Short-term road traffic speed prediction is a long-standing topic in the area of Intelligent Transportation System. Apparently, effective prediction of the traffic speed on the road can not only provide timely details for the navigation system concerned and help the drivers to make better path selection, but also greatly improve the road supervision efficiency of the traffic department. At present, some researches on speed prediction based on GPS data, by adding weather and other auxiliary information, using graph convolutional neural network to capture the temporal and spatial characteristics, have achieved excellent results. In this paper, the problem of short-term traffic speed prediction based on GPS positioning data is further studied. For the processing of time series, we innovatively introduce Dynamic Time Warping algorithm into the problem and propose a Long Short-Term Memory with Dynamic Time Warping (D-LSTM) model. D-LSTM model, which integrates Dynamic Time Warping algorithm, can fine-tune the time feature, thus adjusting the current data distribution to be close to the historical data. More importantly, the fine-tuned data can still get a distinct improvement without special treatment of holidays. Meanwhile, considering that the data under different feature distributions have different effects on the prediction results, attention mechanism is also introduced in the model. Our experiments show that our proposed model D-LSTM performs better than other basic models in many kinds of traffic speed prediction problems with different time intervals, and especially significant in the traffic speed prediction on weekends.

In the context of automated driving, the connected and automated vehicles (CAVs) technology unlock the energy saving potential. This paper develops an LSTM‐based deep learning framework for eco‐driving adaptive identification on Intelligent vehicle multivariate time series data. The framework can be adapted for Driver Assistance Systems (DAS) to reduce fuel consumption. Specifically, considering overtaking on rural road is a critical maneuver for operation and has potential to reduce consumption, a simulated driving experiment with 30 participants was conducted to collect the multivariate time series data of the overtaking operation behaviors in conditional automation driving. Driving behaviors were classified into eco‐driving operation behaviors and high fuel consumption operation behaviors based on fuel consumption calculated by using vehicle specific power (VSP). Significance analysis based on linear regression was adopted to identify operation behaviors, and an eco‐driving behavior identification model was established with the use of long short‐term memory (LSTM) for multivariate classification theory. Meanwhile, the other four classification algorithms were used to establish identification models for comparison. The results indicated that the gear position, lane position, the acceleration pedal depth, the clutch pedal depth, and the brake pedal depth had a significant influence on fuel consumption. The eco‐driving behavior identification model of overtaking demonstrated a high classification power and robustness with a classification accuracy of 89.16%. According to the simulation results, the developed adaptive identification model is with promising performance. The conclusions provide theoretical support for developing an adaptive strategy for connected eco‐driving.

The pricing strategy of remanufactured products has been widely regarded as a significant issue due to the associated economic benefits for both suppliers and retailers awareness worldwide. However, the overall profit determined based on centralized pricing strategy or decentralized pricing strategy is not in line with the actual profit of the manufacturer and the retailer, especially when taking the cost of carbon emission under the system of cap and trade into account. This paper proposes a revenue sharing contract model to coordinate the optimal pricing strategies for both manufacturers and retailers who engages in remanufacturing. The model considers the different Willingness to Pay (WTP) for remanufactured products, and profit distribution of manufacturers and retailers under the Cap and Trade policy. The optimal retail and wholesale price of new products and remanufactured products are derived by solving a convex programming model. Our results identify that: (1) The carbon emission cap is negatively correlated with the retail price of both new products and remanufactured products when the established profits function of the closed-loop supply chain is set to be a constant. The retail price of remanufactured products is also negatively correlated with the value of WTP, while the retail price of new products is positively correlated with WTP. (2) The profit of retailers is positively correlated with carbon emission cap, while the relation of the manufacturer’s profit and carbon limit present an inverted U shape, which indicates the optimal carbon emission limit for manufacturers achieving the maximum profit. Moreover, the profit of the retailer and manufacturer and the coefficient of WTP is positively correlated. Numerical experiments are conducted to examine the feasibility of the proposed pricing strategies and gain optimal managerial insights for carbon cap and trade policy.

In order to solve the problem of cross-regional customized bus (CB) route planning during the COVID-19, we develop a CB route planning method based on an improved Q-learning algorithm. First, we design a sub-regional route planning approach considering commuters’ time windows of pick-up stops and drop-off stops. Second, for the CB route with the optimal social total travel cost, we improve the traditional Q-learning algorithm, including state-action pair, reward function and update rule of Q value table. Then, a setup method of CB stops is designed and the path impedance function is constructed to obtain the optimal operating path between each of the two stops. Finally, we take three CB lines in Beijing as examples for numerical experiment, the theoretical and numerical results show that (i) compared with the current situation, although the actual operating cost of optimized route increases slightly, it is covered by the reduction of travel cost of passengers and the transmission risk of COVID-19 has also dropped significantly; (ii) the improved Q-learning algorithm can solve the problem of data transmission lag effectively and reduce the social total travel cost obviously.

In the latest connected vehicle (CV) message standards, including SAE J2735‐2016 and T‐CASE 53–2017, the basic safety messages (BSMs) are designed specifically as effective measures for traffic safety management and applications. In this study, a testbed on the Nanchang‐Jiujiang Intelligent Highway in Jiangxi, China is illustrated as an example, and the basic architecture and key technologies is introduced for a proactive traffic safety utilisation, where the core basic safety message (BSM) data are sorted and implemented to perceive and predict risky driving behaviours in a field environment. On this basis, an accurate insight into time‐critical driving safety issues can be achieved by investigating raw BSM data, such as the inter‐vehicle distance, driver manipulation, vehicle speed, and acceleration/deceleration. Furthermore, to effectively take advantage of connected vehicle information and perceive the high uncertainty of driving behaviours during an emergency situation and evaluate the driving safety in mixed traffic scenarios, a long short‐term memory (LSTM) based deep learning framework is introduced to build a multi‐horizon vehicle crash risk prediction model using continuous BSMs as the inputs. The experimental results demonstrate the significance of connected vehicle data and deep learning algorithms for improving driving safety and promoting widespread deployment and application of connected vehicles.