1 of 15

Team 4

QUANTUM-AI TRAFFIC OPTIMIZATION

4

2 of 15

3 of 15

The Problem

  • In many African urban centers, persistent traffic congestion suppresses economic growth by wasting billions in lost work hours, consumed fuel, and postponed business dealings  .An estimated amount of US $314 billion is lost in urban cities due to traffic congestion alone.

​

  • $314 billion annually

​

4 of 15

Country

Time Lost (Workdays/Year)

Economic Loss (Million USD/Day)

Notable Health/Social Impacts

🇬🇭 Ghana

35 workdays/year

$1.1 million/day

High PM2.5 levels leads to increase in asthma cases, chest pain, headaches (Accra, Kumasi)

🇲🇬 Madagascar

40 workdays/year (avg est.)

$0.11 million/day

Poor public transport; long delays; major highway upgrade ongoing

🇨🇲 Cameroon

30–45 workdays/year (est. avg)

$9.9 million/day

Increase in road crashes (US $168M/year); severe congestion in Douala/Bamenda

🇳🇬 Nigeria

70 workdays/year

$16.4 million/day

11,000 pollution-related premature deaths/year (mostly in Lagos)

🇲🇦 Morocco

25–30 workdays/year

3.8 million/day (≈ 1.05% GDP / 365)

Cities like Casablanca exceed WHO air quality limits; households spending 10–20% of income on transport; traffic contributes ~⅓ of Morocco's CO₂ emissions

5 of 15

Real data adaptability

Congestion prediction using ML algorithm

Quantum optimization

USSD protocol for users/drivers

OUR SOLUTION

6 of 15

IMPLEMENTATION

    • x₀ = Σ⁽ᵏ⁻¹⁾ 2ⁱ·b₀,ᵢ
    •   x₁ = Σ⁽ᵏ⁻¹⁾ 2ⁱ· b₁,ᵢ
    • b₀,ᵢ and b₁,ᵢ ∈ {0,1} are binary variables

QUBO Encoding:

    • Penalty = λ · (Σ⁽ᵏ⁻¹⁾ 2ⁱ(b₀,ᵢ + b₁,ᵢ) − T)²

Constraint

    • Q = c₀·Σ⁽ᵏ⁻¹⁾ 2ⁱ·b₀,ᵢ + c₁·Σ⁽ᵏ⁻¹⁾ 2ⁱ·b₁,ᵢ + λ·(Σ⁽ᵏ⁻¹⁾ 2ⁱ(b₀,ᵢ + b₁,ᵢ) − T)²

Objective Function

Objective Function

    • Minimize c₀·x₀ + c₁·x₁, where
    • c₀ = 1 / (N₀ + ε),   c₁ = 1 / (N₁ + ε)
    • N₀ and N₁ are vehicle counts, ε is a small positive number.

Decision Variable

    • x₀ ∈ ℤ: green time for main road m ≤ x₀ ≤ T − m
    • x₁ ∈ ℤ: green time for side road m ≤ x₁ ≤ T − m
    • Where m is minimal time for green light and T is total cycle time at the intersection.

7 of 15

    • Run Algorithm

Qiskit

    • Store Results

QALG

    • Retrieve result from database

API

    • Simulations

Pygame

Congestion Prediction

QAOA implementation

One Intersection

Two Intersections

Multiple Intersections

APPROACH

8 of 15

9 of 15

RESULTS

10 of 15

Traffic green light optimization simulation with pygame

11 of 15

QUANTUM ADVANTAGE

12 of 15

B2C (Business-to-Consumer)

Target: Individual drivers, commuters, logistics companies

​

  • Value Proposition: Get real-time traffic insights (via USSD/app)

Reduce travel time and fuel costs

Avoid fines or delays from congestion

Useful for ride-hailing, delivery, private car owners

​

  • Revenue Models: Freemium app (basic vs premium insights)

Daily/weekly subscription via mobile money

Ad-supported (location-based promotions)

Sell premium access to driver unions (e.g., taxi associations)

​

B2G (Business-to-Government)

Target: Municipalities, Transport Ministries, Smart City programs

​

  • Value Proposition: Optimize traffic light timing using real-time data

Reduce congestion and emissions across key intersections

Improve emergency response times

Feed into urban planning, public safety, and infrastructure investment

​

  • Revenue Models: Licensing fees for traffic management system

Public-private partnerships (PPP)

Data-as-a-service (provide analytics dashboards and traffic reports)Implementation contracts (paid consultancy + platform deployment)

​

BUSINESS PLAN

13 of 15

14 of 15

Optimizing green light reduces traffic, making road safer, cities better, the air cleaner, people healthier and more productive.

​

IMPACT

15 of 15

MEDAASE

​

MERCI

​

THANK YOU!

4