A Reinforcement Learning Based Traffic Control Strategy in a Macroscopic Fundamental Diagram Region

Table 2

Results of the Friedman test for different demand levels.

P value		Average intersection queue length			Average intersection waiting time
		Pretimed strategy	Actuated strategy	Perimeter strategy	Pretimed strategy	Actuated strategy	Perimeter strategy
The proposed MARL	Low demand	0	0	0	0	0	0
	Medium demand	0	0	0.0007	0.0028	0.0028	0.0028
	High demand	0.0043	0.0043	0.0012	0	0.0046	0.0027