Short-term forecast model of taxi demand based on time and space heterogeneity

2021 ◽  
pp. 1-12
Author(s):  
Zhiyu Yan ◽  
Shuang Lv

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.

2017 ◽  
Vol 62 ◽  
pp. 3-11 ◽  
Author(s):  
Nicholas E. Johnson ◽  
Olga Ianiuk ◽  
Daniel Cazap ◽  
Linglan Liu ◽  
Daniel Starobin ◽  
...  

2019 ◽  
Author(s):  
Isaac Opper ◽  
William Johnston ◽  
John Engberg ◽  
Lea Xenakis
Keyword(s):  
New York ◽  

2020 ◽  
Vol 45 (2) ◽  
pp. 143-168
Author(s):  
Luca Berardi ◽  
Sandra Bucerius

Sociologists and criminologists have relied on the concept of “turning points” to map individual criminal careers over the life course. Similar to individuals, criminal organizations undergo drastic changes that influence their trajectory over time and space. Using the case of the Almighty Latin King and Queen Nation (ALKQN) in New York City, we introduce the concept of “organizational turning points” to explain the group’s evolution through various legitimate and illegitimate forms. Bringing together conceptual lenses from literature on organizational change, culture and cognition, and criminology, we demonstrate that street gangs can be complex and fluid organisms that change over time and space. Identifying and recognizing organizational turning points in criminal groups can have important implications for scholars and practitioners alike.


Author(s):  
Shay Lehmann ◽  
Alla Reddy ◽  
Chan Samsundar ◽  
Tuan Huynh

Like any legacy subway system that first opened in the early 1900s, the New York City subway system operates using technology that dates from many different eras. Although some of this technology may be outdated, efforts to modernize are often hindered by budgetary limits, competing priorities, and managing the tradeoff between short-term service disruptions and long-term service improvements. At New York City Transit (NYCT), the locations of all trains on all lines are not visible to any one person in any one place and, for much of the system, train locations can only be seen at field towers for the handful of interlockings in its operational jurisdiction as result of the legacy signal system, which may come as a surprise to many daily commuters or personnel at newer metros. In 2019, developers at NYCT gained full access to the legacy signal system’s underlying track circuit occupancy data and developed an algorithm to automatically track trains and match these data with schedules and manual dispatchers’ logs in real time. This data-driven solution enables real-time train identification and tracking long before a full system modernization could be completed. This information is being provided to select personnel as part of a pilot program via several different tools with the aim of improving service management and reporting.


Sign in / Sign up

Export Citation Format

Share Document