Challenge

WeFix is the UK's only call-out mobile phone repair solution, repairing phones and tablets on customer’s doorsteps. For the 100+ technicians travelling around the country repairing devices, a large part of the day can be spent driving to the next job; taking the best route can make a big difference to the company's efficiency. As a fairly small, young company, the initial solution to this problem had been built in an ad hoc way. It was expensive (lots of calls to Google maps API) and often gave the wrong answer. The CTO contacted Ortom to build a route optimisation solution.

Solution

Ortom built a route optimisation application based on OpenStreetMap data. This worked by solving a variant of the travelling salesman problem. This is a type of problem that was first described when salesmen would travel from city to city selling their wares. The challenge was to find the shortest route that visited all cities in a list. The problem is simple with a small number of cities but becomes vastly more complex as more locations are added. Finding the shortest route is a very complex task. A large body of maths, known as combinatorial optimisation has been built up around solving this and similar problems. Knowing which algorithms to use and applying them correctly takes expert knowledge.

Result

Ortom built a solution that reduced journey times by 15% and was cheaper to run. The application ran on scalable cloud infrastructure using docker and used Python's FastAPI framework.

Conclusion

After using Ortom’s solution successfully in the UK, WeFix is now expanding into North America and Australia using the same solution.