Description

This project provides the way to solve multiple variations of Vehicle Routing Problem known as rich VRP. It provides default metaheuristic implementation which can be roughly described as Multi-objective Parthenogenesis based Evolutionary Algorithm with Ruin and Recreate Mutation Operator

Getting started

Please check documentation here.

Design goal

Although performance is constantly in focus, the main idea behind design is extensibility: the project aims to support a wide range of VRP variations known as Rich VRP. This is achieved through various extension points: custom constraints, objective functions, acceptance criteria, etc.

How to use

VRP solver is built in Rust. To install it, use cargo install or pull the source code from master.

Install from source

Once pulled the source code, you can build it using cargo:

cargo build --release

Built binaries can be found in the ./target/release directory.

Alternatively, you can try to run the following script from the project root:

    ./solve_problem.sh examples/data/pragmatic/objectives/berlin.default.problem.json

It will build the executable and automatically launch the solver with the specified VRP definition. Results are stored in the folder where a problem definition is located.

Install from Cargo

You can install vrp solver cli tool directly with cargo install:

cargo install vrp-cli

Ensure that your $PATH is properly configured to source the crates binaries, and then run solver using the vrp-cli command.

Use from command line

vrp-cli crate is designed to use on problems defined in scientific or custom (aka 'pragmatic') format:

vrp-cli solve pragmatic problem_definition.json -m routing_matrix.json --max-time=120`

Please refer to crate docs for more details.

Use from code

If you're using rust, then you can simply use vrp-scientific, vrp-pragmatic crates to solve VRP problem defined in 'pragmatic' or 'scientific' format using default metaheuristic. For more complex scenarios, please refer to vrp-core documentation.

If you're using some other language, e.g java, kotlin, javascript, please check examples section to see how to call the library from it.

Project structure

The project consists of the following parts:

• vrp solver code: the source code of the solver is split into four crates:
  • vrp-core: a core crate with default metaheuristic implementation
  • vrp-scientific: a crate with functionality to solve problems from some of scientific benchmarks on top of the core crate
  • vrp-pragmatic: a crate which provides logic to solve rich VRP using pragmatic json format on top of the core crate
  • vrp-cli: a crate which aggregates logic of others crates and exposes them as a library and application
• docs: a source code of the user guide documentation published here. Use mdbook tool to build it locally.
• examples: provides various examples:
  • data: a data examples such as problem definition, configuration, etc.
  • json-pragmatic: an example how to solve problem in pragmatic json format from rust code using the project crates
  • jvm-interop: a gradle project which demonstrates how to use the library from java and kotlin

Dependant projects

• analysis: provides way to analyze solutions, algorithm behaviour
• API: API prototype built using Rust/AWS/Terraform

Status

Experimental.