We provide 40 instances for the Vehicle Routing Problem with Synchronization Constraints (VRPSCDL). In this problem, each customer receives one or more deliveries that must synchronize at the customer location.

In these files, deliveries are identified by DeliveryID and depots by DepotID. A delivery is served to a customer location, identified as LocationID. Depots also have a LocationID.

Travel times between locations are provided below the line "Location;[TravelTime]+". The first item in each row is always the LocationID; afterwards, we give the distance to each other location in the problem, sorted in ascending order.

For example, in the following travel time matrix, travelling from location 3 to 8 takes 1.234 min., and travelling from 8 to 3 takes 1.345 min. The elements in the main diagonal are always 0.

Location;[TravelTime]+
3;0;1.234
8;1.345;0

Time related variables are always given in minutes. This concerns data under labels PlanningHorizon, ServiceTime, and TravelTime. 

Results in our paper:

Briseida Sarasola, Karl F. Doerner:
Adaptive large neighborhood search for the vehicle routing problem with synchronization constraints at the delivery location. Networks 75(1): 64-85 (2020)

are likewise provided in minutes and were calculated by multipliying minutes by 1000 to achieve positive integers.

The amount of idle time w_u at each customer u location is not explicitly given in the instance files, as it can be derived from the given data (and modified, if necessary). Most results in our paper were calculated using w_u as the sum of the service times of all deliveries to the customer location u, unless otherwise stated.
