Fig 1 - uploaded by Joseph Y. J. Chow
Content may be subject to copyright.
![Network monitoring architecture using method from [8].](profile/Joseph-Chow-3/publication/343695977/figure/fig1/AS:925512797003776@1597670854071/Network-monitoring-architecture-using-method-from-8.png)
Source publication
![Fig. 1. Network monitoring architecture using method from [8].](profile/Joseph-Chow-3/publication/343695977/figure/fig1/AS:925512797003776@1597670854071/Network-monitoring-architecture-using-method-from-8_Q320.jpg)
In prior research, a statistically cheap method was developed to monitor transportation network performance by using only a few groups of agents without having to forecast the population flows. The current study validates this "multi-agent inverse optimization" method using taxi GPS trajectories data from the city of Wuhan, China. Using a controlle...
Contexts in source publication
Context 1
... While the time travel times correspond to real data, (1) the sampling of OD pairs over time is not based on real demand for information and (2) the realized route choices are assumed to be the same as Google queries as opposed to realized choices collected from the field. In practice, this method would need to serve a system design illustrated in Fig. 1. The success of the system design depends on where the sample data from end users are coming from (e.g. crowdsourced participants like Google Waze or GPS data from regulated taxis, or both?). If route data collected from the field only corresponds to certain OD demand, which OD demands sufficiently monitor the network, and how many ...
Context 2
... network by travelers experiencing the congestion. It tests the effectiveness of using sampled data from two OD pairs in the network in acting as both sensors and inference mechanisms. By proving the effectiveness using only two OD pairs, the study gives credence to larger monitoring systems that can make use of multiple OD pair trajectories. As Fig. 1 shows, a monitoring system requires either processing of GPS trajectory data to map it to a network data structure or a data collection device that automatically outputs location data in the network data structure. Since, many data sources are only provided from GPS data, we also propose mapping algorithms to match the location data to ...
Context 3
... forecasting population flows. To elaborate, the results suggest that a practitioner can implement a monitoring system that can observe route choices made along a controlled set of OD pairs over time, and use those results to explain congestion effects throughout the network in real time and evaluate intervention strategies as highlighted in Fig. 1. [8] infers capacity effects throughout a network using only GPS probe samples without the statistically costly step of forecasting population flows. However, the earlier study only provided a theoretical argument and numerical illustration using real data. No validation of the accuracy of the method in monitoring a system is provided. ...
Context 4
... work should implement this system described in Fig. 1 in a real-world setting using GIS tools and use the monitoring with predefined thresholds to set alerts for dual variables in an online dashboard. Related work can also include monitoring a network before, during, and after a disaster to quantify the impact of dual price increases due to capacity degradation. Since user GPS data may ...
Similar publications
Citations
Accurate prediction of traffic flow is of great significance for alleviating urban traffic congestions. Most previous studies used historical traffic data, in which only one model or algorithm was adopted by the whole prediction space and the differences in various regions were ignored. In this context, based on time and space heterogeneity, a Classification and Regression Trees-K-Nearest Neighbor (CART-KNN) Hybrid Prediction model was proposed to predict short-term taxi demand. Firstly, a concentric partitioning method was applied to divide the test area into discrete small areas according to its boarding density level. Then the CART model was used to divide the dataset of each area according to its temporal characteristics, and KNN was established for each subset by using the corresponding boarding density data to estimate the parameters of the KNN model. Finally, the proposed method was tested on the New York City Taxi and Limousine Commission (TLC) data, and the traditional KNN model, backpropagation (BP) neural network, long-short term memory model (LSTM) were used to compare with the proposed CART-KNN model. The selected models were used to predict the demand for taxis in New York City, and the Kriging Interpolation was used to obtain all the regional predictions. From the results, it can be suggested that the proposed CART-KNN model performed better than other general models by showing smaller mean absolute percentage error (MAPE) and root mean square error (RMSE) value. The improvement of prediction accuracy of CART-KNN model is helpful to understand the regional demand pattern to partition the boarding density data from the time and space dimensions. The partition method can be extended into many models using traffic data.
