UK Travel Cost Optimizer - Project Description
Inspiration 💡
Working with service businesses, I constantly heard the same frustration: "We're losing money on travel, but we don't know where."
One client, a cleaning company with multiple freelancers, showed me their planning process - hours spent manually matching cleaners to homes, using Google Maps for each route, maintaining complex spreadsheets, yet still missing parking costs and congestion charges.
That's when I realized: this isn't a spreadsheet problem, it's an optimization problem that needs AI.
What I Learned 🎓
Building this project taught me invaluable lessons about the hidden complexity of service businesses:
Real costs are invisible - Businesses track mileage but miss parking fees, congestion charges, travel time compensation, and zone-specific costs. The true cost of a journey can be several times the fuel cost alone.
Every business is unique - A plumber with a van has different needs than an interpreter using public transport. The solution must adapt to each business model.
Simplicity is non-negotiable - Service businesses don't have IT departments. If it's not dead simple to integrate, it won't be used.
Trust requires transparency - When the system says "Send John instead of Mary," businesses need to understand why. The reasoning must be clear and verifiable.
How I Built It 🛠️
I started by shadowing actual dispatch managers to understand their workflow. This revealed the core challenge: they make hundreds of micro-decisions daily with incomplete information.
The Architecture:
- Data Input Layer - Accepts booking and provider data via API or Excel templates for testing
- Smart Processing Engine - Analyzes all possible combinations considering:
- Travel distance and time
- Individual provider rates
- Transport modes (car, public, mixed)
- UK-specific charges (ULEZ, parking, tolls)
- Provider availability and skills
- Optimization Output - Returns best assignments with complete cost breakdowns
Technology Stack:
- Python + Streamlit for rapid prototyping
- Google Maps API for accurate routing
- Custom algorithms for UK-specific cost calculations
- Modular design for easy customization
Challenges Faced 🚧
Understanding Hidden Complexity:
- UK has multiple congestion zones with different rules - London ULEZ, Birmingham CAZ, Bath Clean Air Zone
- Parking costs vary dramatically by location and time
- Providers use different transport modes with different cost structures
Making It Business-Ready:
- Service businesses use inconsistent address formats - built robust geocoding
- Discovered providers have existing bookings that affect availability
- Different businesses prioritize differently (speed vs cost vs skills)
Technical Hurdles:
- Initial calculations were too slow for real-time use
- Handling edge cases like multi-stop journeys
- Building trust through explainable recommendations
Real-World Impact 🚀
Scenario 1: Interpreting Service
A London interpreting agency manages freelance interpreters across the city. Previously, they assigned jobs based on "who's closest" using basic postcode matching.
With the optimizer:
- Discovered that sending public transport users to central London saved significantly in parking
- Multi-booking routes reduced dead travel time dramatically
- Monthly savings exceeded expectations
Scenario 2: Home Care Provider
A care company with mixed staff (some drive, some use buses) struggled with fair job allocation.
The system revealed:
- Morning shifts in suburban areas suited drivers
- City center visits were cheaper with public transport users
- Optimizing based on true costs improved staff satisfaction and reduced costs substantially
Scenario 3: Emergency Plumbing
A 24/7 plumbing service needed rapid response while controlling costs.
Implementation showed:
- Night jobs incurred fewer congestion charges
- Grouping nearby emergency calls saved significant travel time
- Transparent cost breakdowns helped justify premium charges to customers
Easy Implementation 🔧
For Businesses:
- No software installation required
- Use existing booking data - just IDs and addresses
- Get recommendations via API or web interface
- See exactly why each assignment was made
Adaptability:
- Configure your own provider rates
- Set transport modes per provider
- Define service areas and constraints
- Choose optimization priorities (cost vs speed vs quality)
The Bigger Picture 🌍
This project proves that significant business improvements don't always require complex enterprise software. By focusing on one specific pain point and solving it thoroughly, we can deliver immediate value.
Key Innovation: First UK-specific travel cost optimizer that considers ALL costs, not just distance. Built for the real world where providers use different transport modes and face complex charging schemes.
Future Vision:
- Expand to other countries with local regulations
- Learn from historical data to predict accurate travel times
- Integrate with popular booking systems
- Help reduce carbon emissions through smarter routing
Conclusion
Every service business loses money on inefficient travel planning. This project transforms a complex manual process into a simple API call, saving hours of planning time and reducing travel costs significantly.
By making the invisible visible - showing all hidden costs - businesses can finally make informed decisions about their most expensive resource: people's time.
The future of service businesses is intelligent automation that adapts to their unique needs. This project is proof that it's not only possible but practical.
Built with ❤️ for the thousands of service businesses keeping our communities running
Log in or sign up for Devpost to join the conversation.