Bibliometric Survey on Particle Swarm Optimization Algorithms (2001–2021)

Table 2

Summary of existing works on hybrid particle swarm optimization (PSO).

References	Techniques	Performance metrics	Pros	Cons
Liu et al. [34]	PSO-DE	Objective function values, median, mean, standard deviation	Converges quickly, it solves constrained problems	Higher cost in some problems
Iwata and Fukuyama [35]	DEEPSO	Least square error	Accuracy	Underperforming in unsTable output conditions
Yoshida and Fukuyama [36]	DEEPSO	Mean and standard deviation of objective functions	Faster convergence speed and better accuracy	Parallel distributed processing is not considered
Buba and Lee [37]	Hybrid DE and PSO	Passengers’ and operators’ cost	Diversity of solution, better accuracy	It is not tested on larger and more realistic problem instances with heterogeneous buses
Wang et al. [38]	DEPSO	Mean and standard deviations of the objective functions	Population diversity, higher scalability	Exploration and exploitation need improvement, implementation complexity
Dabhi and Pandya [39]	EVDEPSO	Iterations and mean execution time and mean, max, min, and standard deviation of the objective function, average ranking index	It is superior in terms of the ranking index and average ranking index as compared to the other algorithms	Tests are confined to a limited category of problems
Mirsadeghi and Khodayifar [40]	PSODE	Mean, the best, the worst, and standard deviation of the objective functions, run-time, success rate	High accuracy of solution	Improvement for the exploration and exploitation capabilities, implementation complexity
Marcelino et al. [41]	C-DEEPSO	Mean, median, and standard deviation of objective functions	Better accuracy, high scalability	Improvement for the exploration and exploitation capabilities
Tomar and Pant, 2011 [42]	MPDE	Mean, standard deviation of objective function	Better computation time	Getting trapped in local optima
Yu et al. 2014 [43]	HPSO-DE	Mean, standard deviation and T-value of objective function	Maintains diversity of the population	Slow convergence and high computation time
Lin et al. 2018 [44]	HPSODE	Mean, standard deviation of objective function	Good solution accuracy	Converging at local optima,
Too et al. 2019 [45]	BPSODE	Mean, standard deviation of the objective function	Good solution accuracy	Extra computational cost, premature convergence
Parouha and Das 2016 [46]	DE-PSO	Mean, standard deviation of objective function	Better solution accuracy	Rapid loss of diversity, converging prematurely.