Profit-maximizing Autonomous Mobility-on-Demand (AMoD) operators may exploit consumer surplus through excessive prices and underserve less profitable regions by relocating vehicles away from them. We study how penalty-based regulation of an AMoD operator in a multimodal mobility system can limit surge pricing and improve regional service availability. We model regulation as an outer control loop that acts on a learning-based operator. Static regulation with fixed penalties is formulated as a Markov Decision Process for an operator that jointly controls pricing and rebalancing, and dynamic regulation as a Partially Observable Stochastic Game in which a regulator periodically adjusts region-specific penalties in closed loop. Operator and regulator policies are learned with graph-based multi-agent reinforcement learning. In experiments with real-world Manhattan data, fixed surge pricing penalties reduce the share of travelers facing excessive prices from 22% to 1%. For regional service availability, dynamic regulation improves the regulator's reward by up to 15.6% over fixed penalties of equal mean intensity. The gains in service availability are strongest when spare fleet capacity exists, whereas service deficits persist in peak periods with highly utilized fleets. The learned regulator policy responds by reducing penalties where limited fleet capacity restricts compliance. The results identify trade-offs between regulatory targets, operator profitability, and traveler utility.
Figures & tables
Fig. 1: Decision architecture, consisting of an AMoD control layer and, under dynamic regulation, an AMoD regulation layer. In the AMoD control layer, Step 1 determines prices and target idle-vehicle distribution, Step 2 captures mode choice, queueing, passenger matching, and regulation evaluation, and Step 3 executes rebalancing and updates vehicle locations and queues. In the AMoD regulation layer, penalty levels are updated at each regulatory decision interval. Between regulatory updates, the current penalty levels remain in force.
Fig. 2: Simulation area in southern Manhattan, discretized into 12 regions.
Penalty level ϕSP (US$)
Metric
0
1
5
10
Regulator
Affected travelers (#)
4,965 (144)
3,750 (182)
1,279 (187)
296 (113)
Affected share (%)
22 (1)
17 (1)
6 (1)
1 (1)
Avg. price factor excess
0.07 (0.00)
0.06 (0.00)
0.03 (0.00)
0.01 (0.00)
Penalties (US$)
0 (0)
230 (11)
170 (27)
28 (10)
TABLE I: Performance metrics under static surge pricing regulation.
Hour of day
Penalty level
15
16
17
18
19
20
ϕSP=0
36
27
40
82
97
97
ϕSP=1
37
28
41
84
97
98
ϕSP=5
40
30
44
86
97
98
ϕSP=10
40
30
43
85
97
98
TABLE II: Vehicle utilization by hour in percent under static surge pricing regulation.
Evaluation penalty
Training penalty
Reward (k US$)
ϕSP=0
ϕSP=0
129.5 (0.7)
ϕSP=1
ϕSP=1
128.7 (0.7)
ϕSP=0
129.1 (0.7)
ϕSP=5
ϕSP=5
127.0 (0.7)
ϕSP=0
127.6 (0.7)
ϕSP=10
ϕSP=10
126.5 (0.7)
TABLE III: Operator reward comparison by evaluation and training penalty level under static surge pricing.
Fig. 3: Regulation outcomes, pricing policy, and traveler outcomes with and without static surge pricing regulation by region and hour in one representative evaluation run. Travelers count as affected if the surge pricing factor is above the threshold.
λ
Regulation
Penalty level
Operator reward (US$)
Traveler utility (k US$)
Regulator reward (k US$)
1.0
Dynamic
ϕˉSA=4
125.6 (0.8)
-112.0 (0.9)
-184 (7)
Static
ϕSA=0
129.5 (0.9)
-115.1 (0.8)
-164 (7)
Static
ϕSA=4
128.8 (0.7)
-114.0 (1.0)
-184 (8)
Static
ϕSA=200
118.3 (0.8)
-110.1 (0.9)
-1,521 (5)
0.1
Dynamic
ϕˉSA=22
124.0 (0.7)
-109.5 (0.9)
-124 (5)
Static
ϕSA=0
129.5 (0.9)
-115.1 (0.8)
-164 (7)
TABLE IV: Static and dynamic spatial equity regulation performance comparison.
Static
Dynamic
Static
ϕSA
λ
ϕSA
Metric
0
1.0
0.1
0.05
0.01
200
Regulator
Avg. penalty level (US$)
0
4
22
47
177
200
(0.00)
(0.05)
(0.34)
(0.35)
(0.29)
(0.00)
Avg. service rate (%)
73
74
78
79
81
81
TABLE V: Performance metrics under spatial equity regulation.
Fig. 4: Mean service deficit (%) without regional service availability regulation.
Fig. 5: Regulator outcomes and policy with dynamic regional service availability regulation by region and hour in one representative evaluation run.
Fig. 6: Operator policy and traveler outcomes with and without dynamic regional service availability regulation by region and hour in one representative evaluation run.
Fleet size NAMOD
Regulator decision interval ΔREG
Target service rate sAMOD,∗
Metric
850
1050
1250
20
40
60
0.6
0.7
0.8
Regulator
Reward
-124 (5)
-86 (5)
-72 (6)
-124 (5)
-134 (8)
-138 (5)
-124 (5)
-155 (7)
-202 (6)
Avg. penalty level (US$)
22 (0.34)
14 (0.21)
22 (0.65)
22 (0.34)
17 (0.34)
43 (0.41)
22 (0.34)
25 (0.95)
36 (0.80)
Avg. service rate (%)
78 (0)
82 (0)
85 (0)
78 (0)
77 (1)
79 (0)
78 (0)
80 (0)
80 (0)
Avg. service deficit (%)
5 (0)
3 (0)
2 (0)
5 (0)
5 (0)
4 (0)
5 (0)
7 (0)
10 (0)
TABLE VI: Sensitivity analyses under dynamic spatial equity regulation.
Appendix figures & tables7 assets
Supplementary material from the paper’s appendix.
Appendix
Operator
Regulator
Static
Dynamic
Dynamic
Hyperparameter
SA
SP
SA
SA
Training
Algorithm
SAC
SAC
SAC
A2C
Initialization strategy
WS
FS
WS
FS
Training episodes
20,000
20,000
20,000
20,000
Appendix
TABLE S1: Hyperparameters used for operator and regulator training.
Hour of day
15
16
17
18
19
20
Mean
3,793
2,982
4,346
7,091
9,069
10,021
Std. dev.
67
54
57
84
88
92
Appendix
TABLE S2: Total demand by hour of day.
Fig. S2: Scenario characteristics over 100 simulation samples.
Fig. S3: PT-to-AMoD travel-time ratio by origin region and hour, averaged over 100 simulation samples.
Fig. S4: AMoD reference price in US$ per minute of travel time by origin region and hour, averaged over 100 simulation samples.
Fig. S5: Surge pricing penalties paid by the operator during training per episode for the first 500 training episodes under surge pricing penalty levels ϕSP∈{1,5,10} .
Regulation
Hour of day
15
16
17
18
19
20
Static
ϕSA=0
37
28
42
83
96
97
Dynamic
λ=1.0
41
29
43
84
97
98
λ=0.1
41
33
48
85
97
98
λ=0.05
42
35
48
87
97
98
λ=0.01
43
36
50
86
97
98
Appendix
TABLE S3: Vehicle utilization by hour in percent under regional service availability regulation.
aCOSYS-GRETTIA, Univ Gustave Eiffel, Marne-la-Vallée, France · bDepartment of Electrical Engineering and Computer Sciences, University of California, Berkeley, Berkeley, USA · cDepartment of Technology, Management and Economics, Technical University of Denmark, Denmark