Adaptive Traffic Flow Optimization System (C++)

An Intelligent Transportation System (ITS) simulation that models and optimizes urban traffic flow using graph algorithms, congestion modeling, and adaptive signal control.

This project simulates real-world traffic conditions in major Pakistani cities such as Karachi, Lahore, Multan, and Peshawar, with the goal of reducing congestion and improving travel efficiency.

FOR UML click this link https://usman-irshad1.github.io/Adaptive-Traffic-Simulator-with-Dynamic-Rerouting-Signal-Control/

Objective

The system minimizes overall traffic cost by balancing:

• Waiting time (queue lengths)
• Road congestion (network utilization)

Optimization Function

min Σₜ Σ₍ᵢⱼ₎ [ α Q₍ᵢⱼ₎(t) + β (f₍ᵢⱼ₎(t) / c₍ᵢⱼ₎)² ]

Where:

Q₍ᵢⱼ₎ = Queue length

f₍ᵢⱼ₎ = Traffic flow

c₍ᵢⱼ₎ = Road capacity

Core Features

1. Dynamic Graph-Based Road Network

• Cities and intersections are modeled as nodes
• Roads are modeled as directed edges
• Enables realistic simulation of traffic flow between locations

2. Intelligent Routing (Dijkstra Algorithm)

• Computes shortest paths using dynamic edge weights
• Edge weights update in real time based on congestion
• Vehicles automatically reroute based on current traffic conditions

3. BPR Congestion Model

• Travel time is calculated using the Bureau of Public Roads (BPR) function:
• w₍ᵢⱼ₎(t) = w₍ᵢⱼ₎^free (1 + α (f₍ᵢⱼ₎(t) / c₍ᵢⱼ₎)^β)
• This models how travel time increases as traffic volume approaches road capacity.

4. Adaptive Traffic Signals (Longest Queue First)

• Intersections monitor incoming road queues
• The road with the highest queue length is given the green signal
• Helps reduce bottlenecks and improves flow at congested intersections

5. Backpressure Flow Control

• Enforces road capacity constraints
• Vehicles cannot enter a road if it is full
• Prevents unrealistic movement and models real traffic jams

6. Real-Time Analytics

The system continuously tracks:

Average travel time

Traffic congestion levels

Network throughput

Performance Metrics

The simulator evaluates traffic conditions using:

Network Utilization= (1 / |E|) Σ (f₍ᵢⱼ₎ / c₍ᵢⱼ₎)

Total Delay = Σ (Tₛ d − Tₛd^free)

Where:

Tₛd = Actual travel time

Tₛd^free = Free-flow travel time

Testing Results

In a high-density simulation with 30 vehicles:

• Average Travel Time: ~18.53 units
• Network Throughput: ~0.11 vehicles per tick
• Peak Congestion Level: 7.9% saturation
• Success Rate: 100% (no deadlocks, all vehicles reached destination)