrent unit (GRU) (Vateekul et al, 2021) network introduces memory cells to learn whether the previous hidden layer state needs to be forgotten or updated, which can control the information to be memorized with a shorter training time. The structure of the GRU network is shown in Fig.…”

“…In recent years, deep learning based data modeling algorithms have been supported by scholars for their powerful ability to learn the correlated feature representations (Ke et al, 2021). For the carsharing travel demand prediction tasks, existing studies usually use the long short-term memory (LSTM) neural network (Yu et al, 2020) and gated recurrent unit (GRU) (Vateekul et al, 2021) to capture the temporal characteristics. Although LSTM and GRU structures can adequately extract temporal dependencies from the historical data, they fail to effectively model the spatial relations.…”

Transport Vehicle Demand Prediction Using Context-Aware Neural Networks

“…Neural networks serve as an incredibly valuable tool in this scenario because the data collected follow a sequential pattern (a time series). Deep learning techniques, such as neural networks, excel in handling unorganized data and aid in identifying critical features for analysis, making them especially effective in this context [3]. The use of LSTMs can also help identify long-term changes in trends or patterns in the demand for transport vehicles.…”

“…As is evident from Figure 1, sudden changes in demand have no pattern as such and cannot be identified by any models without multimodal analysis. data and aid in identifying critical features for analysis, making them especially effective in this context [3]. The use of LSTMs can also help identify long-term changes in trends or patterns in the demand for transport vehicles.…”