A Review on Developments in Evolutionary Computation Approaches for Road Traffic Flow Prediction

Widespread traffic congestion significantly impacts the quality of life, posing several problems and challenges. To reduce traffic congestion, it is necessary to have accurate information about traffic flow. Accurately predicting traffic flow is challenging due to its uncertainty, nonlinearity and time-varying characteristics. A traffic management system in a city is the most important component for traffic flow prediction. This can assist drivers in selecting the best routes to their intended destinations. Therefore, cities need a comprehensive system for more precise traffic flow forecasts. Consequently, various Artificial Intelligence (AI) based techniques have been developed over time to address the issues associated with traffic flow forecasts. One of them is computational intelligence (CI), a subset of AI that can be used with AI techniques to deal with the nonlinearity and randomness of traffic flow in a better way. This review presents a detailed analysis of evolutionary computation (EC) methodologies, which belong to the field of computational intelligence. Swarm intelligence (SI) and evolutionary algorithms are types of EC techniques, each presenting optimization frameworks characterized by unique theories and objective functions. Most often employed techniques such as particle swarm optimization (PSO), genetic algorithm (GA), ant colony optimization (ACO) and artificial bee colony (ABC) are discussed and have been used to solve road traffic flow prediction problems. Additionally, hybrid approaches that combine EC techniques with AI-based techniques leverage the strengths of both methods to enhance the prediction accuracy of traffic flow models. This study examines and summarizes the most recent articles on the application of EC techniques in traffic flow prediction. Challenges and possibilities for future research work are also illustrated. The objective of this paper is to contribute to the existing knowledge by compiling, analyzing and evaluating the developments in evolutionary computation techniques as they relate to the prediction of traffic flow on roads.