This article is cited in 29 papers

NUMERICAL METHODS AND DATA ANALYSIS

Big data analysis in a geoinformatic problem of short-term traffic flow forecasting based on a k nearest neighbors method

A. A. Agafonov^a, A. S. Yumaganov^a, V. V. Myasnikov<sup>ab

a</sup> Samara National Research University, 443086, Russia, Samara, Moskovskoye Shosse 34
^b IPSI RAS – Branch of the FSRC “Crystallography and Photonics” RAS, Molodogvardeyskaya 151, 443001, Samara, Russia

Abstract: Accurate and timely information on the current and predicted traffic flows is important for the successful deployment of intelligent transport systems. These data play an essential role in traffic management and control. Using traffic flow information, travelers could plan their routes to avoid traffic congestion, reduce travel time and environmental pollution, as well as improving traffic operation efficiency in general. In this paper, we propose a distributed model for short-term traffic flow prediction based on a k nearest neighbors method, that takes into account spatial and temporal traffic flow distributions. The proposed model is implemented as a MapReduce based algorithm in an Apache Spark framework. An experimental study of the proposed model is carried out on a traffic flow data in the transportation network of Samara, Russia. The results demonstrate that the proposed model has high predictive accuracy and an execution time sufficient for real-time prediction.

Keywords: traffic flow, short-term forecasting, k nearest neighbors, MapReduce.

Received: 03.12.2018
Accepted: 10.12.2018

DOI: 10.18287/2412-6179-2018-42-6-1101-1111