Developer Challenge: Load Bundler

Background

To reduce the CO2 footprint of road-freight, it's important to find the optimal route with as little empty capacity as possible. As a small developer challenge (that will not save the earth but is a great code work-out ;-) ), let's assume the following:

Goal

We want to develop an algorithm that bundles partial truck loads to as few as possible full truck loads, filling the trucks to 34 palettes (or less - but not much less!). For our challenge, all the loads are being picked up at a central warehouse, which is the starting point for all trucks.

Input Data

in the data folder, you will find 2 files:

partial-loads.json

This contains random partial truck loads:

  {
     "destination": "Köln",   // Destination
     "pieces": 3,  // Number of euro palettes that need to go to the destination for this load
     "splitable": true,  // if this load can be split into further partials 
     "loading_window": {    // time frame during which the load must be picked up at the central warehouse
        "start": "2021-03-29T21:46:06.074Z",
        "end": "2021-03-29T23:46:06.074Z"
     },
     "unloading_window": {  // time frame during which this load must be dropped off at the destination
        "start": "2021-03-30T21:46:06.074Z",
        "end": "2021-03-30T23:46:06.074Z"
     }
  },

bundleable-destinations.json

This contains a list of destinations that may be combined.

[
    ["Hamburg","Luebeck","Kiel","Neumuenster","Wedel"],
    ["Bremen","Hannover","Bielefeld"],
    ["Frankfurt", "Köln","Heidelberg"],
    ["Konstanz","Landshut"],
    ["München","Augsburg"]
]

This means, for example: Loads that go to Konstanz are allowed to be combined into 1 truck with loads that go to Landshut, to make it into a multi-stop route. (Negative example: loads that go to Konstanz may not be combined with Loads that go to Hamburg).

Sample Output

Feel free to use whatever output format you want. This here is just an example that would be OK (example for 1 truck, the actual result would be big array of these, one entry for each truck):

   {
      "total_pieces": 31,
      "partials": [
         {
            "destination": "Luebeck",
            "pieces": 26,
            "splitable": true,
            "loading_window": {
               "start": "2021-03-29T22:06:12.974Z",
               "end": "2021-03-30T00:06:12.974Z"
            },
            "unloading_window": {
               "start": "2021-03-30T22:06:12.974Z",
               "end": "2021-03-31T00:06:12.974Z"
            }
         },
         {
            "destination": "Wedel",
            "pieces": 5,
            "splitable": true,
            "loading_window": {
               "start": "2021-03-30T18:03:41.575Z",
               "end": "2021-03-30T20:03:41.575Z"
            },
            "unloading_window": {
               "start": "2021-03-31T18:03:41.575Z",
               "end": "2021-03-31T20:03:41.575Z"
            }
         }
      ],
      "destinations": [
         "Luebeck",
         "Wedel"
      ]
   }

Some Hints

• feel free to start with ignoring the time
• feel free to assume that the time it takes between any drop-off location is 0 (or a standard value you define)
• feel free to start without splitting individual loads