CI Status

License

Creators

Corinna Krebs

Introduction

This solution validator checks the feasibility of a given solution with respect to various loading and routing constraints (see Code Documentation or Literature). It handles the combination of two optimization problems — the vehicle routing problem (VRP) combined with the container loading (CLP), also known as the 2L-CVRP, 2L-VRPTW, 3L-CVRP and 3L-VRPTW.

Requirements

The tool is implemented in C++ and in Java. Therefore, it can be used with either of the two languages. Only standard libraries are used.

C++ Version

• Language level: min. C++11
• Software: e.g. MSBuild via Visual Studio
• Build files: validator.sln, validator.vcxproj

Java Version

• Language level: min. Java 11
• Software: Java JRE, Apache Maven Version 2.0, IDE (e.g. IntelliJ or Eclipse), Lombok
• Build files: pom.xml

Usage

• Examples:
  • C++: Main.cpp
  • Java: Main.java

Code Example

In general, the program consists of two main parts:

• Provide Data (Instance, ConstraintSet and Solution)
• Use Check Methods

General hints are given below. For the concrete usage of the methods, please check the documentation of the code.

// Necessary Data
Instance instance           = Read.readInstanceFile("../Input/Instances/Krebs_Ehmke_Koch_2020/001_n020_m200_bt3.txt");
ConstraintSet constraintSet = Read.readConstraintFile("../Input/Constraint_Sets/P1.txt");
Solution solution           = Read.readSolutionFile("../Input/PackPlan/001_n020_m200_bt3_P1_1.txt", instance);

// Check Constraints
boolean feasibleRouting = checkRoutingConstraints(solution, constraintSet, instance);
boolean feasibleLoading = checkLoadingConstraints(solution, constraintSet, instance);

Input Data

Provide necessary data via Instance, ConstraintSet and Solution objects. Exemplary files are available in Input Folder.

Instance

• Necessary Data: Information about vehicles, customers, items and demands
• Provide Data:
  • use converted Instances from Literature (see this repository) or
  • use Instance Class Constructor (see C++: Instance.h or Java: Instance.java).

Constraint Set

• Necessary Data: Included Loading Constraints and Parameters
• Provide Data:
  • read Constraint File (see examples) or
  • adapt an Example and read this new File or
  • use ConstraintSet Class Constructor (see C++: ConstraintSet.h or Java: ConstraintSet.java).

Solution

• Necessary Data: Objective Values, run time, number of iterations, set of routes and packing plans
• Provide Data:
  • use Solution file (see our repository) or
  • use Solution Class Constructor (see C++: Solution.h or Java: Solution.java) or
  • convert Data via Write Class (see C++: Write.cpp or Java: Write.java) and use the new file.

Check Methods:

• Necessary Data: Input Data is shown above. Provided as parameters to the methods.
• Return Value: Feasibility of the solution (feasible: true; infeasible: false).
• Check all Constraints defined in ConstraintSet: Use checkRoutingConstraints and checkLoadingConstraints
• Check specific Constraint, see
  • C++: ConstraintsLoading.cpp and ConstraintsRouting.cpp
  • Java: Loading.java and Routing.java.

Literature

Loading Constraints and formulas are described in

• Advanced Loading Constraints for 3D Vehicle Routing Problems
  by C. Krebs, J. Ehmke, H. Koch

• Axle Weights in Combined Vehicle Routing and Container Loading Problems
  by C. Krebs, J. Ehmke

• Vertical Stability Constraints in Combined Vehicle Routing and 3D Container Loading Problem
  by C. Krebs, J. Ehmke