Taxi Utilization Rate Maximization by Dynamic Demand Prediction: A Case Study in the City of Chicago

2021 ◽  
pp. 036119812110597
Author(s):  
Tianyi Li ◽  
Guo-Jun Qi ◽  
Raphael Stern
Keyword(s):  
Short Term Memory ◽  
Transportation Network ◽  
Capacity Utilization ◽  
Utilization Rate ◽  
Optimization Strategy ◽  
Dynamic Demand ◽  
Demand Prediction ◽  
Linear Relationships ◽  
Taxi Service ◽  
The City

The explosive popularity of transportation network companies (TNCs) in the last decade has imposed dramatic disruptions on the taxi industry, but not all the impacts are beneficial. For instance, studies have shown taxi capacity utilization rate is lower than 50% in five major U.S. cities. With the availability of taxi data, this study finds the taxi utilization rate is around 40% in June 2019 (normal scenario) and 35% in June 2020 (COVID 19 scenario) in the city of Chicago, U.S. Powered by recent advances in the deep learning of capturing non-linear relationships and the availability of datasets, a real-time taxi trip optimization strategy with dynamic demand prediction was designed using long short-term memory (LSTM) architecture to maximize the taxi utilization rate. The algorithms are tested in both scenarios—normal time and COVID 19 time—and promising results have been shown by implementing the strategy, with around 19% improvement in mileage utilization rate in June 2019 and 74% in June 2020 compared with the baseline without any optimizations. Additionally, this study investigated the impacts of COVID 19 on the taxi service in Chicago.

scholarly journals AI-based network topology optimization system

2021 ◽  
Vol 2 (4) ◽  
pp. 81-90
Author(s):  
Han Zengfu ◽  
Kong Jiankun ◽  
Wang Zhiguo ◽  
Zhang Yiwei ◽  
Liu Ke ◽  
...  
Keyword(s):  
Topology Optimization ◽  
Resource Utilization ◽  
Network Topology ◽  
Model Building ◽  
Short Term Memory ◽  
Capacity Utilization ◽  
Optimization Strategy ◽  
Edge Deletion ◽  
Optimization System ◽  
Network Topology Optimization

Existing network topology planning does not fully consider the increasing network traffic and problem of uneven link capacity utilization, resulting in lower resource utilization and unnecessary investments in network construction. The AI-based network topology optimization system introduced in this paper builds a Long Short-Term Memory (LSTM) model for time series traffic forecasting, which uses NetworkX, a Python library, for graph analysis, dynamically optimizes the network topology by edge deletion or addition based on traffic over nodes, and ensures network load balancing when node traffic increases, mainly introducing the LSTM forecasting model building process, parameter optimization strategy, and network topology optimization in some detail. As it effectively enhances resource utilization, this system is vital to the optimization of complex network topology. The end of this paper looks forward to the future development of artificial intelligence, and suggests the possibility of how to cooperate with operator networks and how to establish cross-border ecological development.

The Key Technology Research for Logistics Vehicle Monitoring System Based on GPS

2014 ◽  
Vol 1049-1050 ◽  
pp. 1189-1192 ◽  
Author(s):  
Li Jun Huang
Keyword(s):  
Transportation Network ◽  
New Technology ◽  
Capacity Utilization ◽  
Utilization Rate ◽  
Transportation Industry ◽  
Logistics Industry ◽  
Network Information ◽  
Control Automation ◽  
Speed Up ◽  
Vehicle Management

As the logistics industry scale expands unceasingly, logistics vehicles, increasing the traditional freight vehicle management method because of the lack of the support of new technology and gradually cannot meet the needs of enterprise development. The research target of this paper is through the transportation network, information, the digital management of vehicles, the vehicle and cargo optimized configuring, and improve the utilization rate of vehicle load and capacity utilization, so as to speed up the efficiency of business operation of the entire transportation industry, enterprise operation standardization and control automation, reduce business operating costs, the use of system and the research will speed up the development of logistics industry.

scholarly journals Optimization strategy of automobile dispatching based on neural network model

2021 ◽  
Vol 271 ◽  
pp. 01020
Author(s):  
Chuyuan Wang
Keyword(s):  
Neural Network ◽  
Public Transportation ◽  
Missing Values ◽  
Supply And Demand ◽  
Sharing Economy ◽  
Optimization Strategy ◽  
Data Standardization ◽  
Demand Prediction ◽  
The City

As a representative product of the sharing economy era and a powerful supplement to public transportation shared cars have the characteristics of convenience, efficiency, environmental protection, and green travel, and to a certain extent alleviate the contradiction between supply and demand, and solve the problem of long-term idle vehicles and overloaded operation of roads problems. But the uneven distribution of shared cars, the coexistence of no cars, and empty seats will happen. To solve the above problems, this article first analyzes data outliers, data missing values, and data standardization processing on the attached data, and then builds a BP neural network demand prediction model to obtain the distribution of shared car usage in the city.

Spatial and Temporal Differences in Weekday Travel Durations Between Private-for-Hire Transportation Services and Transit in the City Center

2021 ◽  
pp. 036119812110197
Author(s):  
Baxter Shandobil ◽  
Ty Lazarchik ◽  
Kelly Clifton
Keyword(s):  
Transportation Network ◽  
Central City ◽  
Temporal Analysis ◽  
Time Of Day ◽  
City Center ◽  
Transportation Services ◽  
Regulatory Frameworks ◽  
Transit Services ◽  
Level Information ◽  
The City

There is increasing evidence that ridehailing and other private-for-hire (PfH) services such as taxis and limousines are diverting trips from transit services. One question that arises is where and when PfH services are filling gaps in transit services and where they are competing with transit services that are publicly subsidized. Using weekday trip-level information for trips originating in or destined for the city center of Portland, OR from PfH transportation services (taxis, transportation network companies, limousines) and transit trip data collected from OpenTripPlanner, this study investigated the temporal and spatial differences in travel durations between actual PfH trips and comparable transit trips (the same origin–destination and time of day). This paper contributes to this question and to a growing body of research about the use of ridehailing and other on-demand services. Specifically, it provides a spatial and temporal analysis of the demand for PfH transportation using an actual census of trips for a given 2 week period. The comparison of trip durations of actual PfH trips to hypothetical transit trips for the same origin–destination pairs into or out of the central city gives insights for policy making around pricing and other regulatory frameworks that could be implemented in time and space.

scholarly journals Understanding the Usage Patterns of Bicycle-Sharing Systems to Predict Users’ Demand: A Case Study in Wenzhou, China

2018 ◽  
Vol 2018 ◽  
pp. 1-21 ◽  
Author(s):  
Xiaomei Xu ◽  
Zhirui Ye ◽  
Jin Li ◽  
Mingtao Xu
Keyword(s):  
Cluster Analysis ◽  
Comparative Analysis ◽  
Prediction Models ◽  
Transportation Network ◽  
Back Propagation ◽  
Back Propagation Neural Network ◽  
Demand Prediction ◽  
Usage Patterns ◽  
Accuracy Of Prediction

Bicycle-sharing systems (BSSs) have become a prominent feature of the transportation network in many cities. Along with the boom of BSSs, cities face the challenge of bicycle unavailability and dock shortages. It is essential to conduct rebalancing operations, the success of which largely depend on users’ demand prediction. The objective of this study is to develop users’ demand prediction models based on the rental data, which will serve rebalancing operations. First, methods to collect and process the relevant data are presented. Bicycle usage patterns are then examined from both trip-based aspect and station-based aspect to provide some guidance for users’ demand prediction. After that, the methodology combining cluster analysis, a back-propagation neural network (BPNN), and comparative analysis is proposed to predict users’ demand. Cluster analysis is used to identify different service types of stations, the BPNN method is utilized to establish the demand prediction models for different service types of stations, and comparative analysis is employed to determine if the accuracy of the prediction models is improved by making a distinction among stations and working/nonworking days. Finally, a case study is conducted to evaluate the performance of the proposed methodology. Results indicate that making a distinction among stations and working/nonworking days when predicting users’ demand can improve the accuracy of prediction models.

scholarly journals Real-Time Return Demand Prediction Based on Multisource Data of One-Way Carsharing Systems

2021 ◽  
Vol 2021 ◽  
pp. 1-14
Author(s):  
Dongbo Liu ◽  
Jian Lu ◽  
Wanjing Ma
Keyword(s):  
Prediction Model ◽  
Real Time ◽  
Supply And Demand ◽  
Optimization Strategy ◽  
Trajectory Data ◽  
Demand Prediction ◽  
Data User ◽  
User Characteristic ◽  
Destination Station ◽  
The One

One-way carsharing system has been widely adopted in the carsharing field due to its flexible services. However, as one of the main limitations of the one-way carsharing system, the imbalance between supply and demand needs to be solved. Predicting pick-up demand has been studied to achieve the goal, but using returned vehicles to reduce unnecessary relocation is also one of the important methods. Nowadays, trajectory data and other data are available for real-time prediction for return demand. Based on the return demand prediction, the relocation response can be more reasonable. Thus, the balance of demand and supply can be largely improved. The multisource data include trajectory data, user application log data, order data, station data, and user characteristic data. Based on these data, a return demand prediction model was used to predict whether the user will return the vehicle in 15 min in real time, and a destination station prediction model was applied to forecast which station the user will park at. Finally, a case study using ten stations’ one-week field data was conducted to test the benefit of the dynamic return demand prediction. The results showed that the return demand prediction improves the efficiency of the relocations by mitigating the condition that the station parking space is full or empty. The potential application of this study would effectively reduce unnecessary relocation and further formulate an active operation optimization strategy to reduce the system’s operational cost and improve the service quality of the system.

Predicting Spatiotemporal Demand of Dockless E-Scooter Sharing Services with a Masked Fully Convolutional Network

2021 ◽  
Vol 10 (11) ◽  
pp. 773
Author(s):  
Santi Phithakkitnukooon ◽  
Karn Patanukhom ◽  
Merkebe Getachew Demissie
Keyword(s):  
Loss Function ◽  
Short Term Memory ◽  
Selection Process ◽  
Convolutional Network ◽  
Fully Convolutional Network ◽  
Demand Prediction ◽  
Learning Techniques ◽  
Last Mile ◽  
Long Short Term Memory ◽  
Highly Correlated

Dockless electric scooters (e-scooter) have emerged as a green alternative to automobiles and a solution to the first- and last-mile problems. Demand anticipation, or being able to accurately predict spatiotemporal demand of e-scooter usage, is one supply–demand balancing strategy. In this paper, we present a dockless e-scooter demand prediction model based on a fully convolutional network (FCN) coupled with a masking process and a weighted loss function, namely, masked FCN (or MFCN). The MFCN model handles the sparse e-scooter usage data with its masking process and weighted loss function. The model is trained with highly correlated features through our feature selection process. Next-hour and next 24-h prediction schemes have been tested for both pick-up and drop-off demands. Overall, the proposed MFCN outperforms other baseline models including a naïve forecasting, linear regression, and convolutional long short-term memory networks with mean absolute errors of 0.0434 and 0.0464 for the next-hour pick-up and drop-off demand prediction, respectively, and the errors of 0.0491 and 0.0501 for the next 24-h pick-up and drop-off demand prediction, respectively. The developed MFCN expands the collection of deep learning techniques that can be applied in the transportation domain, especially spatiotemporal demand prediction.

Sign in / Sign up

Export Citation Format

Share Document