When you're looking to solve a vehicle routing problem (VRP) — whether you're familiar with the space or not — it’s worth thinking through the structure of the problem you're looking to solve. Doing so helps streamline satisfying business requirements like dropping off goods in the appropriate sequence or ensuring all vehicles end their shift at the same location.
Say, for example, you're looking to stand up ride sharing logistics software like Uber or Lyft. You're going to think about structuring that routing problem a bit differently than if you want to solve a routing problem for package distribution at an Amazon, UPS, or Walmart.
These types of vehicle routing problems aren't exclusive to just big-name companies. They're common among businesses of all sizes. Plus, they're possible to start solving without requiring large teams with deep operations research expertise. Let’s have a look at a few examples of routing problems and how to think about structuring your data in the context of Nextmv Cloud.
If you'd like to follow along and try each scenario out, create a free Nextmv Cloud account.
Delivery (and ride sharing) vehicle routing problems
Delivery is probably what comes to mind when you think of a vehicle routing problem: drivers need to service requests (or stops) to handle both pickups and dropoffs.
We often see this play out with restaurant food and grocery delivery as well as the ride sharing space. Oico, a Brazilian construction material delivery company, is another delivery example that dispatches vehicles to several suppliers to pick up and dropoff materials at job sites.
You have a delivery vehicle routing problem if:
- Your requests or stops include pickups and dropoffs
- Your vehicles start and end their routes from anywhere
- Your routes are typically dynamic or being made in real time
In delivery scenarios, it’s common to add in other constraints to represent business logic such as delivery precedence, hard delivery windows, target times, break times, and more. Get started with a delivery tutorial and sample JSON input in our documentation.
Sourcing vehicle routing problem
Sourcing is a type of vehicle routing problem where vehicles are picking materials up and bringing them back to a common location such as a warehouse or a depot.
This is common in areas like waste management and recycling. It’s also prevalent in agriculture and dairy. Take Milk Moovement, a Canadian dairy technology company, for example. They solve a sourcing vehicle routing problem by dispatching a fleet of trucks to farms to pick up raw milk and transport it to processing facilities.
You have a sourcing vehicle routing problem if:
- Your requests or stops only include pickups
- Your vehicles start their route from anywhere
- Your vehicles end their route at the same location
In sourcing scenarios, it's also common for routes to be scheduled in advance and also fixed, but it's possible to re-plan routes on the fly for when things change. Get started with a sourcing tutorial and sample JSON input in our documentation.
Distribution vehicle routing problem
Distribution is a type of vehicle routing problem characterized by the movement of goods from one location to many others. We probably experience this daily with mail and packages going from a centralized location to our mailboxes and front steps. It’s also common for distributors for home goods like Wayfair and Living Spaces.
You have a distribution vehicle routing problem if:
- Your requests or stops only include dropoffs
- Your vehicles start their route at the same location
- Your vehicles end their route anywhere
Similar to sourcing scenarios, distribution scenarios often have fixed routes that are scheduled in advance, but it’s possible to re-plan routes ad hoc. Get started with a distribution tutorial and sample JSON input in our documentation.
End of dispatch
You should have a good sense for how to characterize a few different types of vehicle routing problems and have the starting points to begin solving for each scenario. Give it a whirl with a free Nextmv Cloud account or contact us with questions!
