The document proposes using an Adaptive Feature Fusion Network (AFFN) to predict passenger flow from origin to destination in metro systems. AFFN aims to adaptively fuse spatial dependencies from multiple data sources and capture periodic patterns based on external factors like weather. It also predicts inflow and outflow at each station as a side task to improve origin-destination prediction accuracy. Evaluation on real metro datasets showed AFFN outperforms baselines in accuracy metrics.

1. Base paper Title: Adaptive Feature Fusion Networks for Origin-Destination Passenger Flow
Prediction in Metro Systems
Modified Title: Using Adaptive Feature Fusion Networks to Predict Passenger Flow from
Origin to Destination in Metro Systems
Abstract
The challenges are that OD flows 1) have high temporal dynamics and complex spatial
correlations, 2) are affected by external factors, and 3) have sparse and incomplete data slices.
In this paper, we propose an Adaptive Feature Fusion Network (AFFN) to a) adaptively fuse
spatial dependencies from multiple knowledge-based graphs and even hidden correlations
between stations and b) accurately capture the periodic patterns of passenger flows based on
the auto-learned impact from external factors. To deal with the incompleteness and sparsity of
OD matrices, we extend AFFN to multi-task AFFN to predict the inflow and outflow of each
station as a side-task to further improve OD prediction accuracy. We conducted extensive
experiments on two real-world metro trip datasets collected in Nanjing and Xi’an, China.
Evaluation results show that our AFFN and multi-task AFFN outperform the state-of-theart
baseline techniques and AFFN variants in various accuracy metrics, demonstrating the
effectiveness of AFFN and each of its key components in OD prediction.
Existing System
METRO is one of the most popular and efficient transportation in metropolitan cities.
More than 50% commuters chose metro as their daily transportation in most cities. In Tokyo,
New York, and Hong Kong, the proportion of metro commuters is even higher (80%-90%) [1].
With rapid urbanization and increasing population, metro systems are facing high dynamic
travel demands, and thus need to timely optimize service operations such as scheduling elastic
timetables [2], [3] and planning flexible skip-stop lines [4], [5], which requires accurate origin-
destination (OD) passenger flow predictions. Most existing works have focused on predicting
Inflow and Outflow in metro stations (IO prediction) [6], [7], [8], [9], [10] in individual stations
for metro management [11], [12] and emergency response [13], [14]. Only a few works predict
the number of metro trips between each origin-destination station pair [15], [16]. Although OD
prediction has been well studied for taxi or ride-hailing systems, i.e., predicting the number of
taxi trips from each origin region to the destination region [17], [18], [19]. These techniques,

2. however, cannot be applied directly to the metro as the stations are connected by sparse metro
lines other than dense road networks where Euclidean distances can roughly approximate road
distances. Therefore, we are motivated to study how to accurately predict citywide OD flow on
sparse metro networks.
Drawback in Existing System
 Complexity and Computation: AFFNs often involve complex architectures that
might demand significant computational resources, making them computationally
intensive. This could limit their applicability in real-time prediction scenarios or on
devices with limited computing power.
 Data Requirements: These models typically require a substantial amount of data to
learn effectively. If the available data is sparse, incomplete, or contains biases, the
model's predictive accuracy might be compromised.
 Overfitting: AFFNs can be prone to overfitting, especially if the model's complexity
isn’t managed properly or if the dataset is limited. Overfitting can lead to the model
performing well on training data but failing to generalize to unseen or new data.
 Transferability: Deploying a model trained on data from one metro system to another
might not yield accurate predictions due to differences in commuter behavior,
infrastructure layout, or other factors unique to each system.
Proposed System
 An external factor-based attention module is proposed to collaboratively integrate
periodic data flow with attention weights learned from external factors to improve the
prediction accuracy.
 we also validate the effectiveness of AFFN in different structural metro systems, give
a visual analysis of the prediction results, and demonstrate the runtime efficiency of our
proposed models.
 we propose an Adaptive Feature Fusion Network (AFFN) (see Fig. 2(a)) to predict
Origin-Destination passenger flow between metro stations.
 We proposed an Adaptive Feature Fusion Network (AFFN) to predict origin-
destination passenger flow in a citywide metro system.

3. Algorithm
 Historical Data: Entry and exit counts, timestamps, weather conditions, events, station
connectivity, etc.
 Pre processed Features: Extracted and engineered features influencing passenger
flows.
 Predicted OD Flows: Estimated passenger movements between different stations.
Advantages
 Integration of Diverse Data: AFFNs can effectively fuse various data types such as
temporal, spatial, weather, and event-related data. This integration allows for a holistic
understanding of factors influencing passenger flows.
 Feature Adaptability: These networks dynamically adapt and weigh features,
learning the importance of different variables over time. This adaptability enables the
model to focus on relevant features, improving prediction accuracy.
 Temporal Understanding: Their ability to comprehend temporal aspects, including
daily, weekly, or seasonal patterns, enables better predictions, especially during peak
hours or special events.
 Real-Time Predictions: Once trained and deployed, AFFNs can generate real-time
predictions, aiding metro system operators in making timely decisions related to
resource allocation, scheduling, and crowd management.
Software Specification
 Processor : I3 core processor
 Ram : 4 GB
 Hard disk : 500 GB
Software Specification
 Operating System : Windows 10 /11
 Frond End : Python
 Back End : Mysql Server
 IDE Tools : Pycharm