Swarm intelligence in transportation engineering

Swarm intelligence (SI) is an artificial intelligence technique based on collective behavior of self-organized systems. In the recent years, SI has attracted significant attention from researchers and has been successfully applied to solve optimization problems in transportation engineering. This Special Collection on ‘‘Swarm Intelligence in Transportation Engineering’’ presents 13 papers that reflect the broad range of frontier research activities in Transportation Engineering. These papers provide an insightful and inspiring introduction to this topic, on both conceptual aspects and technical implementation. The first paper titled ‘‘Factor analysis of comprehensive evaluation on rural built environment changes in China under multi-index panel data’’ explored an index system of China’s rural built environment (RBE), analyzed RBE developing trends from a macro-perspective, and used factor analysis to comprehensively evaluate the RBE level with multi-index panel data. A total of 24 samples collected and sorted from statistical yearbooks were analyzed, and the findings coincided with the current stage and policies of urbanization and new rural construction in China. The paper titled ‘‘Allometric relationship between port throughput growth and urban population: A case study of Shanghai port and Shanghai city’’ established an allometric growth model between the port throughput and urban population to verify the delicate relationship between the two. The authors took the throughput of Shanghai port and the population of Shanghai city as an example and studied the relative growth rate ratio. Apart from this, the Logistic model was utilized to forecast Shanghai port throughput and Shanghai urban population in the future. In the paper ‘‘Chinese automobile sales forecasting using economic indicators and typical domestic brand automobile sales data: A method based on econometric model,’’ an econometric model is proposed to analyze the dynamic connections among Chinese automobile sales, typical domestic brand automobile (Chery) sales, and some economic variables. The authors applied a vector error correction model based on co-integration to quantify the long-term impact of endogenous variables on Chinese automobile sales. The results showed a better forecasting performance when the impact of Chery sales is considered. In the paper ‘‘Grout diffusion model in porous media considering the variation in viscosity with time,’’ the authors proposed an assembled diffusion model of spheres and cylinders for grouting using a perforated pipe, taking consideration of the variation in viscosity with time. A numerical simulation method was used to verify the grout diffusion mode. In the paper ‘‘Pedestrian movement intention identification model in mixed pedestrian-bicycle sections based on phase-field coupling theory,’’ the fuzzy logic method was used to build the pedestrian movement intention identification model which synthetically considered the pedestrian safety and the satisfaction with the speed and the space for walking. The mutual influence between pedestrian and its surrounding traffic participants in mixed pedestrian–bicycle sections was comprehensively analyzed. The experimental verifications showed that the result of the pedestrian movement intention identification model was consistent with the actual situation. In the paper ‘‘Analysis of safety characteristics of flight situation in complex low-altitude airspace,’’ the authors proposed several aircraft behavior models taking account of complex low-altitude environment and general aviation flight characteristics, including an individual aircraft behavior model, a multi-flight behavior model, and an individual interaction model. The relationship and influence rules among the flight situation indicators were found in the flight situation evolution process based on a mixed flight situation simulation environment. In the paper ‘‘Reverse deduction of vehicle group situation based on dynamic Bayesian network,’’ the authors applied the probability theory to identify vehicle group situation which was constituted by target vehicle and its neighboring vehicles. And the dynamic Bayesian network was used to build the reverse deduction model of vehicle group situation, which was verified through actual and virtual driving experiments. Verification results showed that the model established in this article was reasonable and feasible.