Weather conditions have considerable influence on cycling travel behavior, especially in tropical countries such as Singapore, which has a hot, humid, and rainy climate. This study examined the effects of weather conditions and weather forecasts on cycling travel behavior in Singapore. Cyclists (n = 553) answered a questionnaire on their perceptions of weather conditions, travel accident risk, pre-trip and during-trip acquisition of weather information, possible travel plan changes, trip duration, trip purpose, trip frequency, and respondents' social demography. The questionnaires were administrated directly on sites around mass rapid transit (MRT) stations and bike parking areas during daylight hours (08:00–18:00 h) and in dry as well as wet weather conditions. Concurrently, real-time weather information from the nearest weather measurement station and weather forecast information were collected from a smart phone application furnished by the National Environment Agency. Cyclists were found to prefer relatively lower temperature (29.5–31.5°C) and humidity (52.3%–62.7%) and no rainfall (in past 60 min). Higher temperature (>30.9°C), humidity (>55.8%), and rainfall (>0.28 mm in past 60 min) tended to elevate cyclists' self-estimated level of traffic accident risk. Nearly 30% of participating cyclists checked the weather forecast information before the onset of the trip, in which internet, radio, and smart phone applications were the main media sources. Social/leisure trip was the main purpose for cyclists during wet weather. Irregular trips were clearly underrepresented during wet weather conditions, suggesting that some trips were postponed or cancelled. Cyclists with before-trip information of wet weather forecast were more likely to change their travel mode, but such influence was weaker for adverse weather forecast acquired during the trip. Research results suggested that the authorities could provide real-time and predicted weather information, especially during wet weather conditions, to help travelers adjust their travel plan in time and improve the level of travel safety.

Traffic flow prediction plays an important role in intelligent transportation systems. To accurately capture the complex non-linear temporal characteristics of traffic flow, this paper adopts a Bi-directional Gated Recurrent Unit (Bi-GRU) model in traffic flow prediction. Compared to Gated Recurrent Unit (GRU), which can memorize information from the previous sequence, this model can memorize the traffic flow information in both previous and subsequent sequence. To demonstrate the model’s performance, a set of real case data at 1-hour intervals from 5 working days was used, wherein the dataset was separated into training and validation. To improve data quality, an augmented dickey-fuller unit root test and differential processing were performed before model training. Four benchmark models were used, including the Autoregressive Integrated Moving Average (ARIMA), Long Short-Term Memory (LSTM), Bidirectional Long Short-Term Memory (Bi-LSTM), and GRU. The prediction results show the superior performance of Bi-GRU. The Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), and Mean Absolute Error (MAE) of the Bi-GRU model are 30.38, 9.88%, and 23.35, respectively. The prediction accuracy of LSTM, Bi-LSTM, GRU, and Bi-GRU, which belong to deep learning methods, is significantly higher than that of the traditional ARIMA model. The MAPE difference of Bi-GRU and GRU is 0.48% which is a small prediction error value. The results show that the prediction accuracy of the peak period is higher than that of the low peak. The Bi-GRU model has a certain lag on traffic flow prediction.