Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Strategic Infrastructure Design via Multi-Agent Congestion Games with Joint Placement and Pricing

Niloofar Aminikalibar, Farzaneh Farhadi, Maria Chli

Abstract

Real-world infrastructure planning increasingly involves strategic interactions among autonomous agents competing over congestible, limited resources. Applications such as Electric Vehicle (EV) charging, emergency response, and intelligent transportation require coordinated resource placement and pricing decisions, while anticipating the adaptive behaviour of decentralised, self-interested agents. We propose a novel multi-agent framework for joint placement and pricing under such interactions, formalised as a bi-level optimisation model. The upper level represents a central planner, while the lower level captures agent responses via coupled non-atomic congestion games. Motivated by the EV charging domain, we study a setting where a central planner provisions chargers and road capacity under budget and profitability constraints. The agent population includes both EV drivers and non-charging drivers (NCDs), who respond to congestion, delays, and costs. To solve the resulting NP-hard problem, we introduce ABO-MPN, a double-layer approximation framework that decouples agent types, applies integer adjustment and rounding, and targets high-impact placement and pricing decisions. Experiments on benchmark networks show that our model reduces social cost by up to 40% compared to placement- or pricing-only baselines, and generalises to other MAS-relevant domains.

Strategic Infrastructure Design via Multi-Agent Congestion Games with Joint Placement and Pricing

Abstract

Real-world infrastructure planning increasingly involves strategic interactions among autonomous agents competing over congestible, limited resources. Applications such as Electric Vehicle (EV) charging, emergency response, and intelligent transportation require coordinated resource placement and pricing decisions, while anticipating the adaptive behaviour of decentralised, self-interested agents. We propose a novel multi-agent framework for joint placement and pricing under such interactions, formalised as a bi-level optimisation model. The upper level represents a central planner, while the lower level captures agent responses via coupled non-atomic congestion games. Motivated by the EV charging domain, we study a setting where a central planner provisions chargers and road capacity under budget and profitability constraints. The agent population includes both EV drivers and non-charging drivers (NCDs), who respond to congestion, delays, and costs. To solve the resulting NP-hard problem, we introduce ABO-MPN, a double-layer approximation framework that decouples agent types, applies integer adjustment and rounding, and targets high-impact placement and pricing decisions. Experiments on benchmark networks show that our model reduces social cost by up to 40% compared to placement- or pricing-only baselines, and generalises to other MAS-relevant domains.
Paper Structure (20 sections, 1 theorem, 6 equations, 2 figures, 1 table, 2 algorithms)

This paper contains 20 sections, 1 theorem, 6 equations, 2 figures, 1 table, 2 algorithms.

Table of Contents

  1. Introduction
  2. Literature Review
  3. Strategic Multi-Agent Model
  4. Network and Agent Definitions
  5. Cost Structure and Agent Behaviour
  6. Government Authority’s Optimisation Problem
  7. Agent Equilibrium Modelling and Reformulation
  8. Coupled Congestion Games
  9. Nash Equilibrium Conditions
  10. Computational Complexity of the Reformulated Problem
  11. Solution Methodology
  12. Layer 1: Behavioural Decomposition Based on EV Penetration
  13. Layer 2: Integer Adjustment for Placement Decisions
  14. Experimental Evaluation
  15. Benchmark Network and Parameters
  16. ...and 5 more sections

Key Result

theorem 1

The GA's optimisation problem, after reformulating the equilibrium constraints as explicit inequalities, is NP-hard.

Figures (2)

  • Figure 1: Nguyen-Dupuis Network:Optimal placement and pricing in the network
  • Figure 2: Left: Social cost vs. budget for the joint model, showing diminishing returns beyond $B{=}7$, with shaded regions indicating standard deviation over random parameters across $e$ values. Right: Joint Optimisation outperforms benchmarks by at least 10% across all budgets, with over 40% gain at $B{=}52$, highlighting the value of joint placement and pricing.

Theorems & Definitions (2)

  • theorem 1
  • proof