A Repair Method for Missing Traffic Data Based on FCM, Optimized by the Twice Grid Optimization and Sparrow Search...

Area of Science:

• Transportation Engineering
• Data Science
• Artificial Intelligence

Background:

• Complete traffic sensor data is crucial for analyzing traffic flow dynamics and developing effective traffic control strategies.
• Missing traffic data is a prevalent challenge in real-world traffic monitoring systems.
• Accurate data imputation is essential for reliable traffic analysis and management.

Purpose of the Study:

• To propose an improved Fuzzy C-Means (FCM) algorithm for repairing missing traffic sensor data.
• To develop a robust data imputation method that considers temporal, spatial, and attribute correlations of traffic flow.
• To enhance the accuracy and reliability of traffic data for better traffic flow analysis and control.

Main Methods:

• An improved Fuzzy C-Means (FCM) algorithm, termed TGO-SSA-FCM, is developed.
• A Twice Grid Optimization (TGO) algorithm is proposed to determine reliable initial clustering centers for FCM.
• The Sparrow Search Algorithm (SSA) is employed to optimize the fuzzy weighting index (m) and classification number (k) of the FCM algorithm.
• Three distinct data repair modes are established based on the correlation of time, space, and attribute values.

Main Results:

• The TGO-SSA-FCM algorithm demonstrated superior performance compared to traditional data imputation algorithms in experimental tests.
• The effectiveness of the proposed algorithm was validated using real-world traffic sensor data from Shunyi District, Beijing.
• Experimental results indicated that the choice of data repair mode should be adapted based on different data missing rates.

Conclusions:

• The TGO-SSA-FCM algorithm provides an effective solution for repairing missing traffic sensor data.
• The developed method offers improved accuracy and reliability for traffic flow analysis and the formulation of traffic control strategies.
• The study highlights the importance of selecting appropriate data repair modes under varying conditions of data missingness.